Adaptive image acquisition system

• US 4,949,391 A
• Filed: 05/08/1987
• Issued: 08/14/1990
• Est. Priority Date: 09/26/1986
• Status: Expired due to Fees

• Current Assignee: Everex TI Corporation
• Original Assignee: Everex TI Corporation
• Inventors: Boatner, Bruce E., Jacobs, Michael W., McCarthy, Timothy J., Gipe, Michael A., Moluf, Allan A., Bennett, Sanford M., Picard, Raymond L., Foden, Edwin S., Menard, Edward J., Faulkerson, James L.
• Primary Examiner(s): Boudreau, Leo H.
• Assistant Examiner(s): Couso, Jose L.
• Application Number:
  US07/048,100
• Time in Patent Office: 1,194 Days
• Field of Search: 382/44, 382/6, 382/59, 382/34, 382/56, 382/41, 382/49, 358/135, 358/136, 358/228
• US Class Current: 382/313
• CPC Class Codes: G06F 3/0238 Programmable keyboards key guide holders G06F3/0224
• CPC Class Codes: G06V 30/142 using hand-held instruments; Constructional details of the instruments
• CPC Class Codes: H04N 1/00204 with a digital computer or a digital computer system, e.g. an internet server programmed control between transmitter and receiver or between image input and image output device H04N1/32561
• CPC Class Codes: H04N 1/047 Detection, control or error compensation of scanning velocity or position H04N1/0402 and H04N1/17 take precedence
• CPC Class Codes: H04N 1/107 with manual scanning
• CPC Class Codes: H04N 1/1077 Arrangements for facilitating movement over the scanned medium, e.g. disposition of rollers
• CPC Class Codes: H04N 1/21 Intermediate information storage H04N1/387, H04N1/41 take precedence ; for control between transmitter and receiver or between image input and image output device H04N1/32358; indexing, editing G11B27/00
• CPC Class Codes: H04N 1/40056 Circuits for driving or energising particular reading heads or original illumination means H04N1/401, H04N1/407 take precedence
• CPC Class Codes: H04N 2201/0414 Scanning an image in a series of overlapping zones
• CPC Class Codes: H04N 2201/04703 using the scanning elements as detectors, e.g. by performing a prescan
• CPC Class Codes: H04N 2201/0471 using dedicated detectors
• CPC Class Codes: H04N 2201/04724 on a separate encoder wheel
• CPC Class Codes: H04N 2201/04734 Detecting at frequent intervals, e.g. once per line for sub-scan control
• CPC Class Codes: H04N 2201/04743 by detecting the image directly
• CPC Class Codes: H04N 2201/04787 by changing or controlling the addresses or values of pixels, e.g. in an array, in a memory, by interpolation

Abstract

An adaptive image acquisition system employing a hand-held optical scanner or camera is disclosed. The system includes an adaptive camera exposure control system to adjust the camera exposure time to the media'"'"'s reflectivity characteristics. The system assembles a "filmstrip" of images from compressed video data, wherein redundant image data appearing in successive image frames has been removed. The system employs a correlation process between successive frames to identify the redundant frame information and determined the camera velocity. By adapting the correlation process to the velocity of the camera, the effects of variations in the velocity may be substantially eliminated.

Claims

1. An adaptive image acquisition system for optically scanning a media surface bearing static information in a particular direction and providing in real time a digital data representation of the scanned portion of the surface, comprising:

a hand-held camera comprising an image sensor array of photosensitive pixel elements arranged in r rows and C columns;

means for exposing said pixel elements for capturing sequential images of the portion of said media surface within the field of view of said array;

means for reading out from said imaging array resulting digital data values for said pixel elements to provide sequential frames of digital image data comprising C columns and R rows of data;

means for real time processing of successive ones of said frames of image data to remove data therefrom which is redundant to image data in the preceding image frame, said means comprising;

means for storing two successive frames of data, the initial frame known as the reference frame and the next frame known as the current frame;

means for selecting a reference window defined by a subset of the pixel element values comprising said reference frame, said selecting means comprising means for adaptively determining said window size in dependence on the rate of relative movement between said imaging array and said media surface to decrease the size of said reference window as said rate increases; and

means for accessing said storing means and comparing selected pixel element values comprising the current frame with the pixel element values comprising said reference window and determining which pixel element data of said current frame is redundant to the pixel element data comprising said reference window; and

means for providing an output image frame data set for each acquired image frame which includes data representative of only the pixel element data determined to be non-redundant to the pixel element data of the next preceding image frame.

2. The adaptive image acquisition system of claim 1 wherein said comparing means comprises means for correlating current frame pixel values with those reference frame values within said reference window to determine the relative position of said reference window in relation to said current frame having the best match between corresponding pixel values of said reference window and current frame pixels.

3. The adaptive image acquisition system of claim 2 wherein said correlating means comprises a means for performing a one dimensional correlation between the pixel elements comprising selected compounds of said current frame and the corresponding pixel elements comprising columns comprising said reference window.

4. The adaptive image acquisition system of claim 3 wherein said hand-held camera further comprises at least one tracking wheel for contacting said surface and mounted for rotation substantially restricted to along said particular direction for inhibiting movement of said camera orthogonal to said direction.

5. The adaptive image acquisition system of claim 3 wherein said correlating means comprises:

means for determining the number of corresponding pixel mismatches between said reference window and corresponding columns of said current frame for each of a number P of incrementally different alignment positions of said reference window and said current frame;

means for searching the respective number of pixel mismatches for respective alignment positions to determine the alignment position having the fewest mismatches; and

means for providing a value representing the offset of said window portion from said current image frame at said position having the fewest mismatches.

6. The adaptive image acquisition system of claim 1 wherein said means for reading out from said imaging array comprises a first reading means for reading only a first set of said pixel elements to provide image frames of a first relative resolution level, second means for reading said first set of pixel elements and a second set of pixel elements to provide image frames which are of higher resolution than said first resolution level, and means for selecting either said first reading means or said second reading means to selectively provide either image frames of said first resolution level or said second resolution level.

7. The adaptive image acquisition system of claim 6 wherein said first set of pixel elements comprises substantially one-half of the pixel elements comprising said imaging array, the pixel elements comprising said first set distributed over the area of said imaging array.

8. The adaptive image acquisition system of claim 7 wherein said second set of pixel elements comprises substantially all of the pixel elements of the imaging array which do not comprises said first set.

9. The adaptive image acquisition system of claim 6 wherein said reading means further comprises third reading means for providing image frames comprising substantially the same number of pixel values as said images provided by said first reading means, wherein each pixel value of said third reading means is the logical result of performing a logical OR function on a respective pixel element value of said first set and corresponding pixel element value of a pixel of said second set which is adjacent said pixel element of said first set, said third means for providing a data-compressed high resolution image.

10. An adaptive image acquisition system for optically scanning a media surface bearing static information in a particular direction and providing in real time a digital data representation of the scanned portion of the surface, comprising:

a hand-held camera comprising an image sensor array of photosensitive pixel elements arranged in R rows and C columns;

means for exposing said pixel elements during sequential exposure intervals of selectable duration for capturing sequential images of the portion of said media surface within the field of view of said array;

means for reading out from said imaging array after each exposure interval thereof resulting digital data values for said pixel elements to provide sequential frames of digital image data comprising C columns and R rows of data;

exposure control means for adaptively selecting the duration of said exposure intervals in dependence on the reflectivity of a background portion of said media surface, said exposure control means comprising means for acquiring a plurality of exposure measurement images of a background portion of said media surface with different exposure interval durations, and means for determining the number of black pixels in said respective images and processing said black pixel numbers to determine an exposure interval duration which yields a minimal number of black pixels for said background portion of said surface;

means for real time processing of successive ones of said frames of image data to remove data therefrom which is redundant to image data in the preceding image frame, said means comprising;

means for storing two successive frames of data, the initial frame known as the reference frame and the next frame known as the current frame;

means for selecting a reference window defined by a subset of the pixel element values comprising said reference frame; and

means for accessing said storing means and comparing selected pixel element values comprising the current frame with the pixel element values comprising said reference window and determining which pixel element data of said current frame is redundant to the pixel element data comprising said reference window; and

means for providing an output image frame data set for each acquired image frame which includes data representative of only the pixel element data determined to be non-redundant to the pixel element data of the next preceding image frame.

11. The adaptive image acquisition system of claim 10 wherein said exposure control means further comprises:

means for fixing a first target black pixel count N_w equal to a predetermined small number of the total number of pixels in said array and a second target black pixel count N_b equal to a predetermined large number of said total number of pixels;

means for determining the exposure interval duration T_w required to provide a background image black pixel count substantially equal to said first target count N_w ;

means for determining the exposure interval duration T_b required to provide a background image black pixel count substantially equal to said second target count N_b ; and

means for calculating an optimum exposure duration T_opt in dependence on said values T_b and T_w.

12. The adaptive image acquisition system of claim 11 wherein said exposure control means further comprises an N-bit binary counter and a clock, and wherein the start count of said binary counter and the clock rate determine said exposure interval duration, and wherein said means for determining said interval duration comprises:

means for acquiring N image frames by separately testing, from the most significant bit to the least significant bit, each bit comprising the start count at the value "1," comparing the resulting image black pixel counter against the target value and setting the bit under test to "0" if the resulting count is less than the target value.

13. An adaptive image acquisition system for optically scanning a media surface bearing static information in a particular direction and providing in real time a digital data representation of the scanned portion of the surface, comprising:

an image sensor array of photosensitive pixel elements arranged in R rows and C columns;

means for exposing said pixel elements for capturing sequential images of the portion of said media surface within the field of view of said array;

means for reading out from said imaging array resulting digital data values for said pixel elements to provide sequential frames of digital image data comprising C columns and R rows of data;

means for real time processing of successive ones of said frames of image data to remove data therefrom which is redundant to image data in the preceding image frame, said means comprising;

means for storing two successive frames of data, the initial frame known as the reference frame and the next frame known as the current frame;

means for selecting a reference window of N image columns, where the number N is less than the number C of pixel columns;

means for adaptively determining the number N in dependence on the rate of relative movement between the image sensor array and said media surface to decrease the size of the reference window as the rate decreases; and

means for accessing said storing means for comparing the pixel elements comprising selected columns of the current frame with the pixel elements comprising N selected columns of the reference frame within said reference window and determining which columns of pixel data of said current frame are redundant to corresponding columns of data within said reference window; and

means for providing an output image frame data set for each acquired image which includes data representative of only those columns of pixel elements determined to be non-redundant to the columns of the next preceding image frame.

14. The adaptive image acquisition system of claim 13 wherein said exposing means is for exposing said pixel elements during sequential exposure intervals of selectable duration, and further comprising exposure selection means for adaptively selecting said exposure interval duration in accordance with the reflectivity of the particular medium surface and the sensitivity of said image sensor array.

15. The adaptive image acquisition system of claim 13 wherein said comparing means comprises means for correlating current frame pixel values with those reference frame values within said N column window to determine the relative column-wise position of said reference frame window in relation to said current frame having the best match between corresponding pixel values of said reference and current frame pixels.

16. The adaptive image acquisition system of claim 15 wherein said correlating means comprises a means for performing a one dimensional correlation between the pixel elements comprising selected columns of said current frame and the corresponding pixel elements comprising said columns comprising said reference window.

17. The adaptive image acquisition system of claim 16 wherein said image sensor array is contained within a hand-held camera body, and further comprising at least one tracking wheel mounted on said body for contacting said surface and mounted for rotation substantially restricted along said particular direction for inhibiting movement of said camera body orthogonal to said direction.

18. The adaptive image acquisition system of claim 15 wherein said correlating means comprises:

means for determining the number of corresponding pixel mismatches between said reference window and corresponding columns of said current frame for each of a number P of incrementally different column-wise alignment positions of said reference window and said current frame;

means for determining the alignment position having the fewest mismatches; and

means for providing a value representing the column-wise offset of said window portion from said current image frame at said position having the fewest mismatches.

19. The adaptive image acquisition system of claim 13 wherein said means for reading out from said imaging array comprises a first reading means for reading only a first set of said pixel elements to provide image frames of first relative resolution level, second means for reading said first set of pixel elements and a second set of pixel elements to provide image frames which are of higher resolution than said first resolution level, and means for selecting either said first reading means or said second reading means to selectively provide either image frames of said first resolution level or said second resolution level.

20. The adaptive image acquisition system of claim 19 wherein said first set of pixel elements comprises substantially one-half of the pixel elements comprising said imaging array, the pixel elements comprising said first set distributed over the area of said imaging array.

21. The adaptive image acquisition system of claim 20 wherein said second set of pixel elements comprises substantially all of the pixel elements of the imaging array which do not comprises said first set.

22. The adaptive image acquisition system of claim 19 wherein said reading means further comprises third reading means for providing image frames comprising substantially the same number of pixel values as said images provided by said first reading means, wherein each pixel value of said third reading means is the logical result of performing a logic al OR function on a respective pixel element value of said first set and corresponding pixel element value of a pixel of said second set which is adjacent said pixel element of said first set, said third means for providing a data-compressed high resolution image.

23. The adaptive image acquisition system of claim 13 wherein said means for exposing said pixel elements comprises a scanner light source for illuminating said surface with illumination light during sequential illumination intervals, and wherein said pixel elements are responsive to illumination light reflected from said surface.

24. An adaptive image acquisition system for processing sequential frames of digital image data to remove redundant data, comprising:

an optical camera means for providing sequential frames of digital image data, each frame representing a captured image of a portion of a surface bearing static information, each said frame comprising rows and columns of image pixel data;

means responsive to a current image frame and the next preceding image frame for;

(i) selecting a reference frame window size of N columns in relation to the current image frame, where the number N is adaptively selected in dependence on the relative rate of movement between said optical camera means and said surface;

(ii) comparing selected columns of the current frame image data with the corresponding columns of said preceding frame image data within said reference frame window at a plurality P of alignment positions of the reference frame window, where the number P is selected in dependence on said relative rate of movement, and determining which columns of data of said current frame are redundant with corresponding columns of data of said preceding frame; and

(iii) discarding said redundant columns and providing as a system output only those remaining columns of said current image frame.

25. A data compression method for processing sequential frames of digitized image data produced by a scanner moving in a predetermined direction in relation to a desired surface bearing static information, comprising a sequence of the following steps:

(a) providing sequential image frames of digitized image pixel data of the portion of said surface within the field of view of said scanner, each frame comprising R rows and C columns of pixel data;

(b) as the image frames are provided, storing the current frame and the next preceding reference frame in an addressable digital memory, the addresses for each frame addressable in the Y or row-wise dimension and in the X or column-wise dimension, where said predetermined movement direction is substantially aligned with said X dimension;

(c) defining a reference frame window formed by N contiguous columns of said reference frame, where the value N is less than the value C, wherein the value N is adaptively determined in dependence on the rate of relative movement between the scanner and the surface to decrease the size of the reference window as the rate increases;

(d) for each of a plurality of contiguous column-wise alignment positions of said reference frame window in relation to contiguous columns of said current frame, comparing the corresponding respective pixel values in said respective current frame and reference frame window to determine the column-wise alignment position of said reference window in relation to said current frame providing the best match between corresponding pixels;

(e) from the best match position, determining which columns of pixel values comprising the current frame are redundant to columns of pixel data of the reference window and saving only the non-redundant columns of the current frame.

26. The method of claim 25 wherein said step (e) comprises:

(i) determining the number (DELTAX) of pixel columns said reference window is offset from said current frame;

(ii) saving the DELTAX value;

(iii) repeating steps (d), e(i) and e(ii) for the next image frame provided, considering the new frame as the current frame and the frame previously considered the current frame as the reference frame, and discarding all columns of said reference frame except the DELTAX columns of said reference frame adjacent the leading edge of said reference frame.

27. The method of claim 25 wherein said step (d) comprises performing multiple-level, row-wise, column-wise and position-wise comparisons between respective current and reference window data operands, by:

(i) fixing a frame row start value;

(ii) fixing a frame column start value equal to the column address of the pixel column in the current data frame at the leading edge of said frame;

(iii) iteratively comparing the current frame data operand defined by said start row and column addresses against each of the possible reference frame data operands which are aligned with said current data operand for the respective alignment positions to determine whether the pixels match, and accumulating said respective comparison results with the results of previous comparisons for the particular position in a position comparison results buffer;

(iv) incrementally varying the column address and repeating step (iii) until all column addresses have been used;

(v) incrementally varying the row address and repeating steps (iii) and (iv) until all row addresses have been used; and

(vi) searching the accumulated comparison results for each alignment position to find the position providing the best match.

28. The method of claim 27 wherein said respective current frame and said next preceding reference frame are stored in a random access memory.

31. The method of claim 28 further comprising the step of determining whether each image is of only the media surface background, and setting the value DELTAX to an estimated value when the image is of only the media background.

32. The method of claim 31 wherein said background determining step comprises:

(i) setting the exposure interval to a predetermined exposure time;

(ii) determining the number of black pixels in the image frame;

(iii) comparing the number of black pixels to a predetermined threshold value; and

(iv) declaring that the image is of only the media background if said black pixel number is less than said threshold value.

33. The method of claim 31 wherein said background determining step comprises:

(i) determining the amount of blackness in the image frame;

(ii) comparing the determined blackness amount against a predetermined threshold value; and

(iii) declaring that the image frame is of only the media background is said determined blackness amount is less than said threshold value.

34. The method of claim 31 wherein said step of setting the value DELTAX to an estimated value comprises setting the value DELTAX to the particular calculated DELTAX value for the last image frame which was not solely of background.

29. A method for sequentially acquiring digitized images of a surface bearing static information, comprising a sequence of the following steps:

(a) providing an optical scanner having an image sensor array of photosensitive pixel elements arranged in R rows and C columns, the rows extending substantially along the X axis and the columns substantially along the Y axis in a Cartesian coordinate system;

(b) imparting relative motion between the optical scanner and said surface substantially in a predetermined direction along the X axis;

(c) sequentially exposing the image sensor array for an exposure duration and then reading the values of the pixel elements to provide a sequential series of digitized image data frames of the portion of the surface within the field of view of said scanner;

(d) correlating each image data frame with a reference frame data window formed by a plurality N of contiguous columns of pixel data values of the preceding frame, where N is less than C, to determine the degree of overlap between the successive image frames and determine a variable value DELTAX representing the number of non-redundant columns for the particular image frame, wherein the value N is adaptively determined in dependence on the rate of relative movement between the image sensor array and the surface to decrease the size of the reference window as the rate increases; and

(e) after the particular image frame has been correlated against the next successive image frame, discarding the (C-DELTAX) columns of image pixel data determined to have been redundant in step (d) and storing only the remaining DELTAX columns of pixel data for that image frame.

30. The method of claim 29 wherein said step (d) of correlating each image data frame comprises:

(i) aligning said reference window with contiguous columns of the particular image frame at a plurality P of incrementally varied alignment positions;

(ii) for each position, making a bit-wise comparison of the respective values of the corresponding pixel element values for the reference window and image frame;

(iii) determining which alignment position results in the best match between the reference window and image frame; and

(iv) determining the value DELTAX as the displacement in columns of said best match alignment position from a predetermined edge of the image frame.

35. The method of claim 30 wherein said step of making a bit-wise determination of said respective values comprises determining the number of pixel value mismatches between corresponding pixel element values for the reference window and image frame, and said step of determining which alignment position results in the best match comprises selecting the alignment position having the fewest number of mismatches.

36. The method of claim 35 further comprising the steps of:

determining the average number of pixel mismatches per alignment position;

comparing the number of pixel mismatches of the best match alignment position with said average mismatch number; and

if the number of pixels for the best match position does not exceed said average number by at least a predetermined threshold value, maintaining the value DELTAX as the value used for the next preceding image frame.

37. The method of claim 29 further comprising the steps of determining whether said DELTAX value equals or exceeds a predetermined maximum value, and if so, indicating that an error condition exists and terminating said correlation.

38. An adaptive image acquisition system for optically scanning across a media surface bearing static information in a predetermined scan direction and providing a digital data representation of the scanned portion of the surface, comprising:

a hand-held optical scanner comprising an image sensor array of photosensitive pixel elements arranged in R rows and C columns, and means for exposing said pixel elements to the media surface during sequential exposure intervals for capturing sequential images of the portion of said medium surface within the field of view of said array, said rows of elements extending generally in the direction of said scan direction and said columns extending generally orthogonally to said scan direction;

an image frame buffer memory coupled to said imaging array for buffering digitized representations of a plurality of said images;

a video processor coupled to said optical scanner and said buffer memory for providing control signals to said imaging array and said exposing means for controlling the start and duration of said exposure intervals, for reading out from said imaging array after each exposure interval resulting digital data values for said pixel elements to sequentially provide said digitized representations of said images, and for storing said digitized representations of a plurality of sequential images at selected memory locations within said buffer memory; and

a central controller coupled to said video processor to control said video processor operations and receive from said video processor compressed digital data representing the non-redundant portions of the sequential images captured by the scanner;

said video processor and said central controller being cooperatively arranged to provide a means for performing a one-dimensional correlation along said scan direction between each image frame and a selectable window portion of the next preceding image frame comprising N columns thereof to identify redundant columns of pixel data appearing in both said image frame and said selectable window portion, and to provide said compressed digital data to said controller representing only non-redundant image data.

39. The system of claim 38 wherein said means for exposing said pixel elements comprises a scanner light source for illuminating said surface with illumination light during said illumination intervals.

40. The system of claim 38 wherein said means for performing a one-dimensional correlation comprises:

means for determining said number of corresponding pixel mismatches between said window portion and the image frame for each of a number P of incrementally different column-wise alignment positions of the window portion and the image frame;

means for determining the position having the fewest mismatches; and

means for providing an offset value representing the column-wise offset of said window portion from said image frame at said position having the fewest mismatches.

41. The system of claim 38 wherein said frame buffer memory comprises memory locations for storing three successive image frames, the most recently acquired or "new" frame, the next preceding or "current" frame and the second preceding or "reference" frame, and wherein said video processor comprises:

means responsive to the termination of the exposure interval for sequentially accessing said frame buffer memory and fetching a selected data operand comprising said current frame and a selected data operand comprising said portion of said preceding frame;

means for buffering a variable number of said selected window operands at selected locations in a reference window buffer, the number thereof dependent on the number of different reference window operands to which said image frame operand will be compared at each of the alignment positions;

means for generating buffer addresses for selectively accessing the window operands stored in said buffering means;

means for sequentially comparing said image frame operand to each of said varying number of buffered reference window operands to determine whether said operands match, and to provide a comparison result value indicating a mismatch for each of said window operands;

read/write comparison results buffer memory having at least P memory locations, one for each alignment position;

means for generating read addresses and write addresses for said results buffer memory, said means synchronized with the operation of said comparing means for addressing the buffer memory location corresponding to the same position as said comparison result value;

means coupled to said comparing means and said results buffer memory for accumulating said respective comparison result value with the value stored in the corresponding position memory location of said results buffer memory; and

means responsive to a signal indicating that the comparison between said window portion and said image frame is complete for subsequentially transferring said accumulated comparison result values from said results buffer memory to said central controller.

42. The system of claim 40 wherein central controller adaptively determines the number of columns N comprising said reference window in dependence on said offset value, wherein said value N varies between predetermined minimum and maximum values and decreases with increasing offset values.

43. The system of claim 40 wherein said central controller adaptively determines the number of alignment positions P in dependence on said offset value, wherein said value P varies between predetermined minimum and maximum values.

44. The system of claim 38 wherein said optical scanner further comprises at least one tracking wheel for contacting said surface and mounted for rotation substantially restricted along said scan direction for inhibiting movement of said optical scanner orthogonal to said scan direction.

45. The system of claim 38 wherein said central controller comprises an image buffer memory for assembling and storing said compressed digital data as a composite image representation of the portion of said surface scanned by said scanner.

46. The system of claim 45 wherein said central controller is remote from said video processor, and further comprising a serial data interface between said video processor and said central controller for transferring said compressed digital data, wherein said serial data interface comprises:

a first serial data path for carrying sequential packets of serial data comprising compressed digital data; and

a second serial data path for carrying sequential packets of control data, each respective packet of control data identifying a corresponding packet of serial data carried over said first data path.

47. Apparatus for sequentially acquiring digitized images of a surface bearing static information, comprising:

a handheld optical scanner having an image sensor array of photosensitive pixel elements arranged in R rows and C columns, the rows extending substantially along the X axis and the columns substantially along the Y axis in a Cartesian coordinate system;

means for sequentially exposing the image sensor array for an exposure duration and then reading the values of the pixel elements to provide a sequential series of digitized image data frames;

means for determining the effective spatial position of one data frame to the next frame in relation to said surface;

means responsive to said spatial position means for determining the data in the each frame which is not redundant to the data in the preceding frame, said means comprising means for adaptively selecting a reference window of N rows of the data in the preceding frame, the size N being adaptively selected in dependence on the rate of relative movement between the sensor array and the surface so as to reduce the size N as the rate increases, and means for comparing the reference frame data and next framed data for determining the non-redundant data; and

means for buffering said non-redundant data with the non-redundant data from previous frames, to assemble a composite image of said surface.

48. The apparatus of claim 47 wherein said hand-held camera comprises a tracking wheel mounted on an axle for contacting said surface and rotating as said scanner moves in relation to said surface, and said means for determining the effective spatial position of said data frames comprises a means for measuring said movement, said movement comprising an encoder device for providing electrical signals indicative of rotation of said wheel by predetermined amounts.

49. An adaptive image acquisition system for optically scanning a media surface in a particular direction and providing in real time a compressed digital data representation of the scanned portion of the surface, comprising:

a hand-held camera comprising an image sensor array of photosensitive pixel elements arranged in R rows and C columns;

means for exposing said pixel elements during sequential exposure intervals of selectable duration for capturing sequential images of the portion of said medium surface within the field of view of said array;

means for reading out from said imaging array after each exposure interval thereof resulting digital data values for said pixel elements to provide sequential frames of digital image data comprising C columns and R rows of data;

means for real time processing of successive ones of said frames of image data to remove data therefrom which is redundant to image data in the preceding image frame, said means comprising;

means for storing two successive frames of data, the initial frame known as the reference frame and the next frame known as the current frame;

means for selecting a reference window comprising a subset of the pixel element values comprising said reference frame;

means for adaptively determining the size of said reference window in dependence on the rate of relative movement between said imaging array and said media surface to decrease the size of said reference window as said rate increases; and

means for accessing said storing means and comparing selected pixel element values comprising the current frame with the pixel element values comprising said reference window and determining which pixel element data of said current frame is redundant to the pixel element data comprising said reference window; and

means for providing an output image frame data set for each acquired image frame which includes data representative of only the pixel element data determined to be non-redundant to the pixel element data of the next preceding image frame.

50. An adaptive image acquisition system for optically scanning a media surface in a particular direction and providing in real time a compressed digital data representation of the scanned portion of the surface, comprising:

an image sensor array of photosensitive pixel elements arranged in R rows and C columns;

means for exposing said pixel elements during sequential exposure intervals of selectable duration for capturing sequential images of the portion of said medium surface within the field of view of said array;

means for reading out from said imaging array after each exposure interval thereof resulting digital data values for said pixel elements to provide sequential frames of digital image data comprising C columns and R rows of data;

means for real time processing of successive ones of said frames of image data to remove data therefrom which is redundant to image data in the preceding image frame, said means comprising;

means for storing two successive frames of data, the initial frame known as the reference frame and the next frame known as the current frame;

means for selecting a reference window of N image columns, where the number N is less than the number C of pixel columns;

means for adaptively determining the size of said reference window in dependence on the rate of relative movement between said imaging array and said media surface to decrease the size of said reference window as said rate increases; and

means for accessing said storing means for comparing the pixel elements comprising selected columns of the current frame with the pixel elements comprising N selected columns of the reference frame within said reference window and determining which columns of pixel data of said current frame are redundant to corresponding columns of data within said reference window; and

means for providing an output image frame data set for each acquired image which includes data representative of only those columns of pixel elements determined to be non-redundant to the columns of the next preceding image frame.

51. The adaptive image acquisition system of claim 50 wherein said processing means further comprises means for providing a data signal indicative of the number of non-redundant frame columns determined by said comparing means, and wherein said window size determining means is responsive to said data signal.

52. A method for sequentially acquiring digitized images of an information bearing surface, comprising a sequence of the following steps:

(a) providing an optical scanner having an image sensor array of photosensitive pixel elements arranged in R rows and C columns, the rows extending substantially along the X axis and the columns substantially along the Y axis in a Cartesian coordinate system;

(b) imparting relative motion between the optical scanner and the information bearing surface substantially in a predetermined direction along the X axis;

(c) sequentially exposing the image sensor array for an exposure duration and then reading the values of the pixel elements to provide a sequential series of digitized image data frames;

(d) correlating each image data frame with a reference frame data window comprising a plurality N of contiguous columns of pixel data values of the preceding frame, where N is less than C, to determine the degree of overlap between the successive image frames and determine a variable value DELTAX representing the number of non-redundant columns for the particular image frame; and

(e) adaptively selecting the size N of the reference window used in correlation the next successive image frame in dependence on the magnitude of the variable value DELTAX, such that the value N has a minimum value and increases with decreasing values of DELTAX to a maximum window size which is less than the number of pixel element columns in the image frame, and

(f) after the particular image frame has been correlated against the next successive image frame, discarding the (C-DELTAX) columns of image pixel data determined to have been redundant in step (d) and storing only the remaining DELTAX columns of pixel data for that image frame.

53. A method for sequentially acquiring digitized images of an information bearing surface, comprising a sequence of the following steps:

(a) providing an optical scanner having an image sensor array of photosensitive pixel elements arranged in R rows and C columns, the rows extending substantially along the X axis and the columns substantially along the Y axis in a Cartesian coordinate system;

(b) imparting relative motion between the optical scanner and the information bearing surface substantially in a predetermined direction along the X axis;

(c) sequentially exposing the image sensor array for an exposure duration and then reading the values of the pixel elements to provide a sequential series of digitized image data frames;

(d) adaptively selecting a reference frame data window comprising a plurality N of contiguous columns of pixel data values of the preceding frame, where N is less than C, in dependence on the magnitude of said relative motion, such that the size N of the window increases as the magnitude of the motion decreases;

(e) correlating each image data frame with said reference frame data window to determine the degree of overlap between the successive image frames and determine a variable value DELTAX representing the number of non-redundant columns for the particular image frame; and

(f) after the particular image frame has been correlated against the next successive image frame, discarding the (C-DELTAX) columns of image pixel data determined to have been redundant in step (d) and storing only the remaining DELTAX columns of pixel data for that image frame.