# US 2015/0085326 A1 · Method and apparatus for using an enlargement operation to reduce visually detected defects in an image

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2015/0085326 A1
- Document ID: US20150085326A1
- Status: Published application
- Kind: A1
- Application number: 14/314,081
- Filed: 2014-06-25
- Published: 2015-03-26
- Earliest claimed: 2013-09-25
- Assignees: ABBYY Development LLC
- Inventors: Mikhail Kostyukov; Ivan Zagaynov
- Web page: https://patentgrep.com/patent/US20150085326A1

## CPC

- H04N 1/409: Edge or detail enhancement; Noise or error suppression
- H04N 1/387: Composing, repositioning or otherwise {geometrically} modifying originals
- H04N 1/3871: {the composed originals being of different kinds, e.g. low- and high-resolution originals}
- H04N 1/393: Enlarging or reducing
- H04N 1/3935: {with modification of image resolution, i.e. determining the values of picture elements at new relative positions}
- H04N 1/4092: {Edge or detail enhancement}
- H04N: PICTORIAL COMMUNICATION, e.g. TELEVISION

## Abstract

Systems and method for improving the quality of document images are provided. One method includes identifying a plurality of image fragments within a previously received document image that includes text. The method further includes separating the plurality of image fragments into a plurality of classes. Each class includes a subset of the plurality of image fragments that are substantially similar to one another. The method further includes, for each of the plurality of classes: (1) processing a class of image fragments to generate a combined and substantially enlarged image fragment for the class; and (2) filtering the combined and substantially enlarged image fragment to generate a filtered image fragment for the class. The method further includes generating an improved document image by replacing or modifying the image fragments within the document image based on the filtered image fragments.

## Claims

### Claim 1

A method for reducing visually detectable defects in a document image, the method comprising:
identifying, using a computing device comprising one or more processors, a plurality of image fragments within the document image;
separating, using the computing device, the plurality of image fragments into a plurality of classes, wherein each class of the plurality of classes includes a subset of the plurality of image fragments that are substantially similar to one another;
for each of the plurality of classes:
processing, using the computing device, the class of image fragments to generate a combined and substantially enlarged image of the class;
filtering, using the computing device, the combined and substantially enlarged image of the class to generate a filtered image for the class; and
generating, using the computing device, an improved document image by replacing or modifying the image fragments within the document image based on the filtered images for the respective classes of the image fragments.

### Claim 2 (depends on 1)

The method of claim 1, wherein processing the class of image fragments to generate the combined and substantially enlarged image of the class comprises:
generating a combined image of the class based on image fragments of the class; and
generating the combined and substantially enlarged image of the class by substantially enlarging the combined image of the class.

### Claim 3 (depends on 1)

The method of claim 1, wherein processing the class of image fragments to generate the combined and substantially enlarged image of the class comprises:
substantially enlarging image fragments of the class to generate substantially enlarged image fragments; and
combining the substantially enlarged image fragments to generate the combined and substantially enlarged image of the class.

### Claim 4 (depends on 1)

The method of claim 1, further comprising, for each of the plurality of classes, normalizing the image fragments of the class prior to processing the class of image fragments to generate the combined and substantially enlarged image of the class.

### Claim 5 (depends on 4)

The method of claim 4, further comprising, for each of the plurality of classes, performing an inverse normalization on each image fragment in the class of image fragments prior to generating the improved document image, wherein generating the improved document image comprises replacing the image fragments within the document image with corresponding improved image fragments within the plurality of classes after the inverse normalization has been performed.

### Claim 6 (depends on 1)

The method of claim 1, wherein processing the class of image fragments to generate the combined and substantially enlarged image of the class comprises averaging the image fragments of the class.

### Claim 7 (depends on 1)

The method of claim 1, wherein filtering the combined and substantially enlarged image of the class comprises performing at least one of a rank filtering procedure and a contour filtering procedure on the combined and substantially enlarged image of the class to generate the filtered image for the class.

### Claim 8 (depends on 1)

The method of claim 1, wherein each of the plurality of image fragments represents one of a single character or a single character part, and separating the plurality of image fragments into a plurality of classes comprises grouping together all image fragments representing a same character or same character part within a single class of the plurality of classes.

### Claim 9 (depends on 1)

The method of claim 1, wherein the plurality of classes comprises a first plurality of classes, and wherein separating the plurality of image fragments comprises separating the plurality of image fragments into the first plurality of classes and a second plurality of classes, wherein each of the second plurality of classes comprises a single image fragment having no other substantially similar image fragment within the plurality of image fragments, wherein the method further comprises, for each of the second plurality of classes, filtering the single image fragment of the class to generate a filtered image fragment for the class.

### Claim 10 (depends on 1)

The method of claim 1, wherein the document image comprises a scanned document image received from a third party, and wherein generating the improved document image comprises restoring at least a portion of image quality lost due to at least one of a scanning procedure or a file size reduction procedure prior to receipt of the document image by the computing device.

### Claim 11 (depends on 10)

The method of claim 10, wherein the document image has been previously received from a scanning device, wherein the document image was scanned by the scanning device from at least one physical document, and wherein a quality of the previously received document image is lower than a quality of the at least one physical document.

### Claim 12 (depends on 11)

The method of claim 11, wherein the quality of the previously received document image is lower than the quality of the at least one physical document due to use of improper scanning device settings during the scanning procedure, and wherein generating the improved document image comprises restoring at least a portion of the image quality lost due to the use of the improper scanning device settings without requiring the document to be rescanned.

### Claim 13 (depends on 1)

The method of claim 1, wherein processing the class of image fragments to generate the combined and substantially enlarged image fragment for the class comprises substantially enlarging one or more image fragments such that they are noticeably larger when presented to a user and defects in the one or more image fragments are more noticeable to the user.

### Claim 14

A system comprising:
at least one computing device operably coupled to at least one memory and configured to:
identify a plurality of image fragments within a document image;
separate the plurality of image fragments into a plurality of classes, wherein each class of the plurality of classes includes a subset of the plurality of image fragments that are substantially similar to one another;
for each of the plurality of classes:
process the class of image fragments to generate a combined and substantially enlarged image of the class;
filter the combined and substantially enlarged image of the class to generate a filtered image for the class; and
generate an improved document image by replacing or modifying the image fragments within the document image based on the filtered images for the respective classes of the image fragments.

### Claim 15 (depends on 14)

The system of claim 14, wherein the at least one computing device is configured to process the class of image fragments to generate the combined and substantially enlarged image of the class by:
averaging image fragments of the class to generate an averaged image of the class; and
generating the combined and substantially enlarged image of the class by substantially enlarging the averaged image of the class.

### Claim 16 (depends on 14)

The system of claim 14, wherein the at least one computing device is configured to process the class of image fragments to generate the combined and substantially enlarged image of the class by:
substantially enlarging image fragments of the class to generate substantially enlarged image fragments; and
averaging the substantially enlarged image fragments to generate the combined and substantially enlarged image of the class.

### Claim 17 (depends on 14)

The system of claim 14, wherein the at least one computing device is further configured to, for each of the plurality of classes, normalize the image fragments of the class prior to processing the class of image fragments to generate the combined and substantially enlarged image of the class.

### Claim 18 (depends on 17)

The system of claim 17, wherein the at least one computing device is further configured to, for each of the plurality of classes, perform an inverse normalization on each image fragment in the class of image fragments prior to generating the improved document image, wherein the at least one computing device is configured to replace the image fragments within the document image with corresponding improved image fragments within the plurality of classes after the inverse normalization has been performed.

### Claim 19 (depends on 14)

The system of claim 14, wherein the at least one computing device is configured to filter the combined and substantially enlarged image of the class by performing at least one of a rank filtering procedure and a contour filtering procedure on the combined and substantially enlarged image of the class to generate the filtered image for the class.

### Claim 20 (depends on 14)

The system of claim 14, wherein each of the plurality of image fragments represents one of a single character or a single character part, and wherein the at least one computing device is configured to separate the plurality of image fragments into a plurality of classes by grouping together all image fragments representing a same character or same character part within a single class of the plurality of classes.

### Claim 21

A computer-readable storage medium having instructions stored thereon that, when executed by a processor, cause the processor to perform operations comprising:
identifying a plurality of image fragments within a document image, wherein each of the plurality of image fragments represents a one of a single character or a single character part;
separating the plurality of image fragments into a plurality of classes, wherein each class of the plurality of classes includes a subset of the plurality of image fragments that are substantially similar to one another, and wherein image fragments representing a same character or same character part are grouped within a single class of the plurality of classes;
for each of the plurality of classes:
normalizing image fragments of the class to generate a normalized set of image fragments of the class;
processing the normalized set of image fragments of the class to generate a combined and substantially enlarged image of the class;
filtering the combined and substantially enlarged image of the class using at least one of a rank filtering procedure and a contour filtering procedure to generate a filtered image for the class;
modifying each image fragment of the normalized set of image fragments of the class based on the filtered image for the class; and
performing an inverse normalization on each image fragment of the normalized set of image fragments to generate a final set of image fragments for the class; and
generating an improved document image by replacing or modifying the image fragments within the document image based on the image fragments in the final set of image fragments for the respective classes of the image fragments.

### Claim 22 (depends on 21)

The computer-readable storage medium of claim 21, wherein processing the normalized set of image fragments to generate the combined and substantially enlarged image of the class comprises:
averaging the normalized set of image fragments to generate an averaged image of the class; and
generating the combined and substantially enlarged image of the class by substantially enlarging the averaged image of the class.

### Claim 23 (depends on 21)

The computer-readable storage medium of claim 21, wherein processing the normalized set of image fragments to generate the combined and substantially enlarged image of the class comprises:
substantially enlarging the normalized set of image fragments to generate two or more substantially enlarged image fragments; and
averaging the two or more substantially enlarged image fragments to generate the combined and substantially enlarged image of the class.
