Systems and methods for removing defects from images
AssigneeUnited Services Automobile Association (USAA)
InventorsReynaldo Medina, III, Bharat Prasad
Life of this patent
Estimating the term…
Abstract
A method is described to increase the efficiency of the removal of defects from document images by reorienting the conceptual framework within which an image is filtered. Rather than arbitrarily applying a filter to an entire landscape of a document image, the disclosure describes a methodology by which a document image is separated into regions of darkness and regions of light, or viewed alternatively, regions of darkness and regions of lack of darkness. Filtering is then adaptively applied to each region to remove defects.
Claims20
A computer implemented method for adaptive binarization to remove a defect from an image, the method comprising:
- receiving image data;
- performing bit plane slicing on the image data to produce a plurality of bit planes;
- analyzing the plurality of bit planes to identify a target bit plane;
- removing recognized text from the target bit plane;
- analyzing the target bit plane to determine a positional boundary for a defect in the image data;
- and
- processing the target bit plane to remove the defect, wherein a first threshold value is used for data within the positional boundary and a second threshold value is used for data outside the positional boundary.
A system comprising:
- a processor;
- and
- a memory, communicatively coupled with the processor, comprising instructions that, when executed by the processor, cause the processor to:
- receive image data;
- perform bit plane slicing on the image data to produce a plurality of bit planes;
- analyze the plurality of bit planes to identify a target bit plane;
- remove recognized text from the target bit plane;
- analyze the target bit plane to determine a positional boundary for a defect in the image data;
- and
- process the target bit plane to remove the defect, wherein a first threshold value is used for data within the positional boundary and a second threshold value is used for data outside the positional boundary.
A computer readable storage medium that is not a signal comprising executable instructions that, when executed by a computing device, cause the computing device to effectuate operations comprising:
- receiving image data;
- performing bit plane slicing on the image data to produce a plurality of bit planes;
- analyzing the plurality of bit planes to identify a target bit plane;
- removing recognized text from the target bit plane;
- analyzing the target bit plane to determine a positional boundary for a defect in the image data;
- and
- processing the target bit plane to remove the defect, wherein a first threshold value is used for data within the positional boundary and a second threshold value is used for data outside the positional boundary.
Dependencies are inferred from the claim text.