Division of images into separate color layers

1. A method comprising:

receiving, by one or more processors, a document image;

detecting a plurality of connected components in the document image;

dividing the document image into a plurality of patches, wherein a first patch of the plurality of patches encompasses a first connected component of the plurality of connected components, wherein the first patch is contained within a second patch of the plurality of patches, and wherein the second patch encompasses a second connected component of the plurality of connected components;

determining, for each patch, whether the patch is monochromatic or polychromatic;

clustering monochromatic patches of the plurality of patches into a plurality of clusters within a color space, wherein the plurality of clusters comprises a predefined number of clusters, and wherein each cluster corresponds to a color layer of a plurality of color layers of the document image;

segmenting each polychromatic patch of the plurality of patches into a corresponding plurality of monochromatic segments;

for each polychromatic patch, associating each monochromatic segment of the corresponding plurality of monochromatic segments with a cluster of the plurality of clusters; and

utilizing the plurality of clusters for performing an information extraction task on the document image.

2. The method of claim 1 , further comprising adjusting one or more properties of the document image and binarizing the document image.

3. The method of claim 1 , further comprising collecting statistical data for each patch of the plurality of patches, wherein the statistical data for each patch comprises at least one of a number of pixels, a centroid in the color space, a measure of dispersion, and a measure of skewness.

4. The method of claim 1 , wherein clustering the monochromatic patches into the plurality of clusters comprises performing an iterative clustering process.

5. The method of claim 1 , wherein each monochromatic segment of the plurality of monochromatic segments corresponds to a respective color layer of the plurality of color layers.

6. The method of claim 1 , wherein performing the information extraction task comprises extracting information from each color layer of the document image and combining the information from each color layer in the document image.

7. A system comprising:

a memory; and

a processor coupled to the memory, the processor configured to:

receive a document image;

detect a plurality of connected components in the document image;

divide the document image into a plurality of patches, wherein a first patch of the plurality of patches encompasses a first connected component of the plurality of connected components, wherein the first patch is contained within a second patch of the plurality of patches, and wherein the second patch encompasses a second connected component of the plurality of connected components;

determine, for each patch, whether the patch is monochromatic or polychromatic;

clusterize monochromatic patches of the plurality of patches into a plurality of clusters within a color space, wherein the plurality of clusters comprises a predefined number of clusters, and wherein each cluster corresponds to a color layer of a plurality of color layers of the document image;

segment each polychromatic patch of the plurality of patches into a corresponding plurality of monochromatic segments;

for each polychromatic patch, associate each monochromatic segment of the corresponding plurality of monochromatic segments with a cluster of the plurality of clusters; and

utilize the plurality of clusters for performing an information extraction task on the document image.

8. The system of claim 7 , wherein the processor is further configured to adjusting one or more properties of the document image comprises and binarize the document image.

9. The system of claim 7 , wherein the processor is further configured to collect statistical data for each patch of the plurality of patches, and wherein the statistical data for each patch comprises at least one of a number of pixels, a centroid in the color space, a measure of dispersion, and a measure of skewness.

10. The system of claim 7 , wherein clustering the monochromatic patches into the plurality of clusters comprises performing an iterative clustering process.

11. The system of claim 7 , wherein each monochromatic segment of the plurality of monochromatic segments corresponds to a respective color layer of the plurality of color layers.

12. The system of claim 7 , wherein performing the information extraction task comprises extracting information from each color layer of the document image and combining the information from each color layer in the document image.

13. A non-transitory machine-readable storage medium including instructions that, when accessed by a processing device, cause the processing device to:

receive a document image;

detect a plurality of connected components in the document image;

divide the document image into a plurality of patches, wherein a first patch of the plurality of patches encompasses a first connected component of the plurality of connected components, wherein the first patch is contained within a second patch of the plurality of patches, and wherein the second patch encompasses a second connected component of the plurality of connected components;

determine, for each patch, whether the patch is monochromatic or polychromatic;

clusterize monochromatic patches of the plurality of patches into a plurality of clusters within a color space, wherein the plurality of clusters comprises a predefined number of clusters, and wherein each cluster corresponds to a color layer of a plurality of color layers of the document image;

segment each polychromatic patch of the plurality of patches into a corresponding plurality of monochromatic segments;

for each polychromatic patch, associate each monochromatic segment of the corresponding plurality of monochromatic segments with a cluster of the plurality of clusters; and

utilize the plurality of clusters for performing an information extraction task on the document image.

14. The non-transitory machine-readable storage medium of claim 13 , wherein the instructions further cause the processing device to collect statistical data for each patch of the plurality of patches, and wherein the statistical data for each patch comprises at least one of a number of pixels, a centroid in the color space, a measure of dispersion, and a measure of skewness.

15. The non-transitory machine-readable storage medium of claim 13 , wherein clustering the monochromatic patches into the plurality of clusters comprises performing an iterative clustering process.

16. The non-transitory machine-readable storage medium of claim 13 , wherein each monochromatic segment of the plurality of monochromatic segments corresponds to a respective color layer of the plurality of color layers.

17. The non-transitory machine-readable storage medium of claim 13 , wherein performing the information extraction task comprises extracting information from each color layer of the document image and combining the information from each color layer in the document image.

18. The method of claim 1 , further comprising:

identifying a first cluster of the plurality of clusters and a second cluster of the plurality of clusters, wherein a distance between a first centroid of the first cluster and a second centroid of the second cluster is a smallest distance between respective centroids of any pair of clusters of the plurality of clusters; and

merging the first cluster and the second cluster.

19. The method of claim 1 , further comprising:

performing pairwise merging of clusters of the plurality of clusters until a pre-determined threshold criterion is met.