Method and apparatus for contextual data enhancement

• US 4,876,726 A
• Filed: 01/07/1986
• Issued: 10/24/1989
• Est. Priority Date: 01/07/1986
• Status: Expired due to Term

• Current Assignee: Printrak International, Inc. (Motorola Solutions, Inc.)
• Original Assignee: De la Rue Printrak Incorporated
• Inventors: Mabry, George R., Capello, Richard D.
• Primary Examiner(s): Boudreau, Leo H.
• Application Number:
  US06/816,865
• Time in Patent Office: 1,386 Days
• Field of Search: 382/21, 382/22, 382/23, 382/4, 382/67, 382/54, 382/5, 382/18
• US Class Current: 382/124
• CPC Class Codes: G06T 5/20 using local operators
• CPC Class Codes: G06V 40/1347 Preprocessing; Feature extraction

Abstract

A fingerprint image enhancement and identification system, wherein image data in digital form, is analyzed in three basic scan windows which are clocked through a matrix array of data elements each containing either intensity data, direction data or both. Using a least summed contrast method, the first window estimates data element directions by analyzing the intensity data in selected direction slits in the window to provide a preliminary direction estimate of a slit direction or "no direction". Thereafter, the second window passes through the matrix to process the preliminary direction estimates, and enhance the direction estimates to identify a slit direction which, along with the directions adjacent to it, are the directions of the predominate number of data elements in the window, or which is the average of the two slit directions which comprise the largest and second largest number of data elements, if they are sufficiently close in direction. The final window examines data elements having enhanced directions and enhances the intensity data of a center data element in the third window, by analysis of the center data element and the neighboring slit data elements in the slit with the direction of the enhanced direction of the center data element.

Claims

1. A method for enhancing data contained in data elements of an N×

M matrix of data elements, each data element containing data representative of an intensity within a range of intensities, the method comprising;

(a) selecting a Y×

Y direction estimate scan window of data elements, from within the N×

M matrix, with Y<

N, Y<

M, and wherein said window has a center data element;

(b) taking the cumulative sum of the differences in the intensity values for each of a plurality of pairs of data elements which are spaced apart by a selected number of data elements, and each comprising one of the data elements in a respective one of a plurality of different slits, wherein each different slit represents a different slit direction and wherein the respective one of a plurality of different slits contains the center data element and is within the direction estimate scan window; and

(c) assigning a direction estimate to the center data element based upon a comparison of the cumulative sum of such differences for the respective slits.

2. The method of claim 1, further comprising the steps of:

assigning to the center data element the direction of the slit having the lowest cumulative sum of the differences in the intensity values for the plurality of pairs of data elements spaced apart by a selected number of data elements within the respective slit.

3. The method of claim 2, further comprising the steps of:

determining the dynamic range of the intensity values of the data elements in the Y×

Y direction estimate scan window, within which dynamic range are contained a selected percentage of the number of data elements in the direction estimate scan window;

assigning "no direction" as the direction estimate, unless the dynamic range exceeds a selected threshold value.

4. The method of claim 3, wherein the dynamic range is the difference between an upper intensity value below which fall the intensity values of substantially 90 percent of the data elements in the direction estimate scan window and a lower intensity value above which fall the intensity values of substantially 90 percent of the data elements in the direction estimate scan window.

5. The method of claim 3, further comprising the steps of:

comparing the lowest cumulative sum of the differences in intensity values to the cumulative sum of the differences in intensity values for the slits generally aligned in direction with direction of the slit having the lowest cumulative sum;

comparing the lowest cumulative sum of the differences in intensity values to the cumulative sum of the differences in intensity values for the slits generally not aligned with the direction of the slit having the lowest cumulative sum;

assigning to the center data element the direction of the slit having the lowest cumulative sum unless a first selected number of the generally aligned slits have a cumulative sum which fails to exceed the lowest cumulative slit sum by a first selected percentage of the lowest cumulative slit sum or a second selected number of the generally not aligned slits have a cumulative sum of the differences in intensities which fails to exceed the lowest cumulative slit sum by a second selected percentage.

6. The method of claim 5, wherein there are twelve separate intersecting slit directions and the generally aligned slits are those four immediately adjacent slits, two on each side of the slit representing the slit direction having the lowest cumulative sum of the differences in intensities, and the first selected number is four, the second selected number is one.

7. The method of claim 6, wherein the first and second percentages are each about 20 percent.

8. The method of claim 7, further comprising the steps of incrementing the direction estimate scan window through the N×

M matrix until each data element of the N×

M matrix has been a center data element of the Y×

Y direction estimate scan window.

9. The method of claim 1, wherein after the steps enumerated therein at least some of the data elements in the N×

M matrix of data elements contains data representative of an assigned direction from within one of a plurality of directions, wherein each of said plurality of directions is generally equally spaced apart in direction, comprising the further steps of;

selecting a Z×

Z enhanced direction scan window of data elements from within the N×

M matrix, with Z<

N, Z<

M, and having a center data element;

determining for each of the plurality of directions the number of data elements within the enhanced direction scan window which have such assigned directions;

assigning to the center data element of the Z×

Z scan window the direction had by the largest number of data elements when the sum of the number of data elements which have the direction of the largest number of data elements and the number of data elements having directions generally aligned with the direction of the largest number of data elements equals or exceeds a selected percentage of the number of data elements in the Z×

Z direction enhancement scan window.

10. The method of claim 2, wherein after the steps enumerated therein at least some of the data elements in the N×

M matrix of data elements contains data representative of an assigned direction from within one of a plurality of directions, wherein each of said plurality of directions is generally equally spaced apart in direction, comprising the further steps of;

selecting a Z×

Z enhanced direction scan window of data elements from within the N×

M matrix, with Z<

N, Z<

M, and having a center data element;

determining for each of the plurality of directions the number of data elements within the enhanced direction scan window which have such directions;

assigning to the center data element of the Z×

Z scan window the direction had by the largest number of data elements when the sum of the number of data elements which have the direction of the largest number of data elements and the number of data elements having directions generally aligned with the direction of the largest number of data elements equals or exceeds a selected percentage of the number of data elements in the Z×

Z direction enhancement scan window.

11. The method of claim 5, wherein after the steps enumerated therein at least some of the data elements in the N×

M matrix of data elements contains data representative of an assigned direction from within one of a plurality of directions, wherein each of said plurality of directions is generally equally spaced apart in direction, comprising the further steps of;

selecting a Z×

Z enhanced direction scan window of data elements from within the N×

M matrix, with Z<

N, Z<

M, and having a center data element;

determining for each of the plurality of directions the number of data elements within the enhanced direction scan window which have such directions;

assigning to the center data element of the Z×

Z scan window the direction had by the largest number of data elements when the sum of the number of data elements which have the direction of the largest number of data elements and the number of data elements having directions generally aligned with the direction of the largest number of data elements equals or exceeds a selected percentage of the number of data elements in the Z×

Z direction enhancement scan window.

12. The method of claim 8, wherein after the steps enumerated therein at least some of the data elements in the N×

M matrix of data elements contains data representative of an assigned direction from within one of a plurality of directions, wherein each of said plurality of directions is generally equally spaced apart in direction, comprising the further steps of;

selecting a Z×

Z enhanced direction scan window of data elements from within the N×

M matrix, with Z<

N, Z<

M, and having a center data element;

determining for each of the plurality of directions the number of data elements with the enhanced direction scan window which have such directions;

assigning to the center data element of the Z×

Z scan window the direction had by the largest number of data elements when the sum of the number of data elements which have the direction of the largest number of data elements and the number of data elements having directions generally aligned with the direction of the largest number of data elements equals or exceeds a selected percentage of the number of data elements in the Z×

Z direction enhancement scan window.

13. The method of claim 9, further comprising the steps of:

determining that the sum of the number of data elements which have the direction of the largest number of data elements and the number of data elements having directions generally aligned with the direction of the largest number of data elements does not exceed the selected percentage of the number of data elements in the Z×

Z direction enhancement scan window,

assigning the center element the direction of the second largest number of data elements when the sum of the number of data elements which have the direction of the second largest number of data elements and the number of data elements having directions generally aligned with the direction of the second largest number of data elements equals or exceeds a selected percentage of the number of data elements in the Z×

Z direction enhancement scan window.

14. The method of claim 10, further comprising the steps of:

determining that the sum of the number of data elements which have the direction of the largest number of data elements and the number of data elements having directions generally aligned with the direction of the largest number of data elements does not exceed the selected percentage of the number of data elements in the Z×

Z direction enhancement scan window,

assigning the center data element the direction of the second largest number of data elements when the sum of the number of data elements which have the direction of the second largest number of data elements and the number of the data elements having directions generally aligned with the direction of the second largest number of data elements equals or exceeds a selected percentage of the number of data elements in the Z×

Z direction enhancement scan window.

15. The method of claim 11, further comprising the steps of:

determining that the sum of the number of data elements which have the direction of the largest number of data elements and the number of data elements having directions generally aligned with the direction of the largest number of data elements does not exceed the selected percentage of the number of data elements in the Z×

Z direction enhancement scan window,

assigning the center data element the direction of the second largest number of data elements when the sum of the number of data elements which have the direction of the second largest number of data elements and the number of data elements having directions generally aligned with the direction of the second largest number of data elements equals or exceeds a selected percentage of the number of data elements in the Z×

Z direction enhancement scan window.

16. The method of claim 12, further comprising the steps of:

determining that the sum of the number of data elements which have the direction of the largest number of data elements and the number of data elements having directions generally aligned with the direction of the largest number of data elements does not exceed the selected percentage of the number of data elements in the Z×

Z direction enhancement scan window,

assigning the center data element the direction of the second largest number of data elements when the sum of the number of data elements which have the direction of the second largest number of data elements and the number of data elements having directions generally aligned with the direction of the second largest number of data elements equals or exceeds a selected percentage of the number of data elements in the Z×

Z direction enhancement scan window.

17. The method of claim 13, wherein the data elements within the Z×

Z direction enhancement window may have "no direction" as an assigned direction, comprising the further steps of;

determining by the preceding steps not to assign to the center data element either the direction of the largest number of data elements or the direction of the second largest number of data elements;

assigning to the center data element the direction of the largest number of data elements if the sum of the number of data elements having the direction of the largest number, and the number of data elements adjacent in direction to the direction of the largest number of data elements, and the number of data elements having "no direction" exceeds a second percentage of the number of data elements in the enhanced direction scan window.

18. The method of claim 14, wherein the data elements within the Z×

Z direction enhancement window may have "no direction" as an assigned direction, comprising the further steps of;

determining by the preceding steps not to assign to the center data element either the direction of the largest number of data elements or the direction of the second largest number of data elements;

assigning to the center data element the direction of the largest number of data elements if the sum of the number of data elements having the direction of the largest number, and the number of data elements adjacent in direction to the direction of the largest number of data elements, and the number of data elements having "no direction" exceeds a second percentage of the number of data elements in the enhanced direction scan window.

19. The method of claim 15, wherein the data elements within the Z×

Z direction enhancement window may have "no direction" as an assigned direction, comprising the further steps of;

determining by the preceding steps not to assign to the center data element either the direction of the largest number of data elements or the direction of the second largest number of data elements;

assigning to the center data elements the direction of the largest number of data elements if the sum of the number of data elements having the direction of the largest number, and the number of data elements adjacent in direction to the direction of the largest number of data elements, and the number of data elements having "no direction" exceeds a second percentage of the number of data elements in the enhanced direction scan window.

20. The method of claim 16, wherein the data elements with the Z×

Z direction enhancement window may have "no direction" as an assigned direction, comprising the further steps of;

determining by the preceding steps not to assign to the center data element either the direction of the largest number of data elements or the direction of the second largest number of data elements;

assigning to the center data element the direction of the largest number of data elements if the sum of the number of data elements having the direction of the largest number, and the number of data elements adjacent in direction to the direction of the largest number of data elements, and the number of data elements having "no direction" exceeds a second percentage of the number of data elements in the enhanced direction scan window.

21. The method of claim 17, comprising the further steps of:

determining by the preceding steps not to assign either the direction of the largest number of data elements or the direction of the second largest number of data elements;

assigning to the center data element of the Z×

Z direction enhancement scan window the direction of the second largest number of data elements if the sum of the number of data elements having the direction of the second largest number of data elements, and the number of data elements adjacent in direction to the direction of the second largest number of data elements, and the number of data elements having a direction of "no direction" exceeds the second percentage of the number of data elements in the Z×

Z direction enhancement scan window.

22. The method of claim 18, comprising the further steps of:

determining by the preceding steps not to assign either the direction of the largest number of data elements or the direction of the second largest number of data elements;

assigning to the center data element of the Z×

Z direction enhancement scan window the direction of the second largest number of data elements if the sum of the number of data elements having the direction of the second largest number of data elements, and the number of data elements adjacent in direction to the direction of the second largest number of data elements, and the number of data elements having a direction of "no direction" exceeds the second percentage of the number of data elements in the Z×

Z direction enhancement scan window.

23. The method of claim 19, comprising the further steps of:

determining by the preceding steps not to assign either the direction of the largest number of data elements or the direction of the second largest number of data elements;

assigning to the center data element of the Z×

Z direction enhancement scan window the direction of the second largest number of data elements if the sum of the number of data elements having the direction of the second largest number of data elements, and the number of data elements adjacent in direction to the direction of the second largest number of data elements, and the number of data elements having a direction of "no direction" exceeds the second percentage of the number of data elements in the Z×

Z direction enhancement scan window.

24. The method of claim 20, comprising the further steps of:

determining by the preceding steps not to assign either the direction of the largest number of data elements or the direction of the second largest number of data elements;

assigning to the center data element of Z×

Z direction enhancement scan window the direction of the second largest number of data elements if the sum of the number of data elements having the direction of the second largest number of data elements, and the number of data elements adjacent in direction to the direction of the second largest number of data elements, and the number of data elements having a direction of "no direction" exceeds the second percentage of the number of data elements in the Z×

Z direction enhancement scan window.

25. The method of claim 21, comprising the further steps of:

determining by the preceding steps not to assign either the direction of the largest number of data elements or the directions of the second largest number of data elements;

assigning to the center data element of the Z×

Z direction enhancement scan window the average direction of the directions of the largest and second largest number of data elements if the sum of the largest and second largest number of data elements exceeds a third selected percentage of the number of data elements in the Z×

Z direction enhancement scan window or the sum of the largest number of data elements and the second largest number of data elements and the number of data elements having direction of "no direction" exceeds a fourth selected percentage of the number of data elements in the Z×

Z direction enhancement scan window, and the difference in direction of the directions of the largest and second largest number of data elements is less than or equal to a selected number of degrees.

26. The method of claim 22, comprising the further steps of:

determining by the preceding steps not to assign either the direction of the largest number of data elements or the directions of the second largest number of data elements;

assigning to the center data element of the Z×

Z direction enhancement scan window the average direction of the directions of the largest and second largest number of data elements if the sum of the largest and second largest number of data elements exceeds a third selected percentage of the number of data elements in the Z×

Z scan window or the sum of the largest number of data elements and the second largest number of data elements and the number of data elements having a direction of "no direction" exceeds a fourth selected percentage of the number of data elements in the Z×

Z direction enhancement scan window, and the difference in direction of the directions of the largest and second largest number of data elements is less than or equal to a selected number of degrees.

27. The method of claim 23, comprising the further steps of:

determining by the preceding steps not to assign either the direction of the largest number of data elements or the directions of the second largest number of data elements;

assigning to the center data element of the Z×

Z direction enhancement scan window the average direction of the directions of the largest and second largest number of data elements if the sum of the largest and second largest number of data elements exceeds a third selected percentage of the number of data elements in the Z×

Z direction enhancement scan window or the sum of the largest number of data elements and the second largest number of data elements and the number of data elements having a direction of "no direction" exceeds a fourth selected percentage of the number of data elements in the Z×

Z direction enhancement scan window, and the difference in direction of the directions of the largest and second largest number of data elements is less than or equal to a selected number of degrees.

28. The method of claim 24, comprising the further steps of:

determining by the preceding steps not to assign either the direction of the largest number of data elements or the directions of the second largest number of data elements;

assigning to the center data element of the Z×

Z direction enhancement scan window the average direction of the directions of the largest and second largest number of data elements if the sum of the largest and second largest number of data elements exceeds a third selected percentage of the number of data elements in the Z×

Z direction enhancement scan window or the sum of the largest number of data elements and the second largest number of data elements and the number of data elements having a direction of "no direction" exceeds a fourth selected percentage of the number of data elements in the Z×

Z direction enhancement scan window, and the difference in direction of the directions of the largest and second largest number of data elements is less than or equal to a selected number of degrees.

29. The method of claim 25, further comprising the steps of:

incrementing the Z×

Z direction enhancement scan window through the N×

M matrix until each data element in the N×

M matrix is a center data element in the Z×

Z direction enhancement scan window.

30. The method of claim 26, comprising the further steps of:

incrementing the Z×

Z direction enhancement scan window through the N×

M matrix until each data element in the N×

M matrix is a center data element in the Z×

Z direction enhancement scan window.

31. The method of claim 27, comprising the further steps of:

incrementing the Z×

Z direction enhancement scan window through the N×

M matrix until each data element in the N×

M matrix is a center data element in the Z×

Z direction enhancement scan window.

32. The method of claim 28, comprising the further steps of:

incrementing the Z×

Z direction enhancement scan window through the N×

M matrix until each data element in the N×

M matrix is a center data element in the Z×

Z direction enhancement scan window.

33. The method of claim 2, wherein at least some of the data elements contain data representative of a feature flow direction, said method further comprising the steps of:

forming an I×

I intensity enhancement scan window of data elements within the N×

M matrix, with I<

N, I<

M, and wherein said intensity enhancement scan window has a center data element,

determining the mean of the intensity values of the intensity data for the data elements within the intensity enhancement scan window;

selecting a plurality of data elements within the intensity enhancement scan window defining a direction slit having the direction represented by the feature flow direction data of the data of the center data element;

modifying the intensity value of the intensity data of the center data element according to the relationship of the intensity data of each data element in the direction slit to the mean, and the relationship of the direction represented by the direction data in each data element in the direction slit to the direction represented by the direction data of the center data element.

36. The method of claim 3, wherein at least some of the data elements contain data representative of a feature flow direction, said method further comprising the steps of:

forming an I×

I intensity enhancement scan window of data elements within the N×

M matrix, with I<

N, I<

M, wherein said intensity enhancement scan window has a center data element;

determining the mean of the intensity values for the data elements within the intensity enhancement scan window;

selecting a plurality of data elements within the intensity enhancement scan window defining a direction slit having the direction represented by the feature flow direction data of the center data element;

modifying the intensity value of the intensity data of the center data element according to the relationship of the intensity data of each data element in the direction slit to the mean, and the relationship of the direction represented by the direction data in each data element in the direction slit to the direction represented by the direction data of the center data element.

37. The method of claim 4, further comprising the steps of a method of enhancing intensity data for each of a plurality of data elements in an N×

M matrix of data elements each including data representative of an intensity within a range of intensities and data representative of a feature flow direction or "no direction," comprising the steps of;

forming an I×

I intensity enhancement window of data elements within the N×

M matrix, with I<

N, I<

M, and having a center data element,

determining the mean of the intensity values of the intensity data for the data elements within the scan window;

selecting a plurality of data elements within the intensity enhancement scan window defining a direction slit having the direction represented by the feature flow direction data of the center data element;

modifying the intensity value of the intensity data of the center data element according to the relationship of the intensity data of each data element in the direction slit to the mean, and the relationship of the direction represented by the direction data in each data element in the direction slit to the direction represented by the direction data of the center data element.

38. The method of claim 5, wherein at least some of the data elements contain data representative of a feature flow direction, said method further comprising the steps of:

forming an I×

I intensity enhancement scan window of data elements within the N×

M matrix, with I<

N, I<

M, wherein said intensity enhancement scan window has a center data element;

determining the mean of the intensity values of the intensity data for the data elements within the intensity enhancement scan window;

selecting a plurality of data elements within the intensity enhancement scan window defining a direction slit having the direction represented by the feature flow direction data of the center data element;

modifying the intensity value of the intensity data of the center data element according to the relationship of the intensity data of each data element in the direction slit to the mean, and the relationship of the direction represented by the direction data in each data element in the direction slit to the direction represented by the direction data of the center data element.

39. The method of claim 6, wherein at least some of the data elements contain data representative of a feature flow direction, said method further comprising the steps of:

forming an I×

I intensity enhancement scan window of data elements within the N×

M matrix, with I<

N, I<

M, wherein said intensity enhancement scan window has a center data element;

determining the mean of the intensity values of the intensity data for the data elements within the intensity enhancement scan window;

selecting a plurality of data elements within the intensity enhancement scan window defining a direction slit having the direction represented by the feature flow direction data of the center data element;

modifying the intensity value of the intensity data of the center data element according to the relationship of the intensity data of each data element in the direction slit to the mean, and the relationship of the direction represented by the direction data in each data element in the direction slit to the direction represented by the direction data of the center data element.

40. The method of claim 9, wherein at least some of the data elements contain data representative of a feature flow direction, said method further comprising the steps of:

forming an I×

I intensity enhancement scan window of data elements within the N×

M matrix, with I<

N, I<

M, wherein said intensity enhancement scan window has a center data element;

determining the mean of the intensity values of the intensity data for the data elements within the intensity enhancement scan window;

selecting a plurality of data elements within the intensity enhancement scan window defining a direction slit having the direction represented by the feature flow direction data of the center data element;

modifying the intensity value of the intensity data of the center data element according to the relationship of the intensity data of each data element in the direction slit to the mean, and the relationship of the direction represented by the direction data in each data element in the direction slit to the direction represented by the direction data of the center data element.

41. The method of claim 10, wherein at least some of the data elements contain data representative of a feature flow direction, said method further comprising the steps of:

forming an I×

I intensity enhancement scan window of data elements within the N×

M matrix, with I<

N, I<

M, wherein said intensity enhancement scan window has a center data element;

determining the mean of the intensity values of the intensity data for the data elements within the intensity enhancement scan window;

selecting a plurality of data elements within the intensity enhancement scan window defining a direction slit having the direction represented by the feature flow direction data of the center data element;

modifying the intensity value of the intensity data of the center data element according to the relationship of the intensity data of each data element in the direction slit to the mean, and the relationship to the direction represented by the direction data in each data element in the direction slit to the direction represented by the direction data of the center data element.

42. The method of claim 11, wherein at least some of the data elements contain data representative of a feature flow direction, said method further comprising the steps of:

forming an I×

I intensity enhancement scan window of data elements within the N×

M matrix, with I<

N, I<

M, wherein said intensity enhancement scan window has a center data element;

determining the mean of the intensity values of the intensity data for the data elements within the intensity enhancement scan window;

selecting a plurality of data elements within the intensity enhancement scan window defining a direction slit having the direction represented by the feature flow direction data of the center data element;

modifying the intensity value of the intensity data of the center data element according to the relationship of the intensity data of each data element in the direction slit to the mean, and the relationship of the direction represented by the direction data in each data element in the direction slit to the direction represented by the direction data of the center data element.

43. The method of claim 12, wherein at least some of the data elements contain data representative of a feature flow direction, said method further comprising the steps of:

forming an I×

I intensity enhancement scan window of data elements within the N×

M matrix, with I<

N, I<

M, wherein said intensity enhancement scan window has a center data element;

determining the mean of the intensity values of the intensity data for the data elements within the intensity enhancement scan window;

selecting a plurality of data elements within the intensity enhancement scan window defining a direction slit having the direction represented by the feature flow direction data of the center data element;

modifying the intensity value of the intensity data of the center data element according to the relationship of the intensity data of each data element in the direction slit to the mean, and the relationship of the direction represented by the direction data in each data element in the direction slit to the direction represented by the direction data of the center data element.

44. The method of claim 13, wherein at least some of the data elements contain data representative of a feature flow direction, said method further comprising the steps of:

forming an I×

I intensity enhancement scan window of data elements within the N×

M matrix, with I<

N, I<

M, wherein said intensity enhancement scan window has a center data element;

determining the mean of the intensity values of the intensity data for the data elements within the intensity enhancement scan window;

selecting a plurality of data elements within the intensity enhancement scan window defining a direction slit having the direction represented by the feature flow direction data of the center data element;

modifying the intensity value of the intensity data of the center data element according to the relationship of the intensity data of each data element in the direction slit to the mean, and the relationship of the direction represented by the direction data in each data element in the direction slit to the direction represented by the direction data of the center data element.

45. The method of claim 14, wherein at least some of the data elements contain data representative of a feature flow direction, said method further comprising the steps of:

forming an I×

I intensity enhancement window of data elements within the N×

M matrix, with I<

N, I<

M, wherein said intensity enhancement scan window has a center data element;

determining the mean of the intensity values of the intensity data for the data elements within the intensity enhancement scan window;

selecting a plurality of data elements within the intensity enhancement scan window defining a direction slit having the direction represented by the feature flow direction data of the center data element;

modifying the intensity value of the intensity data of the center data element according to the relationship of the intensity data of each data element in the directon slit to the mean, and the relationship of the direction represented by the direction data in each data element in the direction slit to the direction represented by the direction data of the center data element.

46. The method of claim 15, wherein at least some of the data elements contain data representative of a feature flow direction, said method further comprising the steps of:

forming an I×

I intensity enhancement window of data elements within the N×

M matrix, with I<

N, I<

M, wherein said intensity enhancement scan window has a center data element;

determining the mean of the intensity values of the intensity data for the data elements within the intensity enhancement scan window;

selecting a plurality of data elements within the intensity enhancement scan window defining a direction slit having the direction represented by the feature flow direction data of the center data element;

modifying the intensity value of the intensity data of the center data element according to the relationship of the intensity data of each data element in the direction slit to the mean, and the relationship of the direction represented by the direction data in each data element in the direction slit to the direction represented by the direction data of the center data element.

47. The method of claim 16, wherein at least some of the data elements contain data representative of a feature flow direction, said method further comprising the steps of:

forming an I×

I intensity enhancement window of data elements within the N×

M matrix, with I<

N, I<

M, wherein said intensity enhancement scan window has a center data element;

determining the mean of the intensity values of the intensity data for the data elements within the intensity enhancement scan window;

selecting a plurality of data elements within the intensity enhancement scan window defining a direction slit having the direction represented by the feature flow direction data of the center data element;

modifying the intensity value of the intensity data of the center data element according to the relationship of the intensity data of each data element in the direction slit to the mean, and the relationship of the directon represented by the direction data in each data element in the direction slit to the direction represented by the direction data of the center data element.

48. The method of claim 17, wherein at least some of the data elements contain data representative of a feature flow direction, said method further comprising the steps of:

forming an I×

I intensity enhancement window of data elements within the N×

M matrix, with I<

N, I<

M, wherein said intensity enhancement scan window has a center data element;

determining the mean of the intensity values of the intensity data for the data elements within the intensity enhancement scan window;

selecting a plurality of data elements within the intensity enhancement scan window defining a direction slit having the direction represented by the feature flow direction data of the center data element;

modifying the intensity value of the intensity data of the center data element according to the relationship of the intensity data of each data element in the direction slit to the mean, and the relationship of the direction represented by the direction data in each data element in the direction slit to the direction represented by the direction data of the center data element.

49. The method of claim 18, wherein at least some of the data elements contain data representative of a feature flow direction, said method further comprising the steps of:

forming an I×

I intensity enhancement window of data elements within the N×

M matrix, with I<

N, I<

M, wherein said intensity enhancement scan window has a center data element;

determining the mean of the intensity values of the intensity data for the data elements within the intensity enhancement scan window;

selecting a plurality of data elements within the intensity enhancement scan window defining a direction slit having the direction represented by the feature flow direction data of the center data element;

modifying the intensity value of the intensity data of the center data element according to the relationship of the intensity data of each data element in the direction slit to the mean, and the relationship of the direction represented by the direction data in each data element in the direction slit to the direction represented by the direction data of the center data element.

50. The method of claim 19, wherein at least some of the data elements contain data representative of a feature flow direction, said method further comprising the steps of:

forming an I×

I intensity enhancement window of data elements within the N×

M matrix, with I<

N, I<

M, wherein said intensity enhancement scan window has a center data element;

determining the mean of the intensity values of the intensity data for the data elements within the intensity enhancement scan window;

selecting a plurality of data elements within the intensity enhancement scan window defining a direction slit having the direction represented by the feature flow direction data of the center data element;

modifying the intensity value of the intensity data of the center data element according to the relationship of the intensity data of each data element in the direction slit to the mean, and the relationship of the direction represented by the direction data in each data element in the direction slit to the direction represented by the direction data of the center data element.

51. The method of claim 20, wherein at least some of the data elements contain data representative of a feature flow direction, said method further comprising the steps of:

forming an I×

I intensity enhancement window of data elements within the N×

M matrix, with I<

N, I<

M, wherein said intensity enhancement scan window has a center data element;

determining the mean of the intensity values of the intensity data for the data elements within the intensity enhancement scan window;

selecting a plurality of data elements within the intensity enhancement scan window defining a direction slit having the direction represented by the feature flow direction data of the center data element;

modifying the intensity value of the intensity data of the center data element according to the relationship of the intensity data of each data element in the direction slit to the mean, and the relationship of the direction represented by the direction data in each data element in the direction slit to the directon represented by the direction data of the center data element.

52. The method of claim 24, wherein at least some of the data elements contain data representative of a feature flow direction, said method further comprising the steps of:

forming an I×

I intensity enhancement window of data elements within the N×

M matrix, with I<

N, I<

M, wherein said intensity enhancement scan window has a center data element;

determining the mean of the intensity values of the intensity data for the data elements within the intensity enhancement scan window;

selecting a plurality of data elements within the intensity enhancement scan window defining a direction slit having the direction represented by the feature flow direction data of the center data element;

modifying the intensity value of the intensity data of the center data element according to the relationship of the intensity data of each data element in the direction slit to the mean, and the relationship of the direction represented by the direction data in each data element in the direction slit to the direction represented by the direction data of the center data element.

53. The method of claim 28, wherein at least some of the data elements contain data representative of a feature flow direction, said method further comprising the steps of:

forming an I×

I intensity enhancement window of data elements within the N×

M matrix, with I<

N, I<

M, wherein said intensity enhancement scan window has a center data element;

determining the mean of the intensity values of the intensity data for the data elements within the intensity enhancement scan window;

selecting a plurality of data elements within the intensity enhancement scan window defining a direction slit having the direction represented by the feature flow direction data of the center data element;

modifying the intensity value of the intensity data of the center data element according to the relationship of the intensity data of each data element in the direction slit to the mean, and the relationship of the direction represented by the direction data in each data element in the direction slit to the direction represented by the direction data of the center data element.

54. The method of claim 32, wherein at least some of the data element contain data representative of a feature flow direction, said method further comprising the steps of:

forming an I×

I intensity enhancement window of data elements within the N×

M matrix, with I<

N, I<

M, wherein said intensity enhancement scan window has a center data element;

determining the mean of the intensity values of the intensity data for the data elements within the intensity enhancement scan window;

selecting a plurality of data elements within the intensity enhancement scan window defining a direction slit having the direction represented by the feature flow direction data of the center data element;

modifying the intensity value of the intensity data of the center data element according to the relationship of the intensity data of each data element in the direction slit to the mean, and the relationship of the direction represented by the direction data in each data element in the direction slit to the direction represented by the direction data of the center data element.

55. The method of claim 54, further comprising the steps of:

for each of the data elements in the direction slit having the direction represented by the direction data in the center data element, where the center data element has other than "no direction" as its direction data, comparing the intensity data to an augmented mean, which is a mean proportioned to the mean of the intensity data of all the data elements in the intensity enhancement scan window, and comparing the direction represented by the direction data of the center data element to the direction represented by the direction data of each respective direction slit data element;

generating an intensity correction with a magnitude dependent upon the absolute difference between the intensity data of the respective direction slit data element and the augmented mean and, when the direction represented by the direction data of the respective slit data element is within a selected number of degrees from the direction represented by the direction data of the center data element, having a sign to decrease the value of the intensity data of the center data element when the intensity of the respective direction slit data element exceeds the augmented mean, and having a sign to increase the intensity of the center data element when the intensity of the respective direction slit data element is less than the augmented mean, and, when the direction represented by the direction data of the respective direction slit data element is not within a selected number of degrees from the direction represented by the direction data of the center data element, having a sign to increase the value of the intensity data of the center data element when the intensity of the respective direction slit data element exceeds the augmented mean, and having a sign to decrease the intensity of the center data element when the value of the intensity of the respective direction slit data element is less than the augmented mean.

56. The method of claim 55, further comprising the steps of:

incrementing the I×

I intensity enhancement scan window through the N×

M matrix until each data element in the N×

M matrix is a center data element in the I×

I direction enhancement scan window.

57. The method of claim 56, comprising the further steps of:

storing the enhanced intensity data in a memory having a memory location corresponding to each of the column and row positions of the N×

M matrix.

58. The method of claim 57, further comprising the steps of:

reading the memory locations having the enhanced intensity data to form a new N×

M matrix of data elements and repeating the previous method steps.

72. The method of claim 1, comprising the further steps of:

estimating the dynamic range of intensity values for a matrix of data elements each having its own intensity value represented by n bits of digital data, within the range I_min to I_max, wherein the dynamic range is defined as the difference between the intensity value above which lie a first percentile of the number of data elements and the intensity value below which lie a second percentile of the number of data elements comprising;

scanning the data elements and determining the s most significant bits of the n digital data bits for each data element;

forming a histogram containing s groups of bins formed according to a hierarchy of most significant bits, and each group containing the possible combinations of the s, s-1, s-2 . . . , s-(s-1) most significant bits for the respective group of bins, with 0 in the least significant bit location of the combination corresponding to each bin of the respective group;

comparing the number contained in the bin of the first group of said s groups of bins, corresponding to the number of data elements which have the most significant bit in their intensity value set to 0, i.e., which are less than (I_max -I_min)/2, with the number of data elements constituting the first percentile, to determine if the number in the bin of the first group is equal to or greater than the number constituting the first percentile, and comparing the number contained in the bin of the first group with the number of data elements constituting the second percentile, to determine if the number in the bin of the first group is equal to or greater than the number constituting the second percentile;

selecting a first percentile selected bin and a second percentile selected bin from the second group of said s group of bins, based upon the results of each of the preceding comparisons, such that, if the number in the bin of the first group of bins is equal to or greater than the respective percentile, a most significant bit of 0 is retained and the bin in the second group having a 0 for the most significant bit is selected for comparison, and if the number in the bin of the first group is less than the respective percentile a most significant bit of 1 is retained, and the bin in the second group having a 1 in the most significant bit is selected for comparison;

comparing the number in the first percentile selected bin in the second group with the first percentile, to determine if the number in the first percentile selected bin is equal to or greater than the first percentile, and, if so, retaining an 0 as the second most significant bit, and selecting for comparison the bin in the third group of said s group of bins having the retained value for the most significant bit as 0 as the second most significant bit, and, if not, retaining a 1 as the next most significant bit, and selecting the bin in the third group having the retained value for the immediately more significant bit and a 1 as the next most significant bit, and comparing the number in the second percentile selected bin in the second second group with the second percentile to determine if the number in the second percentile selected bin in the second group is equal to or greater than the second percentile and, if so, retaining a 0 as the second most significant bit and selecting for comparison the bin in the third group having the retained value for the immediately more significant bit and 0 in the second most significant bit, and, if not, retaining a 1 as the next most significant bit, and selecting the bin in the third group having the retained value for the immediately more significant bit and a 1 as the second next most significant bit;

repeating such comparison and selection for each successive group of the s groups respecting the lesser significant bits to determine the state of s bits of data representative of the s most significant bits of the estimated intensity of the first and of the second percentiles.

73. The method of claim 9, comprising the further steps of:

estimating the dynamic range of intensity values for a matrix of data elements each having its own intensity value represented by n bits of digital data, within the range I_min to I_max, wherein the dynamic range is defined as the difference between the intensity value above which lie a first percentile of the number of data elements and the intensity value below while lie a second percentile of the number of data elements comprising;

scanning the data elements and determining the s most significant bits of the n digital data bits for each data element;

forming a histogram containing s groups of bins formed according to a hierarchy of most significant bits, and each group containing the possible combinations of the s, s-1, s-2 . . . , s-(s-1) most significant bits for the respective group of bins, with 0 in the least significant bit location of the combination corresponding to each bin of the respective group;

comparing the number contained in the bin of the first group of said s group of bins, corresponding to the number of data elements which have the most significant bit in their intensity value set to 0, i.e., which are less than (I_max -I_min)/2, with the number of data elements constituting the first percentile, to determine if the number in the bin of the first group is equal to or greater than the number constituting the first percentile, and comparing the number contained in the bin of the first group with the number of data elements constituting the second percentile, to determine if the number in the bin of the first group is equal to or greater than the number constituting the second percentile;

selecting a first percentile selected bin and a second percentile selected bin from the second group of said s group of bins, based upon the results of each of the preceding comparisons, such that, if the number in the bin of the first group of bins is equal to or greater than the respective percentile, a most significant bit of 0 is retained and the bin in the second group having a 0 for the most significant bit is selected for comparison, and if the number in the bin of the first group is less than the respective percentile a most significant bit of 1 is retained, and the bin in the second group having a 1 in the most significant bit is selected for comparison;

comparing the number in the first percentile selected bin in the second group with the first percentile, to determine if the number in the first percentile selected bin is equal to or greater than the first percentile, and, if so, retaining an 0 as the second most significant bit, and selecting for comparison the bin in the third group of said s group of bins having the retained value for the most significant bit as 0 as the second most significant bit, and, if not, retaining as 1 as the next most significant bit, and selecting the bin in the third group having the retained value for the immediately more significant bit and a 1 as the next most significant bit, and comparing the number in the second percentile selected bin in the second second group with the second percentile to determine if the number in the second percentile selected bin in the second group is equal to or greater than the second percentile and, if so, retaining a 0 as the second most significant bit and selecting for comparison the bin in the third group having the retained value for the immediately more significant bit and 0 in the second most significant bit, and, if not, retaining a 1 as the next most significant bit, and selecting the bin in the third group having the retained value for the immediately more significant bit and a 1 as the second next most significant bit;

repeating such comparison and selection for each successive group of the s groups respecting the lesser significant bits to determine the state of s bits of data representative of the s most significant bits of the estimated intensity of the first and of the second percentiles.

74. The method of claim 1, comprising the further steps of:

updating a compilation of the data contained in the data elements within a matrix of data elements, forming a scan window having a center data element, while incrementing the matrix through an N×

M matrix of data elements, comprising the steps of;

incrementing the scan window through the N×

M matrix column by column and row by row;

for a first center data element position, determining the change in the compilation of the data due to the addition of the data in a data element "X" in the same row as a leading corner of the S×

S scan window in the direction of incrementation and immediately adjacent the leading corner in the direction of incrementation, the deletion of the data in a data element "W," in the same column as the data element "X" and S rows below the data element "X," in a dirrection opposite the direction of incrementation, the deletion of the data in a data element "Z," S columns below the data element "X," in a direction opposite the direction of incrementation, and the addition of the data in a data element "Y," S columns and S rows before the data element "X," in a direction opposite the direction of incrementation, in order to form a first delta;

storing the first delta;

for a second center data element position, determining the change in the compilation of the data due to the addition of data elements "X" and "Y," and the deletion of data elements "W" and "Z," positioned as recited in the second proceeding subparagraph with respect to the second center data element position in the N×

M matrix, to form a second delta;

determining the net change to the compilation of the data due to the first delta and the second delta to form a net delta;

modifying the compilation of the data by the net delta.

75. The method of claim 9, comprising the further steps of:

updating a compilation of the data contained in the data elements within a matrix of data elements, forming a scan window having a center data element, while incrementing the matrix through an N×

M matrix of data elements, comprising the steps of;

incrementing the scan window through the N×

M matrix column by column and row by row;

for a first center data element position, determining the change in the compilation of the data due to the addition of the data in a data element "X" in the same row as a leading corner of the S×

S scan window in the direction of incrementation and immediately adjacent the leading corner in the direction of incrementation, the deletion of the data in a data element "W"," in the same column as the data element "X" and S rows before the data element "X," in a direction opposite the direction of incrementation, the deletion of the data in a data element "Z," S columns before the data element "X," in a direction opposite the direction of incrementation, and the addition of the data in a data element "Y," S columns and S rows before the data element "X," in a direction opposite the direction of incrementation, in order to form a first delta;

storing the first delta;

for a second center data element position, determining the change in the compilation of the data due to the addition of data elements "X" and "Y," and the deletion of data elements "W" and "Z," positioned as recited in the second proceeding subparagraph with respect to the second center data element position in the N×

M matrix, to form a second delta;

determining the net change to the compilation of the data due to the first delta and the second delta to form a net delta;

modifying the compilation of the data by the net delta.

77. The method of claim 72, comprising the further steps of:

updating a compilation of the data contained in the data elements within a matrix of data elements, forming a scan window having a center data element, while incrementing the matrix through an N×

M matrix of data elements, comprising the steps of;

incrementing the scan window through the N×

M matrix column by column and row by row;

for a first center data element position, determining the change in the compilation of the data due to the addition of the data in a data element "X" in the same row as a leading corner of the S×

S scan window in the direction of incrementation and immediately adjacent the leading corner in the direction of incrementation, the deletion of the data in a data element "W," in the same column as the data element "X" and S rows before the data element "X," in a direction opposite the direction of incrementation, the deletion of the data in a data element "Z," S columns before the data element "X," in a direction opposite the direction of incrementation, and the addition of the data in a data element "Y," S columns and S rows before the data element "X," in a direction opposite the direction of incrementation, in order to form a first delta;

storing the first delta;

for a second center data element position, determining the change in the compilation of the data due to the addition of data elements "X" and "Y," and the deletion of data elements "W" and "Z," positioned as recited in the second proceeding subparagraph with respect to the second center data element position in the N×

M matrix, to form a second delta;

determining the net change to the compilation of the data due to the first delta and the second delta to form a net delta;

modifying the compilation of the data by the net delta.

89. A method for processing fingerprint images in the form of data elements wherein said method includes the data enhancing method according to claim 1.

90. A method for processing a fingerprint image in the form of data elements wherein said method includes the data enhancing method according to claim 2.

91. A method for processing fingerprint images in the form of data elements wherein said method includes the data enhancing method according to claim 3.

92. A method for processing fingerprint images in the form of data elements wherein said method includes the data enhancing method according to claim 9.

93. A method for processing fingerprint images in the form of data elements wherein said method includes the data enhancing method according to claim 18.

94. A method for processing fingerprint images in the form of data elements wherein said method includes the data enhancing method according to claim 33.

104. A method for processing fingerprint images in the form of data elements wherein said method includes the data enhancing method according to claim 73.

105. A system for processing fingerprint images in the form of data elements wherein said system includes the data enhancement filter according to claim 77.

34. The method of 33, further comprising the steps of:

for each of the data elements in the direction slit having the direction represented by the direction in the center data element, where the center data element has other than "no direction" as its direction data, comparing the intensity data to an augmented mean, proportional to the mean of the intensity data of all the data elements in the intensity enhancement scan window, and comparing the direction represented by the direction data of the center data element to the direction represented by the direction data of each respective direction slit data element;

generating an intensity correction with a magnitude dependent upon the absolute difference between the intensity data of the respective direction slit data element and the augmented means and, when the direction represented by the direction data of the respective direction slit data element is within a selected number of degrees from the direction represented by the direction data of the center data element, having a sign to decrease the value of the intensity data of the center data element when the intensity of the respective direction slit data element exceeds the augmented mean, and having a sign to increase the intensity of the center data element when the intensity of the respective direction slit data element is less than the augmented mean, and when the direction represented by the direction data of the respective direction slit data element is not within a selected number of degrees from the direction represented by the direction data of the center data element;

having a sign to increase the value of the intensity data of the center data element when the intensity of the respective direction slit data element exceeds the augmented mean, and having a sign to decrease the intensity of the center data element when the value of the intensity of the respective direction slit data elements is less than the augmented mean.

35. The method of claim 34, further comprising the steps of:

incrementing the I×

I intensity enhancement scan window through the N×

M matrix until each data element in the N×

M matrix is a center data element in the I×

I direction enhancement scan window.

95. A method for processing fingerprint images in the form of data elements wherein said method includes the data enhancing method according to claim 34.

59. A data enhancement filter for enhancing data contained in data elements of an N×

M matrix of data elements, each containing data representative of an intensity within a range of intensities, comprising;

scanning means for selecting a Y×

Y direction estimate scan window of data elements, from within the N×

M matrix, with Y<

N, Y<

M, and having a center data element;

slit summing means for determining the cumulative sum of the differences in the intensity value for each of a plurality of pairs of data elements, spaced apart by a selected number of data elements, and each comprising one of the data elements in a respective one of a plurality of slits, containing the center data element, and generally equally separated within the window;

direction estimate means for assigning a direction estimate to the center data element based upon a comparison of the cumulative sum of such differences for each of the respective slits.

60. The apparatus of claim 59, further comprising:

the direction estimate means includes direction estimate logic means for assigning to the center data element the direction of the slit having the lowest cumulative sum of the differences in the intensity values for the plurality of pairs of data elements spaced apart by a selected number of data elements within the slit.

61. The apparatus of claim 60, further comprising:

dynamic range determining means for determining the dynamic range of the intensity values of the data elements in the Y×

Y direction estimate scan window, within which dynamic range are contained a selected percentage of the number of data elements in the direction estimate scan window;

the direction estimate logic means includes means for assigning "no direction" as the direction estimate, unless the dynamic range exceeds a selected threshold value.

62. The apparatus of claim 60, further comprising:

the direction estimate logic means includes slit sum comparator means for comparing the lowest cumulative sum of the differences in intensity values to the cumulative sum of the differences in intensity values for the slits generally aligned in direction with direction of the slit having the lowest cumulative sum, and for comparing the lowest cumulative sum of the differences in intensity values to the cumulative sum of the differences in intensity values for the slits generally not aligned with the direction of the slit having the lowest cumulative sum; and

,

means for assigning to the center data element the direction of the slit having the lowest cumulative sum, unless a first selected number of the generally aligned slits have a cumulative sum which fails to exceed the lowest cumulative sum by a first selected percentage of the lowest cumulative sum or a second selected number of the generally not aligned slits have a cumulative sum of the differences in intensities which fails to exceed the lowest cumulative sum by a second selected percentage.

63. The apparatus of claim 62, further comprising:

incrementing means for incrementing the direction estimate scan window through the N×

M matrix until each data element of the N×

M matrix has been a center data element of the Y×

Y direction estimate scan window.

78. The apparatus of claim 59, further comprising:

means for estimating the dynamic range of intensity values for a matrix of data elements each having its own intensity value represented by n bits of digital data, within the range I_min to I_max, wherein the dynamic range is defined as the difference between the intensity value above which lie a first percentile of the number of data elements and the intensity value below which lie a second percentile of the number of data elements, comprising;

scanning means for scanning the data elements and determining the s most significant bits of the n digital data bits for each data element;

histogram formation means for forming a histogram containing s groups of bins formed according to a hierarchy of most significant bits, and each group containing the possible combinations of the s, s-1, s-2 . . . , s-(s-1) most significant bits for the respective group of bins, with 0 in the least significant bit location of the combination corresponding to each bin of the respective group,

comparator means for comparing the number contained in the bin of the first group of said s group of bins, corresponding to the number of data elements which have the most significant bit in their intensity value set to 0, i.e., which are less than (I_max -I_min)/2, with the number of data elements constituting the first percentile, to determine if the number in the bin of the first group is equal to or greater than the number constituting the first group is equal to or greater than the number constituting the first percentile, and for comparing the number contained in the bin of the first group with the number of data elements constituting the second percentile, to determine if the number in the bin of the first group is equal to or greater than the number constituting the second percentile,

logic means for selecting a first percentile selected bin and a second percentile selected bin from the second group of said s group of bins, based upon the output of the comparator means, such that if the number in the bin in the first group is equal to or greater than the respective percentile a most significant bit of 0 is retained and the bin in the second group having a 0 for the most significant bit is selected for comparison, and if the number in the bin in the first group is less than the respective percentile a most significant bit of 1 is retained, and the bin in the second group having a 1 in the most significant bit is selected for comparison;

comparator and logic means for comparing the number in the first percentile selected bin in the second group with the first percentile, to determine if the number in the first percentile selected bin is equal to or greater than the first percentile, and, if so, retaining an 0 as the next most significant bit, and selecting for comparison the bin in the third group of said s group of bins having the retained value for the most significant bits and 0 as the next most significant, and, if not, retaining a 1 as the next most significant bit, and selecting the bin in the third group having the retained value of the most significant bit and a 1 as the next most significant bit, and for comparing the number in the second percentile selected bin in the second group with the second percentile to determine if the number in the second percentile selected bin in the second group is equal to or greater than the second percentile and, if so, retaining a 0 as the next most significant bit and selecting for comparison the bin in the third group having the retained value for the most significant bit and 0 in the second most significant bit, and, if not, retaining a 1 as the next most significant bit, and selecting the bin in the third group having the retained value for the immediately more significant bit and a 1 in the next most significant bit, and for repeating such comparison and selection for each successive group of the s groups respecting the lesser significant bits to determine the state of s bits of data representative of the s most significant bits of the estimated intensity of the first and of the second percentiles.

81. The apparatus of claim 59, further comprising:

means for updating an operation upon a compilation of data contained in the data elements within an S×

S scan window having a center data element, while incrementing the S×

S scan window through an N×

M matrix of data elements, where N>

S, M>

S, comprising;

scanning means for scanning the content of a data element "X" in the leading row of the S×

S scan window which in the next clock period will enter the S×

S scan window,

storage means for storing the content of a data element "W" in the same column and S rows before the data element "X," in a direction opposite the direction of incrementation;

storage means for storing the content of a data element "Z," S columns before the data element "X," in a direction opposite the direction of incrementation;

storage means for storing the content of data element "Y," S columns and S rows before the data element "X," in a direction opposite the direction of incrementation;

logic means for determining the change in the result of the operation upon the compilation of data in the data elements within the S×

S scan window due to the addition of the data elements "X" and "Y" and the deletion of the data elements "W" and "Z," to form a first delta;

storage means for storing the first delta until the center data element of the S×

S scan window is in the subsequent position in the same column and one row later, in the direction of incrementation, from the position at which the first delta was formed, and in which a new set of data elements "X," "Y," "W" and "Z" exist;

logic means for determining a second delta from the new data elements "X," "Y," "W" and "Z";

accumulating means for accumulating a net delta from the first and second deltas;

updating means for updating the result of the operation upon the compilation of data elements in the subsequent position of the center data element of the S×

S scan window according to the net delta.

83. The apparatus of claim 59, further comprising:

an apparatus for computing slit sums, defined as the sum of the differences in magnitude of data values of selected data elements within a window of data elements positioned generally along one of a plurality of generally equally spaced-apart slit directions through the center data element of the window of data elements, which selected data elements are each spaced apart by a specified number of one or more data elements within the respective slit, comprising;

a plurality of slit sum circuits each associated with a slit corresponding to one of the plurality of directions;

a column scanning means for selecting a particular one or more data elements from within a column of data elements adjacent the window of data elements in a direction of incrementation of the window of data elements to form a selected input of data from one or more data elements into each of the plurality of slit sum circuits;

delay means within each slit sum circuit for holding the input from the column scanning means for a selected number of incrementation periods;

comparator means within each slit sum circuit for comparing the delayed input from the column scanning means to a current selected input from the column scanning means to form a magnitude subtraction between spaced-apart data elements;

a first grouping means within each slit sum circuit for forming the sum of groups of magnitude subtractions output from the comparator means to form the sum of the magnitude substraction for a first side of the respective slit on one side of the center data element for a selected center data element position;

a second grouping means within each slit sum circuit for forming the sum of groups of magnitude subtractions output from the comparator means to form the sum of the magnitude subtraction for groups of magnitude subtractions of the data elements for the second side of the respective slit on the other side of the center data element;

storage means within each slit sum circuit for storing the sum of the magnitude subtractions for the first side of the slit until the output of the second grouping means comprises the sum of the magnitude subtractions for groups of data elements comprising the second side of the respective slit for the selected center data element position;

adder means within each slit sum circuit for adding the output of first and second grouping means to form the slit sum for the selected center data element position.

86. The apparatus of claim 83, further comprising:

the first and second grouping means comprise a single grouping means with an output of the single grouping means comprising the sum of the magnitude subtractions for the first side of the slit at one time period and the sum of the magnitude subtractions for the second side at another time period.

87. The apparatus of claim 59, further comprising:

means for selecting an S×

S scan window of data elements from within the N×

M matrix, with X<

N, S<

M;

means for incrementing the S×

S scan window through the N×

M matrix for a given number of incrementation periods;

scanning and delay circuit means for selecting an abbreviated scan window having a plurality of selected data elements in a selected pattern within a J×

J matrix in the leading corner of the S×

S scan window in the direction of incrementation of with J=.sup.(s-1) /2 with said abbreviated scan window having a leading row and a trailing row;

control means connected to the scanning and delay circuit means for reading the direction data in the incoming data element incremented into the abbreviated scan window and storing that direction data for J-2 incrementation periods, and for controlling the selection of one of the data elements within the abbreviated scan window for each of a plurality of center data element positions within the leading row of the abbreviated scan window, and one of the data elements within the abbreviated scan window for each of the plurality of center data element positions within the trailing row of the abbreviated scan window, according to the direction data in the respective center data element position, such that after a number of incrementation periods, equal to the number of data elements in the slit within the S×

S scan window on one side of the center data element, each of the data elements on one side of the slit having the direction of a respective one of the center data element positions is selected for the center data element positions in the leading row of the abbreviated scan window, and each of the data elements on the other side of the slit having the direction of the respective center data element position is selected for the center data element positions in the trailing row of the abbreviated scan window;

comparator and logic means for comparing the data in the selected data element within the abbreviated scan window to the data in the respective center data element position in the leading row or the trailing row of the abbreviated scan window to form a correction factor based upon the comparison;

correction factor summing means for summing the output of the comparator means for each respective center data element position, to form a correction factor for each respective center data element position;

data enhancement means for applying the correction factor to the data in the respective center data element position.

96. A system for processing fingerprint images in the form of data elements wherein said system includes the data enhancement filter according to claim 59.

97. A system for processing fingerprint images in the form of data elements wherein said system includes the data enhancement filter according to claim 60.

98. A system for processing fingerprint images in the form of data elements wherein said system includes the data enhancement filter according to claim 86.

106. A system for processing fingerprint images in the form of data elements wherein said system includes the data enhancement filter according to claim 81.

108. A system for processing fingerprint images in the form of data elements wherein said system includes the data enhancement filter according to claim 86.

109. A system for processing fingerprint images in the form of data elements wherein said system includes the data enhancement filter according to claim 87.

64. A data enhancement filter for enhancing intensity data for each of a plurality of data elements in an N×

M matrix of data elements each including data representative of an intensity within a range of intensities and data representative of a feature flow direction or "no direction," comprising;

scanning means for selecting an I×

I intensity enhancement window of data elements within the N×

M matrix, with I<

N, I<

M, and having a center data element;

averaging means for determining the mean of the intensity values for the data elements within the scan window;

slit selection means for selecting a plurality of data elements within the intensity enhancement scan window defining a direction slit having the direction represented by the feature flow direction data of the center data element;

intensity enhancement means for modifying the intensity value of the intensity data of the center data element according to the relationship of the intensity data of each data element in the direction slit to the mean, and the relationship of the direction represented by the direction data in each data element in the direction slit to the direction represented by the direction data of the center data element.

65. The apparatus of claim 64, wherein the intensity enhancement means further comprises:

intensity comparator means for comparing, for each of the data elements in the direction slit, where the center data element has other than "no direction" as its direction data, the intensity data of each respective direction slit data element to an augmented means which is a mean proportional to the mean of the intensity data of all the data elements in the intensity enhancement scan window;

direction comparator means for comparing the direction represented by the direction data of the center data element, having a direction other than "no direction," to the direction represented by the direction data of each respective direction slit data element, having other than "no direction";

intensity correction logic means for generating an intensity correction with a magnitude dependent upon the absolute difference between the intensity data in the respective direction slit data element and the augmented mean and, when the direction represented by the direction data of the respective direction slit data element is within a selected number of degrees of the direction represented by the direction data of the center data element, having a sign to decrease the value of the intensity data of the center data element when the intensity of the respective direction slit data element exceeds the augmented mean, and having a sign to increase the intensity of the center data element when the intensity of the respective direction slit data element is less than the augmented mean, and, when the direction represented by the direction data of the respective direction slit data element is not within a selected number of the direction represented by the direction data of the center data element, having a sign to increase the value of the intensity data of the center data element when the intensity of the respective direction slit data element exceeds the augmented mean, and having a sign to decrease the intensity of the center data element when the value of the intensity of the respective direction slit data element is less than the augmented mean.

67. The apparatus of claim 65, further comprising:

incrementing means for incrementing the I×

I intensity enhancement scan window through the N×

M matrix until each data element in the N×

M matrix is a center data element in the I×

I direction enhancement scan window.

69. The apparatus of claim 67, further comprising:

input means from the data storage means to the scanning means for the Y×

Y scan window for reading the memory locations having the enhanced intensity data to form a new N×

M matrix of data elements.

71. The apparatus of claim 69, wherein the first and second grouping means comprise a single grouping means with an output of the single grouping means representing the sum of the magnitude subtractions for the first side of the slit at one time period and the sum of the magnitude subtractions for the second side at another time period.

76. The method of claim 71, comprising the further steps of:

updating a compilation of the data contained in the data elements within a matrix of data elements, forming a scan window having a center data element, while incrementing the matrix through an N×

M matrix of data elements, comprising the steps of;

incrementing the scan window through the N×

M matrix column by column and row by row;

for a first center data element position, determining the change in the compilation of the data due to the addition of the data in a data element "X" in the same row as a leading corner of the S×

S scan window in the direction of incrementation and immediately adjacent the leading corner in the direction of incrementation, the deletion of the data in a data element "W," in the same column as the data element "X" and S rows before the data element "X," in a direction opposite the direction of incrementation, the deletion of the data in a data element "Z," S columns before the data element "X," in a direction opposite the direction of incrementation, and the addition of the data in a data element "Y," S columns and S rows before the data element "X," in a direction opposite the direction of incrementation, in order to form a first delta;

storing the first delta;

for a second center data element position, determining the change in the compilation of the data due to the addition of data elements "X" and "Y," and the deletion of data elements "W" and "Z," positioned as recited in the second proceeding subparagraph with respect to the second center data element position in the N×

M matrix, to form a second delta;

determining the net change to the compilation of the data due to the first delta and the second delta to form a net delta;

modifying the compilation of the data by the net delta.

80. The apparatus of claim 64, further comprising:

means for estimating the dynamic range of intensity values for a matrix of data elements each having its own intensity value represented by n bits of digital data, within the range I_min to I_max, wherein the dynamic range is defined as the difference between the intensity value above which lie a first percentile of the number of data elements and the intensity value below which lie a second percentile of the number of data elements, comprising;

scanning means for scanning the data elements and determining the s most significant bits of the n digital data bits for each data element;

histogram formation means for forming a histogram containing s groups of bins formed according to a hierarchy of most significant bits, and each group containing the possible combinations of the s, s-1, s-2 . . . , s-(s-1) most significant bits for the respective group of bins, with 0 in the least significant bit location of the combination corresponding to each bin of the respective group,

comparator means for comparing the number contained in the bin of the first group of said s group of bins, corresponding to the number of data elements which have the most significant bit in their intensity value set to 0, i.e., which are less than (I_max -I_min)/2, with the number of data elements constituting the first percentile, to determine if the number in the bin of the first group is equal to or greater than the number constituting the first percentile, and for comparing the number contained in the bin of the first group with the number of data elements constituting the second percentile, to determine if the number in the bin of the first group is equal to or greater than the number constituting the second percentile,

logic means for selecting a first percentile selected bin and a second percentile selected bin from the second group of said s group of bins, based upon the output of the comparator means, such that if the number in the bin in the first group is equal to or greater than the respective percentile a most significant bit of 0 is retained and the bin in the second group having a 0 for the most significant bit is selected for comparison, and if the number in the bin in the first group is less than the respective percentile a most significant bit of 1 is retained, and the bin in the second group having a 1 in the most significant bit is selected for comparison;

comparator and logic means for comparing the number in the first percentile selected bin in the second group with the first percentile, to determine if the number in the first percentile selected bin is equal to or greater than the first percentile, and, if so, retaining an 0 as the next most significant bit, and selecting for comparison the bin in the third group of said s group of bins having the retained value for the most significant bits and 0 as the next most significant, and, if not, retaining a 1 as the next most significant bit, and selecting the bin in the third group having the retained value for the most significant bit and a 1 as the next most significant bit, and for comparing the number in the second percentile selected bin in the second group with the second percentile to determine if the number in the second percentile selected bin in the second group is equal to or greater than the second percentile and, if so, retaining a 0 as the next most significant bit and selecting for comparison the bin in the third group having th retained value for the most significant bit and 0 in the second most significant bit, and, if not, retaining a 1 as the next most significant bit, and selecting the bin in the third group having the retained value for the immediately more significant bit and a 1 in the most significant bit, and for repeating such comparison and selection for each successive group of the s groups respecting the lesser significant bits to determine the state of s bits of data representative of the s most significant bits of the estimated intensity of the first and of the second percentiles.

82. The apparatus of claim 64, further comprising:

means for updating an operation upon a compilation of data contained in the data elements within an S×

S scan window having a center data element, while incrementing the S×

S scan window through an N×

M matrix of data elements, where N>

S, M>

S, comprising;

scanning means for scanning the content of a data element "X" in the leading row of the S×

S scan window which in the next clock period will enter the S×

S scan window,

storage means for storing the content of a data element "W" in the same column and S rows before the data element "X," in a direction opposite the direction of incrementation;

storage means for storing the content of a data element "Z," S columns before the data element "X," in a direction opposite the direction of incrementation;

storage means for storing the content of data element "Y," S columns and S rows before the data element "X," in a direction opposite the direction of incrementation;

logic means for determining the change in the result of the operation upon the compilation of data in the data elements within the S×

S scan window due to the addition of the data elements "X" and "Y" and the deletion of the data elements "W" and "Z," to form a first delta;

storage means for storing the first delta until the center data element of the S×

S scan window is in the subsequent position in the same column and one row later, in the direction of incrementation, from the position at which the first delta was formed, and in which a new set of data elements "X," "Y," "W" and "Z" exist;

logic means for determining a second delta from the new data elements "X," "Y," "W" and "Z";

accumulating means for accumulating a net delta from the first and second deltas;

updating means for updating the result of the operation upon the compilation of data elements in the subsequent position of the center data element of the S×

S scan window according to the net delta.

84. The apparatus of claim 64, further comprising:

an apparatus for computing slit sums, defined as the sum of the differences in magnitude of data values of selected data elements within a window of data elements positioned generally along one of a plurality of generally equally spaced-apart slit directions through the center data element of the window of data elements, which selected data elements are each spaced apart by a specified number of one or more data elements within the respective slit, comprising;

a plurality of slit sum circuits each associated with a slit corresponding to one of the plurality of directions;

a column scanning means for selecting a particular one or more data elements from within a column of data elements adjacent the window of data elements in a direction of incrementation of the window of data elements to form a selected input of data from one or more data elements into each of the plurality of slit sum circuits;

delay means within each slit sum circuit for holding the input from the column scanning means for the specified number of incrementation periods;

comparator means within each slit sum circuit for comparing the delayed input from the colummn scanning means to a current selected input from the column scanning means to form a magnitude subtraction between spaced-apart data elements;

a first grouping means within each slit sum circuit for forming the sum of groups of magnitude subtractions output from the comparator means to form the sum of the magnitude subtractions for a first side of the respective slit on one side of the center data element for a selected center data element position;

a second grouping means within each slit sum circuit for forming the sum of groups of magnitude subtractions output from the comparator means to form the sum of the magnitude subtractions for groups of magnitude subtractions of the data elements for the second side of the respective slit on the other side of the center data element;

storage means within each slit sum circuit for storing the sum of the magnitude subtractions for the first side of the slit until the output of the second grouping means comprises the sum of the magnitude subtractions for groups of data elements comprising the second side of the respective slit for the selected center data element position;

adder means within each slit sum circuit for adding the output of first and second grouping means to form the slit sum for the selected center data element position.

85. The apparatus of claim 82, further comprising:

the first and second grouping means comprise a single grouping means with an output of the single grouping means comprising the sum of the magnitude subtractions for the first side of the slit at one time period and the sum of the magnitude subtractions for the second side at another time period.

88. The apparatus of claim 64, further comprising:

means for selecting an S×

S scan window of data elements from within the N×

M matrix, with S<

N, S<

M;

means for incrementing the S×

S scan window through the N×

M matrix for a given number of incrementation periods;

scanning and delay circuit means for selecting an abbreviated scan window having a plurality of selected data elements in a selectrd pattern within a J×

J matrix in the leading corner of the S×

S scan window with J=.sup.(s-1)/ 2, wthh said abbreviated scan window having a leading row and a trailing row;

control means connected to the scanning and delay circuit means for reading the direction data in the incoming data element incremented into the abbreviated scan window and storing that direction data for J-2 incrementation periods, and for controlling the selection of one of the data elements within the abbreviated scan window for each of a plurality of center data element positions within the leading row of the abbreviated scan window, and one of the data elements within the abbreviated scan window for each of the plurality of center data element positions within the trailing row of the abbreviated scan window, according to the direction data in the respective center data element position, such that after a number of incrementation periods, equal to the number of data elements in the slit within the S×

S scan window on one side of the center data element, each of the data elements on one side of the slit having the direction of a respective one of the center data element positions is selected for the center data element positions in the leading row of the abbreviated scan window, and each of the data elements on the other side of the slit having the direction of the respective center data element position is selected for the center data element positions in the trailing row of the abbreviated scan window;

comparator and logic means for comparing the data in the selected data element within the abbreviated scan window to the data in the respective center data element position in the leading row or the trailing row of the abbreviated scan window to form a correction factor based upon the comparison;

correction factor summing means for summing the output of the comparator means for each respective center data element position, to form a correction factor for each respective center data element position;

data enhancement means for applying the correction factor to the data in the respective center data element position.

99. A system for processing fingerprint images in the form of data elements wherein said system includes the data enhancement filter according to claim 64.

100. A system for processing fingerprint images in the form of data elements wherein said system includes the data enhancement filter according to claim 65.

101. A system for processing fingerprint images in the form of data elements wherein said system includes the apparatus for computing slit sums according to claim 69.

103. A method for processing fingerprint images in the form of data elements wherein said method includes the data enhancing method according to claim 71.

107. A system for processing fingerprint images in the form of data elements wherein said system includes the data enhancement filter according to claim 64.

66.

68. The apparatus of claim 66, further comprising:

data storage means for storing the enhanced intensity data in a memory having a memory location corresponding to each of the column and row positions of the N×

M matrix.

70. An apparatus for computing slit sums, defined as the sum of the differences in magnitude of data values of selected data elements within a window of data elements positioned generally along one of a plurality of generally equally spaced apart slit directions through the center data element of the window of data elements, which selected data elements are each spaced apart by a specified number of one or more data elements within the respective slit, the computing of slit sums being accomplished while incrementing the window in a given direction for a specified number of incrementation periods through a matrix of data elements arranged in columns, comprising:

a plurality of slit sum circuits each associated with a slit corresponding to one of the plurality of slit directions;

a column scanning means for selecting a particular one or more data elements from within a column of data elements adjacent the window of data elements in a direction of incrementation of the window of data elements to form a selected input of data from one or more data elements into each of the plurality of slit sum circuits;

delay means within each slit sum circuit for holding the input from the column scanning means for the specified number of incrementation periods;

comparator means within each slit sum circuit for comparing the delayed input from the column scanning means to a current selected input from the column scanning means to form a magnitude subtraction between spaced-apart data elements;

a first grouping means within each slit sum circuit for forming the sum of groups of magnitude subtractions output from the comparator means to form the sum of the magnitude subtractions for a first side of the respective slit on one side of the center data element for a selected center data element position;

a second grouping means within each slit sum circuit for forming the sum of groups of magnitude subtractions output from the comparator means to form the sum of the magnitude subtractions for groups of magnitude subtractions of the data elements for the second side of the respective slit on the other side of the center data element;

storage means within each slit sum circuit for storing the sum of the magnitude subtractions for the first side of the slit until the output of the second grouping means represents the sum of the magnitude subtractions for groups of data elements comprising the second side of the respective slit for the selected center data element position;

adder means within each slit sum circuit for adding the output of first and second grouping means to form the slit sum for the selected center data element position.

102. A system for processing fingerprint images in the form of data elements wherein said system includes the apparatus for computing slit sums according to claim 70.

79. A data enhancement filter for enhancing data contained in data elements in an N×

M matrix of data elements, each containing data representative of a direction from within one of a plurality of directions, generally equally spaced apart in direction, comprising;

scanning means for selecting a Z×

Z enhanced direction scan window of data elements from within the N×

M matrix, with Z<

N, Z<

M, and having a center data element;

compilation and comparison means for determining the number of data elements within the enhanced direction scan window which have each of the plurality of directions;

direction enhancement logic means for assigning to the center data element of the Z×

Z scan window the direction had by the largest number of data elements when the sum of the number of data elements which have the direction of the largest number of data elements and the number of data elements having directions generally aligned with the direction of the largest number of data elements equals or exceeds a selected percentage of the number of data elements in the Z×

Z direction enhancement scan window,

means for estimating the dynamic range of intensity values for a matrix of data elements each having its own intensity value represented by n bits of digital data, within the range of I_min to I_max, wherein the dynamic range is defined as the difference between the intensity value above which lie a first percentile of the number of data elements and the intensity value below which lie a second percentile of the number of data elements, comprising;

scanning means for scanning the data elements and determining the s most significant bits of the n digital data bits for each data element;

histogram formation means for forming a histogram containing s groups of bins formed according to a hierachy of most significant bits, and each group containing the possible combinations of the s, s-1, s-2 . . . , s-(s-1) most significant bits for the representative groups of bins, with 0 in the least significant bit location of the combination corresponding to each bin of the respective group,

comparator means for comparing the number contained in the bin of the first group, corresponding to the number of data elements which have the most significant bit in their intensity value set to 0, i.e., which are less than (I_max -I_min)/2, with the number of data elements constituting the first percentile, to determine if the number in the bin of the first group is equal to or greater than the number constituting the first percentile, and for comparing the number contained in the bin of the first group with the number of data elements constituting the second percentile, to determine if the number in the bin of the first group is equal to or greater than the number constituting the second percentile,

logic means for selecting a first percentile selected bin and a second percentile selected bin from the second group of bins, based upon the output of the comparator means, such that if the number in the bin in the first group is equal to or greater than the respective percentile a most significant bit of 0 is retained and the bin in the second group having a 0 for the most significant bit is selected for comparison, and if the number in the bin in the first group is less than the respective percentile a most significant bit of 1 is retained, and the bin in the second group having a 1 in the most significant bit is selected for comparison;

comparator and logic means for comparing the number in the first percentile selected bin in the second group with the first percentile, to determine if the number in the first percentile selected bin is equal to or greater than the first percentile, and, if so, retaining an 0 as the next most significant bit, and selecting for comparison the bin in the third group having the retained value for the most significant bits and 0 as the next most significant, and, if not, retaining 1 as the next most significant bit, and selecting the bin in the third group having the retained value for the most significant bit and a 1 as the next most significant bit, and for comparing the number in the second percentile selected bin in the second group with the second percentile to determine if the number in the second percentile selected bin in the second group is equal to or greater than the second percentile and, if so, retaining a 0 as the next most significant bit and selecting for comparison the bin in the third group having the retained value for the most significant bit and 0 in the second most significant bit, and, if not, retaining a 1 as the next most significant bit, and selecting the bin in the third group having the retained value for the immediately more significant bit and a 1 in the next most significant bit, and for repeating such comparison and selection for each successive group of the s groups respecting the lesser significant bits to determine the state of s bits of data representative of the s most significant bits of the estimated intensity of the first and of the second percentiles.