Claims:WE CLAIM:
1. A method for detecting tampering in a document image, the method comprising:
determining, by an image processing device, an image quality of the document image by analyzing one or more quality features extracted from the document image;
pre-processing, by the image processing device, the document image based on a pre-defined ontology of documents when the image quality is above a pre-defined quality threshold;
segmenting, by the image processing device, the pre-processed document image into one or more region of interests based on the pre-defined ontology of documents; and
detecting, by the image processing device, tampering in a region of interest of the one or more region of interests in the document image by processing each region of interest to detect tampering in the document image.
2. The method of claim 1, wherein pre-processing the document image comprises:
extracting one or more image characteristics from the document image; and
pre-processing the one or more image characteristics based on the pre-defined ontology of documents, wherein the pre-defined ontology defines a type of pre-processing required on an image characteristic of the one or more image characteristics.
3. The method of claim 2, wherein the one or more image characteristics comprises contrast, sharpness, layout, and orientation.
4. The method of claim 1, wherein the region of interest in the document image is at least one of photo, logo, printed text, handwritten text, and signature.
5. The method of claim 1, wherein the pre-defined ontology of documents defines the one or more region of interests for the document image based on a document type.
6. The method of claim 1, wherein the tampering in the region of interest is assigned a pre-defined weightage.
7. The method of claim 6, wherein the tampering in the document image is detected when a weighted average of the pre-defined weightage for each tampering in the region of interest is above a pre-defined threshold value.
8. The method of claim 1, wherein the pre-defined ontology is dynamically updated in each iteration based on at least one of type of pre-processing on the document image, the one or more region of interests obtained through segmentation on the pre-processed document image, and the tampering detected in the region of interest.
9. An image processing device for detecting tampering in a document image, the image processing device comprising:
a processor; and
a memory communicatively coupled to the processor, wherein the memory stores processor instructions, which, on execution, causes the processor to:
determine an image quality of the document image by analyzing one or more quality features extracted from the document image;
pre-process the document image based on a pre-defined ontology of documents when the image quality is above a pre-defined quality threshold;
segment the pre-processed document image into one or more region of interests based on the pre-defined ontology of documents; and
detect tampering in a region of interest in the document image by processing each region of interest of the one or more region of interests to detect tampering in the document image.
10. The image processing device of claim 9, wherein the processor is further caused to:
extract one or more image characteristics from the document image; and
pre-process the one or more image characteristics based on the pre-defined ontology of documents, wherein the pre-defined ontology defines a type of pre-processing required on an image characteristic of the one or more image characteristics.
11. The image processing device of claim 10, wherein the one or more image characteristics comprises contrast, sharpness, layout, and orientation.
12. The image processing device of claim 9, wherein the region of interest in the document image is at least one of photo, logo, printed text, handwritten text, and signature.
13. The image processing device of claim 9, wherein the pre-defined ontology of documents defines the one or more region of interests for the document image based on a document type.
14. The image processing device of claim 9, wherein the tampering in the region of interest is assigned a pre-defined weightage.
15. The image processing device of claim 14, wherein the tampering in the document image is detected when a weighted average of the pre-defined weightage for each tampering in the region of interest is above a pre-defined threshold value.
16. The image processing device of claim 9, wherein the pre-defined ontology is dynamically updated in each iteration based on at least one of type of pre-processing on the document image, the one or more region of interests obtained through segmentation on the pre-processed document image, and the tampering detected in the region of interest.

Dated this 28th day of March, 2016

R Ramya Rao
Of K&S Partners
Agent for the Applicant
, Description:TECHNICAL FIELD
This disclosure relates generally to image processing, and more particularly to methods and systems for detecting tampering in a document image by processing the document image.