Method and apparatus for removing a mark in a document image

1. A method for removing a mark in a document image, the method comprising:

extracting connected components from a binary image corresponding to the document image;

clustering the connected components based on grayscale features of the connected components to obtain one clustering center;

searching, within numerical ranges of a clustering radius R from the clustering center and a grayscale threshold T, for a combination (R, T) which causes an evaluation value based on the grayscale features of the connected components to be higher than a first evaluation threshold; and

removing the mark in the document image based on the grayscale threshold in the combination;

wherein the grayscale features of the connected components comprise: a minimum one of grayscale values of pixels in the document image, which correspond to all black pixels in each of connected components;

wherein the removing the mark in the document image based on the grayscale threshold in the combination comprises:

removing the connected components, the grayscale features of which are greater than the grayscale threshold, as the mark in the document image.

2. The method according to claim 1 , wherein the connected components are clustered further based on intrinsic features of the connected components to obtain the one clustering center; wherein the intrinsic features of the connected components comprise at least one of: a number of black pixels in each of the connected components, a ratio of the number of the black pixels in the connected component to a number of pixels in a bounding box of the connected component, and an aspect ratio of the bounding box of the connected component, in the binary image.

3. The method according to claim 1 , wherein the clustering the connected components based on grayscale features of the connected components to obtain the one clustering center comprises:

counting a number of connected components within a range of a given clustering radius when each connected component serves as a clustering center, in a feature space of the grayscale features; and

determining a connected component corresponding to a maximum one of counted numbers as the clustering center.

4. The method according to claim 1 , wherein the evaluation value is further based on a number of black pixels in one connected component in the binary image.

5. The method according to claim 1 , wherein the evaluation value reflects a degree of matching of a result of classifying the connected components based on the clustering center and the clustering radius with a result of classifying the connected components based on the grayscale threshold.

6. The method according to claim 1 , wherein, when the evaluation value is lower than a second evaluation threshold, it is judged that there is no mark in the document image.

7. The method according to claim 1 , further comprising: removing the mark in the document image based on the clustering center and the clustering radius in the combination.

8. The method according to claim 1 , wherein the mark comprises one of a watermark and a copymark.

9. An apparatus for removing a mark in a document image, the apparatus comprising:

an extracting device configured to extract connected components from a binary image corresponding to the document image;

a clustering device configured to cluster the connected components based on grayscale features of the connected components to obtain one clustering center;

a searching device configured to search, within numerical ranges of a clustering radius R from the clustering center and a grayscale threshold T, for a combination (R, T) which causes an evaluation value based on the grayscale features of the connected components to be higher than a first evaluation threshold; and

a fine removing device configured to remove the mark in the document image based on the grayscale threshold in the combination;

wherein the grayscale features of the connected components comprise: a minimum one of grayscale values of pixels in the document image, which correspond to all black pixels in each of connected components;

wherein the fine removing device is further configured to remove the connected components, the grayscale features of which are greater than the grayscale threshold, as the mark in the document image.

10. The apparatus according to claim 9 , wherein the clustering device is further configured to cluster the connected components based on the grayscale features and intrinsic features of the connected components to obtain the one clustering center; wherein the intrinsic features of the connected components comprise at least one of: a number of black pixels in each of the connected components, a ratio of the number of the black pixels in the connected component to a number of pixels in a bounding box of the connected component, and an aspect ratio of the bounding box of the connected component, in the binary image.

11. The apparatus according to claim 9 , wherein the clustering device is further configured:

to count a number of connected components within a range of a given clustering radius when each connected component serves as a clustering center, in a feature space of the grayscale features; and

to determine a connected component corresponding to a maximum one of counted numbers as the clustering center.

12. The apparatus according to claim 9 , wherein the evaluation value is further based on a number of black pixels in one connected component in the binary image.

13. The apparatus according to claim 9 , wherein the evaluation value reflects a degree of matching of a result of classifying the connected components based on the clustering center and the clustering radius with a result of classifying the connected components based on the grayscale threshold.

14. The apparatus according to claim 9 , further comprising a judging device configured to judge that there is no mark in the document image when the evaluation value is lower than a second evaluation threshold.

15. The apparatus according to claim 9 , further comprising a rough removing device configured to remove the mark in the document image based on the clustering center and the clustering radius in the combination.

16. The apparatus according to claim 9 , wherein the mark comprises one of a watermark and a copymark.