Method for auto-cropping a scanned image

1. A method for auto-cropping scanned images, the method comprising:

determining locations of one or more borderline pixels for a current scanned image, the one or more borderline pixels being background pixels that circumscribe an image within an Area of Interest (AOI) image;

approximating average values for R, G, B channels for the one or more borderline pixels based, at least in part, on the scanned images, and setting the approximated R, G, B average values as R, G, B threshold values of the respective borderline pixels;

determining an attribute of each pixel of the current scanned image based on the R, G, B threshold values;

for each pixel of the current scanned image—

determining that the individual pixel is a borderline pixel when the R, G, B levels of the individual pixel are smaller than the R, G, B threshold values; and

recording the position of the individual pixel; and

cropping the AOI image within a crop range defined by each of the borderline pixels of the current scanned image.

2. The method of claim 1 , further comprising:

determining that an individual pixel is an AOI pixel when the standard differences of the R, G, B values of the individual pixel are larger than the R, G, B threshold values.

3. The method of claim 1 wherein determining locations of one or more borderline pixels for a current scanned image comprises:

sequentially comparing the differences of the R, G, B levels for each pixel on the first row of the current scanned image;

determining if the differences of the R, G, B levels of the pixel on the first row are smaller than a predetermined level;

determining the pixel on the first row is a candidate borderline pixel when the standard differences of the R, G, B levels of the candidate borderline pixel are smaller than the predetermined level;

sequentially comparing the standard differences of the R, G, B levels for each pixel on the same column of the candidate borderline pixel; and

determining that the column of the candidate borderline pixel is a borderline when the standard differences of the R, G, B levels of each pixel on the same column of the candidate borderline pixel are smaller than a predetermined level.

4. The method of claim 1 wherein determining locations of one or more borderline pixels for a current scanned image comprises:

sequentially comparing the differences of the R, G, B levels for each pixel on the first column of the current scanned image;

determining if the differences of the R, G, B levels of the pixel on the first column are smaller than a predetermined level;

determining the pixel on the first column as a candidate borderline pixel when the standard differences of the R, G, B levels of the candidate borderline pixel are smaller than the predetermined level;

sequentially comparing the standard differences of the R, G, B levels for each pixel on the same row of the candidate borderline pixel; and

determining that the row of the candidate borderline pixel is a borderline when the standard differences of the R, G, B levels of each pixel on the same row of the candidate borderline pixel are smaller than a predetermined level.

5. The method of claim 3 , further comprising:

determining that an individual pixel is an AOI pixel when the standard differences of the R, G, B values of the individual pixel are larger than the predetermined level.

6. The method of claim 1 wherein approximating average values for R, G, B channels comprises:

storing the R, G, B average values of the borderline pixels of previous scanned images;

averaging the R, G, B average values of the borderline pixels of the previous scanned images and the R, G, B levels of current borderline pixels of the current scanned image, wherein the average value comprises the approximated R, G, B threshold values;

determining—

if the difference between the approximated R, G, B threshold values and the previous R, G, B threshold values is smaller than a predetermined value; and

if the difference between the current R, G, B average values and the previous R, G, B average values is smaller than a predetermined value;

when the determination result is affirmative, setting the sum of the approximated R, G, B threshold values and a weight default value as updated R, G, B threshold values; and

setting the current R, G, B average values as updated R, G, B average values.

7. The method of claim 6 , further comprising:

when the difference of the approximated R, G, B threshold values and the previous R, G, B threshold values is larger than a predetermined value, and the difference of the current R, G, B average values and the previous R, G, B average values is larger than said predetermined value, repeating—

averaging the R, G, B values;

determining differences between the approximated R, G, B threshold values and the current R, G, B values and the previous respective R, G, B values; and

setting the sum of the approximated R, G, B threshold values and the current R, G, B values as the respective updated R, G, B values.

8. A method for auto-cropping scanned images, the method comprising:

determining locations of borderline pixels for a current scanned image, wherein the borderline pixels are background pixels that circumscribe an image within Area of Interest (AOI) image;

storing R, G, B average values of previous borderline pixels of previous scanned images;

averaging R, G, B average values of the borderline pixels of previous scanned images and R, G, B levels of current borderline pixels of the current scanned image, the average value being the approximated R, G, B threshold values;

determining—

if a difference between the approximated R, G, B threshold values and the previous R, G, B threshold values is smaller than a predetermined value; and

if the difference between the current R, G, B average values and the previous R, G, B average values is smaller than a predetermined value;

when the result of the determining is affirmative, setting the sum of the approximated R, G, B threshold values and a weight default value as updated R, G, B threshold values;

when the result of the determining is negative, repeating the storing R, G, B average values, averaging R, G, B average values, and determining steps;

determining an attribute of each pixel of the current scanned image based on the approximated R, G, B threshold values;

determining that a current pixel is a borderline pixel and recording the position of the current pixel when the R, G, B levels of the current pixel are smaller than the R, G, B threshold values; and

cropping an AOI image within a crop range defined by each of the borderline pixels.

9. The method of claim 8 , further comprising:

determining that an individual pixel is an AOI pixel when the differences of the R, G, B values of the individual pixel are larger than the R, G, B threshold values.

10. The method of claim 8 wherein determining locations of borderline pixels for a current scanned image comprises:

sequentially comparing the differences of the R, G, B levels for each pixel on the first row of the current scanned image;

determining if the differences of the R, G, B levels of the pixel on the first row are smaller than a predetermined level;

determining that the pixel on the first row is a candidate borderline pixel when the differences of said R, G, B levels of the candidate borderline pixel are smaller than the predetermined level;

sequentially comparing the differences of the R, G, B levels for each pixel on the same column of the candidate borderline pixel; and

determining that the column of the candidate borderline pixel is a borderline when the differences of the R, G, B levels of each pixel on the same column of the candidate borderline pixel are smaller than a predetermined level.

11. The method of claim 8 wherein determining locations of borderline pixels for a current scanned image comprises:

sequentially comparing the differences of the R, G, B levels for each pixel on the first column of the current scanned image;

determining if the differences of the R, G, B levels of the pixel on the first column are smaller than a predetermined level;

determining that the pixel on the first column is a candidate borderline pixel when the differences of the R, G, B levels of the candidate borderline pixel are smaller than the predetermined level;

sequentially comparing the differences of the R, G, B levels for each pixel on the same row of the candidate borderline pixel; and

determining that the row of the candidate borderline pixel is a borderline when the differences of the R, G, B levels of each pixel on the same row of the candidate borderline pixel are smaller than a predetermined level.