Image reading apparatus

• US 7,733,548 B2
• Filed: 07/23/2007
• Issued: 06/08/2010
• Est. Priority Date: 07/28/2006
• Status: Active Grant

• Current Assignee: Canon Kabushiki Kaisha (Canon Inc.)
• Original Assignee: Canon Kabushiki Kaisha (Canon Inc.)
• Inventors: Kobayashi, Koji, Makino, Yoichiro, Bannai, Nobuyuki
• Primary Examiner(s): NGUYEN, ANH-VINH THI
• Application Number:
  US11/781,691
• Publication Number:
  US 20080024845A1
• Time in Patent Office: 1,051 Days
• Field of Search: 358/518, 358/538, 358/520, 358/1.2, 358/1.1, 358/1.9, 358/505, 358/474, 358/504, 358/526, 382/274, 355/25, 355/82, 399/362
• US Class Current: 358/520
• CPC Class Codes: H04N 1/387 Composing, repositioning or otherwise geometrically modifying originals

Abstract

A method is provided for processing image data read from a book document placed on a platen of an image reading apparatus. The method includes determining a background pixel of the image data based on a luminance difference or a color difference between neighboring pixel arrays, identifying a shadow region based on a luminance change between the neighboring background pixel arrays, discriminatively detecting a shadow of a binding portion and a shadow of a front edge in a document region with reference to a changing direction of shadow luminance, and performing a luminance correction on the shadow region according to characteristics of the shadows.

Claims

1. A method for processing image data read from a book document placed on a platen, the method comprising:

determining a target pixel of the image data as a background pixel when a difference between the target pixel and a neighboring pixel is less than a threshold, the neighboring pixel being in a first direction perpendicular to a binding portion of the book document (step

     1);

performing a preliminary luminance correction calculation using a same correction value candidate for a luminance value of a background pixel positioned on a line parallel to the binding portion of the book document (step

     2);

integrating an absolute value of a difference between a value obtained by the preliminary luminance correction calculation and a luminance value of a neighboring background pixel positioned in the first direction with respect to background pixels on the line, and selecting a correction value candidate that minimizes the integrated value among a plurality of correction value candidates as a correction value of the background pixel on the line (step

     3);

calculating a cumulative luminance correction value for each line by successively performing a luminance correction calculation in neighboring order of the line based on the selected correction value (step

     4);

extracting a shadow region of at least one of a shadow of a binding portion or a shadow of a front edge portion involved in the image data based on the cumulative luminance correction value (step

     5); and

performing a luminance correction on the shadow region based on the cumulative luminance correction value (step

     6).

2. The method according to claim 1, further comprising:

calculating the cumulative luminance correction value on an assumption that a main scanning direction of the platen is parallel to the binding portion of the book document, and identifying the calculated cumulative luminance correction value as a cumulative luminance correction value in a sub scanning direction (step

     7);

calculating the cumulative luminance correction value on an assumption that the sub scanning direction perpendicular to the main scanning direction of the platen is parallel to the binding portion of the book document, and identifying the calculated cumulative luminance correction value as a cumulative luminance correction value in the main scanning direction (step

     8); and

determining a direction of the binding portion of the book document based on a comparison between a distribution of the cumulative luminance correction value in the sub scanning direction and a distribution of the cumulative luminance correction value in the main scanning direction (step

     9),

wherein the method performs the step 5 and the step 6 based on the cumulative luminance correction value corresponding to the determined direction of the binding portion.

3. The method according to claim 1, wherein the difference between neighboring pixels of the image data is a luminance difference.

4. The method according to claim 1, wherein the difference between neighboring pixels of the image data is a color difference.

5. The method according to claim 1, further comprising:

listing combinations of candidates representing at least one of a shadow of the binding portion and a shadow of the front edge portion involved in the image data based on a distribution of the cumulative luminance correction value; and

selecting a target combination of candidates to be corrected from among the listed combinations of candidates.

6. The method according to claim 1, further comprising:

applying image processing to low-resolution image data obtained by reading the book document at a low resolution;

selecting a combination including at least one of a shadow of the binding portion and a shadow of the front edge to be corrected; and

applying the image processing of the step 6 corresponding to the selected at least one of the shadow of the binding portion and the shadow of the front edge to high-resolution image data obtained by reading the book document at a high resolution.

7. The method according to claim 6, further comprising:

calculating the correction value for the high-resolution image data by interpolating the correction value obtained from the low-resolution image data; and

applying the image processing of the step 6 to the high-resolution image data.

8. The method according to claim 1, wherein the luminance correction in the step 6 includes multiplying a same correction magnification with a luminance correction value of a neighboring pixel as a correction calculation for a pixel on the same line.

9. The method according to claim 1, further comprising:

generating an extended analysis area wider than the shadow region extracted in the step 5;

generating two partial analysis areas in the analysis area which are spaced in a direction of the binding portion;

performing the processing of the steps 1, 2, 3, and 4 in the two partial analysis areas;

generating cumulative luminance correction values in the respective partial analysis areas (step

     10); and

performing an interpolation calculation of a correction value of each pixel involved in the shadow region based on a relationship between the cumulative luminance correction values in the respective partial analysis areas generated in the step 10 (step

     11).

10. The method according to claim 9, wherein the interpolation calculation in the step 11 is performed based on a feature quantity table calculated from the cumulative luminance correction values generated in the step 10.

11. The method according to claim 9, wherein the two partial analysis areas are located away from top and bottom edges of the book document.

12. The method according to claim 9, wherein the step 11 includes performing interpolation processing for correcting a shadow width difference between the two partial analysis areas for a shadow of the binding portion, and performing interpolation processing without correcting the shadow width difference for a shadow of the front edge.

13. The method according to claim 9, wherein the luminance correction is performed based on the shadow region of at least one of the shadow of the binding portion or the shadow of the front edge portion extracted in the step 5, if the shadow region of at least one of the shadow of the binding portion or the shadow of the front edge portion extracted in the step 5 disagrees with a shadow region of at least one of a shadow of the binding portion and a shadow of the front edge portion extracted from the cumulative luminance correction values generated in the step 10, of if the cumulative luminance correction values are not generated in the step 10.

14. The method according to claim 9, further comprising:

comparing features between a first luminance correction table obtained from cumulative luminance correction values calculated based on the entire image data in the step 4 and a second luminance correction table obtained from cumulative luminance correction values calculated based on image data in the partial analysis areas in the step 4; and

performing the luminance correction based on the first luminance correction table or a composite table obtained from the first and second luminance correction tables according to a degree of agreement in the comparison.

15. The method according to claim 1, further comprising:

correcting the cumulative luminance correction value using an approximate curve; and

performing the luminance correction based on the corrected cumulative luminance correction value.

16. A computer-readable medium storing instructions which, when executed by an apparatus, causes the apparatus to perform operations comprising:

obtaining image data generated by scanning and reading a book document placed on a platen of an image scanner;

determining a target pixel of the image data as a background pixel when a difference between the target pixel and a neighboring pixel is less than a threshold, the neighboring pixel being in a first direction perpendicular to a binding portion of the book document;

performing a preliminary luminance correction calculation using a correction value candidate for a luminance value of a background pixel positioned on a line parallel to the binding portion of the book document;

selecting a correction value candidate from a plurality of correction value candidates as a correction value of the background pixel on the line;

calculating a cumulative luminance correction value for each line by successively performing a luminance correction calculation in neighboring order of the line based on the selected correction value;

extracting a shadow region of at least one of a shadow of a binding portion or a shadow of a front edge portion included in the image data based on the cumulative luminance correction value; and

performing a luminance correction on the shadow region based on the cumulative luminance correction value.

17. The computer-readable medium according to claim 16, wherein the selecting the correction value candidate comprises:

integrating an absolute value of a difference between a value obtained by the preliminary luminance correction calculation and a luminance value of a neighboring background pixel positioned in the first direction with respect to background pixels on the line; and

selecting a correction value candidate that minimizes the integrated value among the plurality correction value candidates as a correction value of the background pixel on the line.