Background intensity correction of a scan of a document

1. A method for background intensity correction of a scan of pixel values of a document, the scan having at least one background pixel area and at least one non-background pixel area, comprising:

segmenting the scan into the at least one background pixel area and the at least one non-background pixel area;

calculating a gray background pixel value of the at least one background pixel area, wherein the calculating depends upon at least one peak in a histogram of the pixel values of the at least one background pixel area; and

compensating only the pixel values of the at least one background pixel area for background intensity using the calculated gray background pixel value, wherein pixels in the at least one background pixel area are compensated by stretching the pixel values between zero and the at least one peak in the histogram,

wherein at least one of the segmenting, calculating and compensating is performed by a scanning apparatus.

2. The method of claim 1 , wherein the calculating of the gray background pixel value comprises calculating the gray background pixel value of at least some of the pixels of a strip of pixels of a plurality of scan lines proximate a leading edge of the document.

3. The method of claim 1 , wherein the calculating of the gray background pixel value comprises calculating the gray background pixel value of all of the pixels of a strip.

4. The method of claim 1 , wherein the calculating of the gray background pixel value comprises calculating the gray background pixel value of only background pixels of a strip.

5. The method of claim 1 , further comprising down sampling the scan, wherein the segmenting of the scan includes segmenting the down-sampled scan into the at least one background pixel area and the at least one non-background pixel area.

6. The method of claim 5 , wherein the down sampling of the scan uses a down sampling ratio of four to one.

7. The method of claim 1 , wherein the segmenting of the scan comprises:

separating the pixel values into background pixels and non-background pixels;

extracting each area of connected non-background pixels; and

labeling each un-extracted area of connected background pixels as belonging to the at least one background pixel area.

8. The method of claim 1 , wherein the segmenting of the scan comprises:

separating the pixel values into background pixels and non-background pixels;

extracting each area of connected background pixels; and

labeling each extracted area of connected background pixels as belonging to the at least one background pixel area.

9. The method of claim 8 , wherein the separating of the pixel values into background pixels and non-background pixels comprises creating a binary image of the scan using an adaptive thresholding process.

10. A method for background intensity correction of a scan of pixel values of a document, the scan having at least one background pixel area and at least one non-background pixel area, the at least one non-background pixel area including at least one text pixel area and at least one image pixel area, comprising:

segmenting the scan into the at least one background pixel area, the at least one text pixel area, and the at least one image pixel area;

calculating a gray background pixel value of the at least one background pixel area, wherein the calculating depends upon at least one peak in a histogram of the pixel values of the at least one background pixel area; and

compensating only the pixel values of the at least one background pixel area for background intensity using the calculated gray background pixel value, wherein pixels in the at least one background pixel area are compensated by stretching the pixel values between zero and the at least one peak in the histogram,

wherein at least one of the segmenting, calculating and compensating is performed by a scanning apparatus.

11. The method of claim 10 , wherein the calculating of the gray background pixel value comprises calculating the gray background pixel value of at least some of the pixels of a strip of pixels of a plurality of scan lines proximate a leading edge of the document.

12. The method of claim 10 , wherein the calculating of the gray background pixel value comprises calculating the gray background pixel value of all of the pixels of a strip.

13. The method of claim 11 , wherein the calculating of the gray background pixel value comprises calculating the gray background pixel value of only background pixels of a strip.

14. The method of claim 10 , further comprising down sampling the scan, wherein the segmenting of the scan includes segmenting the down-sampled scan into the at least one background pixel area and the at least one non-background pixel area.

15. The method of claim 14 , wherein the down sampling uses a down sampling ratio of four to one.

16. The method of claim 10 , wherein the segmenting of the scan comprises:

separating the pixel values into background pixels and non-background pixels;

extracting each area of connected non-background pixels;

labeling each unextracted area of connected background pixels as belonging to the at least one background pixel area; and

labeling each extracted area of connected non-background pixels as belonging to one of a text pixel area and an image pixel area.

17. The method of claim 10 , wherein the segmenting of the scan comprises:

separating the pixel values into background pixels and non-background pixels;

extracting each area of connected background pixels; labeling each extracted area of connected background pixels as belonging to the at least one background pixel area; and

labeling each unextracted area of connected non-background pixels as belonging to one of a text pixel area and an image pixel area.

18. The method of claim 17 , wherein the separating of the pixel values into background pixels and non-background pixels comprises creating a binary image of the scan using an adaptive thresholding process.

19. A method for background intensity correction of a scan of pixel values of a document, the scan having at least one background pixel area and at least one non-background pixel area, comprising:

segmenting the scan into the at least one background pixel area and the at least one non-background pixel area;

calculating a gray background pixel value of the at least one background pixel area, wherein the calculating depends upon at least one peak in a histogram of the pixel values of the at least one background pixel area; and

compensating only the pixel values of the at least one background pixel area for background intensity using the calculated gray background pixel value, wherein pixels in the at least one background pixel area are compensated by stretching the pixel values between zero and the at least one peak in the histogram, wherein the segmenting of the scan includes separating the pixel values into background pixels and non-background pixels and includes extracting pixels, wherein the separating of the pixel values into background pixels and non-background pixels includes creating a binary image of the scan using an adaptive thresholding process, and wherein the extracting of the pixels includes using binary morphology to extract each area of connected pixels which belong to one of a connected background pixel area and a connected non-background pixel area,

wherein at least one of the segmenting, calculating and compensating is performed by a scanning apparatus.

20. The method of claim 19 , further comprising labeling each connected non-background pixel area as belonging to one of a text pixel area and an image pixel area using a size and an aspect ratio of each connected non-background pixel area.