US 11,756,285 B2
Image processing system and method for image noise removal
MadhuSudhanan Krishnamoorthy, Tamil Nadu (IN); and Ramaswamy M, Tamil Nadu (IN)
Assigned to Bank of America Corporation, Charlotte, NC (US)
Filed by Bank of America Corporation, Charlotte, NC (US)
Filed on Jun. 10, 2021, as Appl. No. 17/343,975.
Prior Publication US 2022/0398407 A1, Dec. 15, 2022
This patent is subject to a terminal disclaimer.
Int. Cl. G06T 5/00 (2006.01); G06V 10/30 (2022.01); G06V 30/418 (2022.01)
CPC G06V 10/30 (2022.01) [G06T 5/002 (2013.01); G06V 30/418 (2022.01); G06T 2207/20081 (2013.01); G06T 2207/30176 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A system for removing noise artifacts from documents, comprising:
a memory operable to store a training clean image of a document, wherein the training clean image is free of noise artifacts;
a processor, operably coupled with the memory, and configured to:
receive an image of the document, wherein the image contains a noise artifact at least partially obstructing a portion of the image;
extract a first set of features from the image, wherein the first set of features represents at least one of shapes, symbols, numbers, and text in the image;
identify noise artifact features from the first set of features, wherein the noise artifact features represent pixel values of the noise artifact;
generate a second set of features by removing the noise artifact features from the first set of features;
generate a test clean image of the document based at least in part upon the second set of features as an input;
compare a portion of the test clean image that previously displayed the noise artifact with a counterpart portion of the training clean image to determine whether the noise artifact is removed from the test clean image;
determine whether the portion of the test clean image that previously displayed the noise artifact corresponds to the counterpart portion of the training clean image; and
in response to determining that the portion of the test clean image that previously displayed the noise artifact corresponds to the counterpart portion of the training clean image, output the test clean image of the document.