Adaptive vision system

• US 5,481,620 A
• Filed: 09/27/1991
• Issued: 01/02/1996
• Est. Priority Date: 09/27/1991
• Status: Expired due to Term

• Current Assignee: E. I. du Pont de Nemours and Company (Corteva, Inc.)
• Original Assignee: E. I. du Pont de Nemours and Company (Corteva, Inc.)
• Inventors: Vaidyanathan, Akhileswar G.
• Primary Examiner(s): Couso, Yon J.
• Assistant Examiner(s): Fox, David
• Application Number:
  US07/767,339
• Time in Patent Office: 1,558 Days
• Field of Search: 382/6, 382/18, 382/48, 382/50, 382/51, 364/555, 377/10, 377/11
• US Class Current: 382/169
• CPC Class Codes: G06T 11/40 Filling a planar surface by adding surface attributes, e.g. colour or texture
• CPC Class Codes: G06T 2207/30024 Cell structures in vitro; Tissue sections in vitro
• CPC Class Codes: G06T 2207/30128 Food products
• CPC Class Codes: G06T 7/0012 Biomedical image inspection
• CPC Class Codes: G06T 7/11 Region-based segmentation
• CPC Class Codes: G06T 7/136 involving thresholding
• CPC Class Codes: G06T 9/20 Contour coding, e.g. using detection of edges
• CPC Class Codes: G06V 10/46 Descriptors for shape, contour or point-related descriptors, e.g. scale invariant feature transform [SIFT] or bags of words [BoW]; Salient regional features colour feature extraction G06V10/56
• CPC Class Codes: G06V 20/695 Preprocessing, e.g. image segmentation

Abstract

The present invention relates to image analysis methods and systems for identifying objects in a background by generating a description, which may be either a histogram or co-occurrence matrix, of the gray level space of the image by using an entropic kernel to recursively analyze the gray level space for candidate objects and validating the presence of valid objects by comparing the candidate object attribute values to a defined set of valid object attribute values contained in a driver. The present invention includes recursive, iterative and parallel processing methods. The methods may be used in a wide variety of industrial inspection techniques, including colony counting and the identification of discrete features in carpets and of pigment elements embedded in a polymer.

Claims

1. A method of identifying at least one valid object having at least one predetermined attribute value in a background, comprising the steps of:

(a) generating an image of the object and the background;

(b) generating a gray level histogram of the image, the gray level histogram having an entropy function;

(c) entropically selecting a threshold gray level such that the entropy function of the histogram is maximized;

(d) searching the image using the entropically selected threshold gray level for at least one candidate object, wherein the candidate object has at least one candidate object attribute value;

(e) validating the candidate object having the valid object predetermined attribute value to identify [the]a valid object;

(f) subdividing the gray level histogram into an upper histogram and a lower histogram using the entropic threshold gray level as defined by step (c) as an upper delimiter and a lower delimiter; and

(g) recursively repeating steps (c)-(f) for each of the upper and lower histograms, wherein the repetition of step (c) selects a next successive entropic threshold gray level, thereby recursively partitioning the gray level histogram until a predetermined minimum number of new valid objects is identified.

3. The method as claimed in claim 1, wherein the predetermined minimum number is zero.

4. The method as claimed in claim 1, wherein the selecting step includes the sub-steps of:

(i) sequentially partitioning the gray level histogram at each gray level into a first and a second partition,

(ii) computing the entropy function for each partition, where the entropy function of the histogram is defined as the sum of the entropy functions for the first and second partitions, and

(iii) selecting the threshold gray level such that the entropy function of the histogram is maximized.

7. The method as claimed in claim 1, wherein the validating step includes the sub-steps of:

(i) calculating the candidate object attribute values, and

(ii) comparing the candidate object attribute values to the valid object predetermined attribute values to validate candidate objects.

8. The method as claimed in claim 7, wherein the calculating sub-step further includes the sub-step of storing the candidate object attribute values.

9. The method as claimed in claim 1, wherein the validating step further includes the sub-step of checking whether the candidate objects are relatively lighter than the background.

10. The method as claimed in claim 1, wherein the validating step further includes the sub-steps of:

(i) inverting the image after step (a), and

(ii) checking whether the candidate objects are relatively darker than the background.

11. The method as claimed in claim 1, wherein the validating step further includes the sub-step of checking for redundancies of the valid object and resolving the redundancies to prevent multiple identification of the valid object.

12. The method as claimed in claim 11, wherein the valid object comprises an homogeneous object and inhomogeneous object and further wherein the redundancy checking sub-step further includes the sub-steps of:

(i) comparing the areas of a plurality of valid objects and designating one of the valid objects as a large valid object and another of the valid objects as a small valid object, and

(ii) determining whether the small valid object is contained within the large valid object as defined by the four extremum points of the large valid object.

13. The method as claimed in claim 12, wherein the redundancy checking sub-step is performed to resolve redundancies in the inhomogeneous objects and further includes the sub-step of deleting the large object when the large object contains more than one small object.

14. The method as claimed in claim 12, wherein the redundancy checking sub-step is performed to resolve redundancies in the inhomogeneous objects and further includes the sub-steps of:

(i) calculating the average edge contrast of the large and small valid objects, and

(ii) deleting the object having the smaller edge contrast when the large valid object contains only one small valid object.

15. The method as claimed in claim 12, wherein the redundancy checking sub-step is performed to resolve redundancies in the homogeneous objects and further includes the sub-steps of:

(i) calculating the average edge contrast of the large and small valid objects, and

(ii) deleting the large object where the average edge contrast of the large object is less than the average edge contrast of the small object and is less than a predetermined minimum edge contrast.

16. The method as claimed in claim 12, wherein the redundancy checking sub-step is performed to resolve redundancies in the homogeneous objects and further includes the sub-steps of:

(i) calculating the edge contrast of the large and small valid objects, and

(ii) deleting the small valid object where the average edge contrast of the large valid object is greater than the average edge contrast of the small valid object and is greater than a predetermined minimum edge contrast.

17. The method as claimed in claim 1, further including the step of performing a final check for redundancies of the valid object and resolving the redundancies to prevent multiple identification of the valid object.

18. The method as claimed in claim 17, wherein the valid objects comprise a plurality of homogeneous objects or a plurality of inhomogeneous objects and further wherein the final redundancy checking step further includes the sub-steps of:

(i) comparing the areas of a plurality of valid objects and designating one of the valid objects as a large valid object and the other of the valid objects as a small valid object, and

(ii) deleting the large valid object when the small valid object and the large valid object overlap.

19. The method as claimed in claim 1, further including the step of filtering the image after step (e), and wherein the valid object comprises a homogenous valid object or an inhomogeneous valid object.

20. The method as claimed in claim 1, further including the step of filtering the image after step (g) and wherein the valid object comprises a homogeneous valid object or an inhomogeneous valid object.

21. The method as claimed in claim 20, wherein the filtering step includes the sub-step of deleting the inhomogeneous valid object if both of the following conditions are met:

(i) the inhomogeneous valid object has an area less than a predetermined minimum area,

(ii) the inhomogeneous valid object is located within a predetermined minimum pixel distance from the edge of the image boundary.

22. The method as claimed in claim 20, wherein the filtering step includes the sub-step of determining when the number of inhomogeneous valid objects retained after the final redundancy check is performed is greater than or equal to a predetermined maximum number.

23. The method as claimed in claim 22, wherein the filtering step further includes the sub-step of deleting the inhomogeneous valid object if:

(i) the inhomogeneous valid object was never contained in another valid object,

(ii) the inhomogeneous valid object has an area less than a predetermined minimum area, and

(iii) at least one of the following conditions is met;

(A) the inhomogeneous valid object is located outside a predetermined minimum pixel distance from the edge of the image boundary, and

(B) the inhomogeneous valid object has an area greater than or equal to a predetermined minimum area.

24. The method as claimed in claim 22, wherein the filtering step further includes the sub-step of deleting the inhomogeneous valid object if:

(i) the inhomogeneous valid object was never contained in another valid object,

(ii) the inhomogeneous valid object has an area greater than or equal to a predetermined minimum area,

(iii) both of the following conditions are met;

(A) the inhomogeneous object has an edge contrast less than a predetermined minimum edge contrast, and

(B) the inhomogeneous object has a center of mass gray level less than the entropically selected threshold gray level for the lower histogram of step (f), and

(iv) at least one of the following conditions is met;

(A) the inhomogeneous valid object is located outside a predetermined minimum pixel distance from the edge of the image boundary, and

(B) the inhomogeneous valid object has an area greater than or equal to a predetermined minimum area.

25. The method as claimed in claim 20, wherein the filtering step includes the sub-step of determining when the number of inhomogeneous valid objects retained after the final redundancy check is performed is greater than a predetermined minimum number and less than a predetermined maximum number.

26. The method as claimed in claim 25 wherein the filtering step includes the sub-step of deleting the inhomogeneous valid object if:

(i) the inhomogeneous object was never contained within another valid object,

(ii) the inhomogeneous object area is less than a predetermined area, and

(iii) at least one of the following conditions is met;

(A) the inhomogeneous valid object is located outside a predetermined minimum pixel distance from the edge of the image boundary, and

(B) the inhomogeneous valid object has an area greater than or equal to a predetermined minimum area.

27. The method as claimed in claim 25, wherein the filtering step includes the sub-step of deleting the inhomogeneous valid object if:

(i) the inhomogeneous valid object was previously contained in another valid object;

(ii) all of the following conditions are met;

(A) the inhomogeneous valid object edge contrast is less than a predetermined minimum edge contrast,

(B) the inhomogeneous valid object center of mass gray level is less than the entropically selected threshold gray level as defined in step (c), and

(C) the inhomogeneous valid object area is less than or equal to a predetermined minimum area, and

(iii) at least one of the following conditions is met;

(A) the inhomogeneous valid object is located outside a predetermined minimum pixel distance from the edge of the image boundary, and

(B) the inhomogeneous valid object has an area greater than or equal to a predetermined minimum area.

28. The method as claimed in claim 25, wherein the filtering step includes the sub-step of deleting the inhomogeneous valid object if:

(i) the inhomogeneous valid object was never contained in another valid object,

(ii) the inhomogeneous valid object area is greater than or equal to a predetermined minimum area,

(iii) both of the following conditions are met;

(A) the inhomogeneous valid object edge contrast is less than a predetermined minimum edge contrast, and

(B) the inhomogeneous object center of mass gray level is less than the entropically selected threshold gray level as defined in step (c), and

(iv) at least one of the following conditions is met;

(A) the inhomogeneous valid object is located outside a predetermined minimum pixel distance from the edge of the image boundary, and

(B) the inhomogeneous valid object has an area greater than or equal to a predetermined minimum area.

29. The method as claimed in claim 20, wherein the filtering step includes the sub-step of determining when the number of inhomogeneous valid objects retained after the final redundancy check is performed is less than or equal to a predetermined minimum number.

30. The method as claimed in claim 29, wherein the filtering step includes the sub-step of deleting the inhomogeneous valid object if:

(i) the object was never contained in another valid object,

(ii) the object area is less than a predetermined minimum area, and

(iii) at least one of the following conditions is met;

(A) the inhomogeneous valid object is located outside a predetermined minimum pixel distance from the edge of the image boundary, and

(B) the inhomogeneous valid object has an area greater than or equal to a predetermined minimum area.

31. The method as claimed in claim 29, wherein the filtering step includes the sub-step of deleting the inhomogeneous valid object if:

(i) the inhomogeneous valid object was never contained in another valid object,

(ii) the inhomogeneous valid object area is greater than or equal to a predetermined minimum area,

(iii) both of the following conditions are met;

(A) the inhomogeneous valid object edge contrast is less than a predetermined minimum edge contrast, and

(B) the inhomogeneous valid object area is greater than a predetermined minimum area and less than a predetermined maximum area, and

(iv) at least one of the following conditions is met;

(A) the inhomogeneous valid object is located outside a predetermined minimum pixel distance from the edge of the image boundary, and

(B) the inhomogeneous valid object has an area greater than or equal to a predetermined minimum area.

32. The method as claimed in claim 29, wherein the filtering step includes the sub-step of deleting the inhomogeneous valid object if:

(i) the inhomogeneous valid object was previously contained in another valid object,

(ii) both of the following conditions are met.

(A) the inhomogeneous valid object center of mass gray level is less than the entropically selected threshold gray level of step (c),

(B) the inhomogeneous object area is greater than a predetermined minimum area, and

(c) the inhomogeneous object area is greater than or equal to a predetermined maximum area,

(iii) at least one of the following conditions is met;

(A) the inhomogeneous valid object is located outside a predetermined minimum pixel distance from the edge of the image boundary, and

(B) the inhomogeneous valid object has an area greater than or equal to a predetermined minimum area.

33. The method as claimed in claim 29, wherein the filtering step includes the sub-step of deleting the inhomogeneous valid object if:

(i) the inhomogeneous valid object was previously contained within another valid object,

(ii) at least one of the following conditions is met;

(A) the inhomogeneous valid object center of mass gray level is greater than or equal to the entropically selected threshold gray level as defined in step (c),

(B) the object area is less than or equal to a predetermined minimum area, and

(C) the inhomogeneous object area is greater than or equal to a predetermined maximum area,

(iii) the inhomogeneous object center of mass gray level is less than the entropically selected threshold gray level for the lower histogram of step (f), and

(iv) at least one of the following conditions is met;

(A) the inhomogeneous valid object is located outside a predetermined minimum pixel distance from the edge of the image boundary, and

(B) the inhomogeneous valid object has an area greater than or equal to a predetermined minimum area.

34. The method as claimed in claim 20, wherein the filtering step includes the sub-step of deleting the homogenous valid object if both of the following conditions are met:

(i) the homogenous valid object is located within a predetermined minimum pixel distance from the edge of the image boundary, and

(ii) the homogeneous valid object has an area less than a predetermined minimum area.

35. The method as claimed in claim 20, wherein the filtering step includes the sub-step of deleting the homogeneous valid object if:

(i) the homogeneous valid object has an area smaller than a predetermined minimum area, and

(ii) at least one of the following conditions is met;

(A) the homogeneous valid object is located outside a predetermined minimum pixel distance from the edge of the image boundary, and

(B) the homogenous valid object area is greater than or equal to a predetermined minimum area.

36. The method as claimed in claim 20, wherein the filtering step includes the sub-step of deleting the homogeneous valid object if:

(i) the homogeneous valid object has an area greater than or equal to a predetermined minimum area, and

(ii) at least one of the following conditions is met;

(A) the homogenous object is located outside a predetermined minimum pixel distance from the edge of the image boundary, and

(B) the homogenous object area is greater than or equal to a predetermined minimum area, and

(iii) both of the following conditions are met;

(A) the homogeneous object edge contrast is less than a predetermined minimum edge contrast, and

(B) the homogeneous object area is less than a predetermined minimum area.

37. The method as claimed in claim 20, wherein the filtering step includes the sub-step of deleting the homogeneous valid object if:

(i) the homogeneous valid object has an area greater than or equal to a predetermined minimum area,

(ii) at least one of the following conditions is met;

(A) the homogenous object is located outside a predetermined minimum pixel distance from the edge of the image boundary, and

(B) the homogenous object area is greater than or equal to a predetermined minimum area,

(iii) at least one of the following conditions is met;

(A) the homogeneous object edge contrast is greater than or equal to a predetermined minimum edge contrast, and

(B) the homogenous object area is greater than or equal to a predetermined maximum area, and

(iv) both of the following conditions are met;

(A) the homogenous object center of mass gray level is less than the entropically selected threshold gray level of step (c), and

(B) the homogeneous object edge contrast is less than a predetermined minimum edge contrast.

38. The method as claimed in claim 20, wherein the filtering step includes the sub-step of deleting the homogeneous valid object if:

(i) the homogeneous valid object has an area greater than or equal to a predetermined minimum area,

(ii) at least one of the following conditions is met;

(A) the homogenous object is located outside a predetermined minimum pixel distance from the edge of the image boundary, and

(B) the homogenous object area is greater than or equal to a predetermined minimum area,

(iii) at least one of the following conditions is met;

(A) the homogeneous object edge contrast is greater than or equal to a predetermined minimum edge contrast, and

(B) the homogenous object area is greater than or equal to a predetermined maximum area,

(iv) at least one of the following conditions is met;

(A) the homogenous object center of mass gray level is greater than or equal to the

entropically selected threshold gray level of step (c), and

(B) the homogeneous object edge contrast is greater than or equal to a predetermined minimum edge contrast, and

(v) both of the following conditions are met;

(A) the homogenous object shape factor is less than a predetermined shape factor, and

(B) the homogenous object edge contrast is less than a predetermined minimum edge contrast.

39. The method as claimed in claim 20, wherein the filtering step includes the sub-step of deleting the homogeneous valid object if:

(i) the homogeneous valid object has an area greater than or equal to a predetermined minimum area,

(ii) at least one of the following conditions is met;

(A) the homogenous object is located outside a predetermined minimum pixel distance from the edge of the image boundary, and

(B) the homogenous object area is greater than or equal to a predetermined minimum area,

(iii) at least one of the following conditions is met;

(A) the homogeneous object edge contrast is greater than or equal to a predetermined minimum edge contrast, and

(B) the homogenous object area is greater than or equal to a predetermined maximum area,

(iv) at least one of the following conditions is met;

(A) the homogenous object center of mass gray level is greater than or equal to the entropically selected threshold gray level of step (c),

(B) the homogeneous object edge contrast is greater than or equal to a predetermined minimum edge contrast, and

(v) at least one of the following conditions is met;

(A) the homogenous object shape factor is greater than or equal to a predetermined shape factor, and

(B) the homogenous object edge contrast is greater than or equal to a predetermined minimum edge contrast, and

(vi) the homogenous object edge contrast is less than a predetermined absolute minimum edge contrast.

40. The method as claimed in claim 1, wherein the image comprises at least one isolated object identified by the degree of circularity thereof and at least one candidate clump of valid objects, the candidate clump having an area greater than the mean area of the isolated valid objects and a shape factor less than a predetermined clump shape factor, further including the steps of:

(i) identifying the candidate clump, and

(ii) determining the number of valid objects in the candidate clump.

41. The method as claimed in claim 40, wherein the determining sub-step includes the sub-steps of:

(i) creating a distance buffer, wherein the distance buffer comprises a plurality of distance values representing the distance of each perimeter point of the candidate clump to the center of mass of the candidate clump, the distance values including a plurality of valleys and peaks;

(ii) searching the buffer for a first valley using a hysteresis-based, valley-finding algorithm,

(iii) shifting the distance buffer by a distance equal to the distance to the first valley from a first distance value as defined by a first predetermined point on the perimeter of the candidate clump, and

(iv) counting the number of peaks in the shifted buffer using a hysteresis-based, peak-finding algorithm to calculate a first value for the number of valid objects in the clump.

42. The method as claimed in claim 41, wherein the counting sub-step further includes the sub-step of dividing the number of peaks by two to obtain a result, rounding the result up to the next higher integer and adding one to the next higher integer.

43. The method as claimed in claims 41, wherein the determining step further includes the sub-step of calculating the area of the candidate clump and dividing the area of the candidate clump by the mean area of isolated valid object to calculate a second value for the number of valid objects in the clump.

44. The method as claimed in claims 43, wherein the determining step further includes the sub-step of calculating the area of the candidate clump and dividing the area of the candidate clump by the mean area of the isolated valid objects minus one-half the standard deviation of the mean area of the isolated valid object to calculate a third value for the number of valid objects in the clump.

45. The method as claimed in claim 44, wherein the determining step further includes the sub-steps of:

(i) computing the average value of the second and third values,

(ii) comparing the average value to the first value, and

(iii) selecting the average value as the number of valid objects if the average value is greater than the first value.

46. The method as claimed in claim 44, wherein the determining step further includes the sub-steps of:

(i) computing the average value of the second and third values,

(ii) comparing the average value to the first value, and

(iii) selecting the first value as the number of valid objects if the average value is less than or equal to the first value.

47. The method as claimed in claim 44, wherein the determining step further includes the sub-steps of:

(i) computing the average value of the second and third values,

(ii) comparing the average value to the first value, and

(iii) selecting the average value as the number of valid objects if the first value is greater than three.

48. The method as claimed in claim 44, wherein the determining step further includes the sub-steps of:

(i) computing the average value of the second and third values,

(ii) comparing the average value to the first value, and

(iii) selecting the first value as the number of valid objects if the average value is less than or equal to three.

2. A method of identifying at least one valid discrete feature in a carpet, the discrete feature having at least one valid feature predetermined attribute value, comprising the steps of:

(a) generating an image of the feature and the carpet;

(b) generating a gray level histogram of the image, the gray level histogram having an entropy function;

(c) entropically selecting a threshold gray level such that the entropy function of the histogram is maximized;

(d) searching the image using the entropically selected threshold gray level for at least one candidate feature, wherein the candidate feature has at least one candidate feature attribute value;

(e) validating the candidate feature having the valid feature predetermined attribute value to identify a valid feature;

(f) subdividing the gray level histogram into an upper histogram and a lower histogram using the entropically selected threshold gray level as defined in step (c) as an upper delimiter and a lower delimiter; and

(g) recursively repeating steps (c)-(f) for each of the upper and lower histograms, wherein the repetition of step (c) selects a next successive entropic threshold gray level, thereby recursively partitioning the gray level histogram until a predetermined minimum number of new valid features is identified.

6. The method as claimed in claim 2, wherein the searching step includes the sub-steps of:

(i) scanning a portion of the image for at least one candidate object using the entropically selected threshold gray level, wherein the portion of the image comprises a plurality of pixels and each pixel has a gray level value less than the upper delimiter plus a predetermined increment, and

(ii) tracing the candidate object having boundary gray levels determined by the entropically selected threshold gray level.

5. A method of identifying at least one valid pigment element embedded in a polymer, the pigment element having at least one valid pigment element predetermined attribute value, comprising the steps of:

(a) generating an image of the pigment element and the polymer;

(b) generating a gray level histogram of the image, the gray level histogram having an entropy function;

(c) entropically selecting a threshold gray level such that the entropy function of the histogram is maximized;

(d) searching the image using the entropically selected threshold gray level for at least one candidate pigment element, wherein the candidate pigment element has at least one candidate pigment element attribute value;

(e) validating the candidate pigment element having the valid pigment element predetermined attribute value to identify a valid pigment element;

(f) subdividing the gray level histogram into an upper histogram and a lower histogram using the entropically selected threshold gray level as defined in step (c) as an upper delimiter and a lower delimiter; and

(g) recursively repeating steps (c)-(f) for each of the upper and lower histograms, wherein the repetition of step (c) selects a next successive entropic threshold gray level, thereby recursively partitioning the gray level histogram until a predetermined minimum number of new valid elements is identified.

49. The method as claimed in 45 or 47, further including the step of tracing the clump in a singular color indicative of the number of valid objects in the candidate clump.

50. The method as claimed in 46 or 48, further including the step of tracing the clump in a color plurality of colors indicative of the number of valid objects in the candidate clump, wherein all the points of the perimeter of the clump of valid objects between the adjacent valleys on the distance buffer are traced in the same color.

51. A system for identifying at least one valid object having at least one predetermined attribute value in a background, comprising:

(a) means for generating an image of the object and the background; and

(b) computer means including;

(i) a driver for storing the definition of the valid object, and

(ii) an entropic kernel for;

(A) generating a gray level histogram of the image,

(B) entropically selecting a threshold gray level such that the entropy function of the histogram is maximized,

(C) searching the image using the entropically selected threshold gray level for at least one candidate object, wherein the candidate object has at least one candidate object attribute value,

(D) validating the candidate object having the valid object predetermined attribute value to identify a valid object,

(E) subdividing the gray level histogram into an upper histogram and a lower histogram using the entropic threshold gray level as defined by step (B) as an upper delimiter and a lower delimiter, and

(F) recursively repeating steps (B)-(E) for each of the upper and lower histograms, wherein the repetition of step (B) selects a next successive entropic threshold gray level, thereby recursively partitioning the gray level histogram until a predetermined minimum number of new valid objects is identified.

52. The system as claimed in claim 51, wherein the image generating means comprises a camera.

53. The system as claimed in claim 51, wherein the computer means further includes a frame grabber for storing the image.

54. The system as claimed in claim 51, wherein the computer means further includes a video digitizer for

55. The system as claimed in claim 51, wherein the computer means further includes a post-scan filter for removing redundancies in overlapping valid objects.

56. A method of counting at least one valid biological colony having at least one predetermined attribute value in a background, comprising the steps of:

(a) generating an image of the colony and the background;

(b) generating a gray level histogram of the image, the gray level histogram having an entropy function;

(c) entropically selecting a threshold gray level such that the entropy function of the histogram is maximized;

(d) searching the image using the entropically selected threshold gray level for at least one candidate colony, wherein the candidate colony has at least one candidate colony attribute value;

(e) validating the candidate colony having the valid colony predetermined attribute value to identify a valid colony;

(f) subdividing the gray level histogram into an upper histogram and a lower histogram using the entropically selected threshold gray level as defined in step (c) as an upper delimiter and a lower delimiter; and

(g) recursively repeating steps (c)-(f) for each of the upper and lower histograms, wherein the repetition of step (c) selects a next successive entropic threshold gray level, thereby recursively partitioning the gray level histogram until a predetermined minimum number of new valid colonies is counted.

57. The method as claimed in claim 56, wherein the selecting step includes the sub-steps of:

(i) sequentially partitioning the gray level histogram at each gray level into a first and a second partition,

(ii) computing the entropy function for each partition, where the entropy function of the histogram is defined as the sum of the entropy functions for the first and second partitions, and

(iii) selecting a threshold gray level such that the entropy function of the histogram is maximized.

58. The method as claimed in claim 56, wherein the searching step includes the sub-steps of:

(i) scanning the image for at least one candidate colony using the entropically selected threshold gray level, and

(ii) tracing the candidate colony having at least one boundary gray level determined by the entropically selected threshold gray level.

59. The method as claimed in claim 56, wherein the validating step includes the sub-steps of:

(i) calculating the candidate colony attribute values, and

(ii) comparing the candidate colony attribute values to the valid colony predetermined attribute values to validate the candidate colony.

60. The method as claimed in claim 56, wherein the validating step further includes the sub-step of checking for redundancies of the validated colony and resolving the redundancies to prevent multiple identification of the valid colony.

61. The method as claimed in claim 56, further including the step of performing a final check for redundancies of the valid object and resolving the redundancies to prevent multiple identification of the valid object.

62. The method as claimed in claim 56, further including the step of filtering the image after step g.

63. The method as claimed in claim 56, further including the step of determining the number of valid colonies in a candidate clump, wherein the candidate clump comprises a plurality of valid colonies.

64. A method of identifying at least one valid object having at least one predetermined attribute value in a background, comprising the steps of:

(a) generating an image of the object and the background;

(b) generating a histogram representative of the entire image, the gray level histogram having an entropy function;

(c) generating a plurality of gray level upper and lower histograms of the image from the histogram, each gray level upper and lower histogram having a respective entropy function;

(d) automatically selecting a threshold gray level value for each upper and lower histogram such that the entropy function of each histogram is maximized;

(e) searching the image using the automatically selected threshold gray level value for each upper and lower histogram for at least one candidate object, wherein the candidate object has at least one candidate object attribute value; and

(f) validating the candidate object having the valid object predetermined attribute value to identify the valid object.