Method for enhancing a digitized document

1. A method for enhancing a document image, the method implemented at least in part by a data processing system and comprising:

receiving a digitized image of a document containing text or graphics, the digitized image including an array of pixels having pixel values;

decomposing the digitized image into a pedestal image and a residual image, low-spatial-frequency content in the digitized image being primarily represented by the pedestal image and high-spatial-frequency content in the digitized image being primarily represented by the residual image;

determining a first gain factor image wherein the pixel values of the first gain factor image are a continuous monotonic function of the corresponding residual image pixel values, the first gain factor image pixel values being substantially equal to a first maximum gain for non-negative residual image pixel values, and the first gain factor image pixel values being substantially equal to a first minimum gain for residual image pixel values that are less than a first residual threshold;

determining a second gain factor image wherein the pixel values of the second gain factor image are a continuous monotonic function of the corresponding residual image pixel values, the second gain factor image pixel values being substantially equal to a second minimum gain for non-negative residual image pixel values, and the second gain factor image pixel values being substantially equal to a second maximum value for residual image pixel values less than a second residual threshold;

determining a scaled pedestal image by multiplying the pedestal image by the first gain factor image;

determining a scaled residual image by multiplying the residual image by the second gain factor image;

determining an enhanced document image by combining the scaled pedestal image and the scaled residual image; and

storing the enhanced document image in a processor accessible memory.

2. The method of claim 1 further including determining a current white, and wherein the first maximum value is equal to a ratio of a target white point to the current white.

3. The method of claim 2 wherein the current white is determined by analyzing a distribution of pedestal image pixel values.

4. The method of claim 3 wherein the current white is determined by calculating a mean and standard deviation of the pedestal image pixel values, and setting the current white to be equal to the mean of the pedestal image pixel values minus a predefined number of standard deviations of the pedestal image pixel values.

5. The method of claim 1 further including determining a current black and a target black, and wherein the second maximum value is equal to a ratio of the target black to the current black.

6. The method of claim 5 wherein the current black is determined by analyzing a distribution of residual image pixel values.

7. The method of claim 6 wherein the current black is determined by calculating a mean and a standard deviation of the residual image pixel values, and setting the current black to be equal to the mean of the residual image pixel values minus a predefined number of standard deviations of the residual image pixel values.

8. The method of claim 5 wherein the target black is determined by analyzing a distribution of pedestal image pixel values.

9. The method of claim 8 wherein the target black is determined by calculating a mean of the pedestal image pixel values, and setting the target black to be equal to negative one times the mean of the pedestal image pixel values.

10. The method of claim 1 wherein the first minimum gain is substantially equal to one and the second minimum gain is substantially equal to one.

11. The method of claim 1 further including the step of convolving the first gain factor image with a low-pass filter before using it to determine the scaled pedestal image.

12. The method of claim 1 further including the step of convolving the second gain factor image with a low-pass filter before using it to determine the scaled residual image.

13. The method of claim 1 wherein the digitized image is captured using a digital camera or a digital scanner.

14. The method of claim 1 wherein the document has a substantially uniform background.

15. The method of claim 1 wherein the residual image is equal to the digitized image minus the pedestal image.

16. The method of claim 1 wherein a wavelet transform is used to decompose the digitized image into the pedestal image and the residual image.

17. The method of claim 1 further including applying a geometric transform to the digitized image to correct for lens aberrations, perspective warping or rotation introduced when the document was digitized to provide the digitized image.

18. The method of claim 1 further including applying a uniformity correction operation to compensate for non-uniformity introduced when the document was digitized to provide the digitized image.

19. The method of claim 1 wherein the digitized image is a grayscale digitized image that was formed from a color digitized image.

20. The method of claim 1 wherein the digitized image is a luminance channel of a color digitized image.

21. The method of claim 1 further including printing the enhanced document image using a digital printing device.

22. A digital imaging system comprising:

an image sensor for capturing a digitized image;

an optical system for forming an image of a document onto the image sensor;

a data processing system; and

a memory system communicatively connected to the data processing system and storing instructions configured to cause the data processing system to implement a method for enhancing a document image, wherein the method includes:

using the image sensor to capture a digitized image of a document containing text or graphics, the digitized image including an array of pixels having pixel values;

decomposing the digitized image into a pedestal image and a residual image, the low-spatial-frequency content in the digitized image being primarily represented by the pedestal image and the high-spatial-frequency content in the digitized image being primarily represented by the residual image;

determining a first gain factor image wherein the pixel values of the first gain factor image are a continuous monotonic function of the corresponding residual image pixel values, the first gain factor image pixel values being substantially equal to a first maximum gain for non-negative residual image pixel values, and the first gain factor image pixel values being substantially equal to a first minimum gain for residual image pixel values that are less than a first residual threshold;

determining a second gain factor image wherein the pixel values of the second gain factor image are a continuous monotonic function of the corresponding residual image pixel values, the second gain factor image pixel values being substantially equal to a second minimum gain for non-negative residual image pixel values, and the second gain factor image pixel values being substantially equal to a second maximum value for residual image pixel values less than a second residual threshold;

determining a scaled pedestal image by multiplying the pedestal image by the first gain factor image;

determining a scaled residual image by multiplying the residual image by the second gain factor image;

determining an enhanced document image by combining the scaled pedestal image and the scaled residual image;

storing the enhanced document image in a processor accessible memory.

23. The method of claim 22 wherein the digital imaging system is a digital scanner, a digital camera or a camera phone.