# US 8,542,931 B2 · Ruled line extraction technique based on comparision results and indentifying noise based on line thickness

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

- Number: US 8,542,931 B2
- Document ID: US8542931B2
- Status: Granted
- Kind: B2
- Application number: 12/552,627
- Filed: 2009-09-02
- Published: 2010-08-05
- Granted: 2013-09-24
- Earliest claimed: 2009-02-05
- Assignees: Fuji Xerox Co., Ltd.
- Applicants: Kunikazu Ueno; Shinichi Yada
- Inventors: Kunikazu Ueno; Shinichi Yada
- Web page: https://patentgrep.com/patent/US8542931B2

## CPC

- G06V 20/52: Surveillance or monitoring of activities, e.g. for recognising suspicious objects (recognising microscopic objects G06V20/69)
- G06F 18/22: Matching criteria, e.g. proximity measures
- G06V 10/761: {Proximity, similarity or dissimilarity measures}
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

An image processing apparatus includes: a ruled line extracting unit that counts the number of pixels within an image, compares the counted number of pixels with a threshold value, and extracts a ruled line based on a result of the comparison; and an identifying unit that identifies a noise component in the ruled line extracted by the ruled line extracting unit based on thickness of the ruled line extracted by the ruled line extracting unit and the threshold value.

## Claims

### Claim 1

An image processing apparatus comprising at least one processor; and
at least one memory, the memory storing instructions which when executed by the at least one processor cause the at least one processor to function as:
a ruled line extracting unit that counts the number of pixels within an image, compares the counted number of pixels with a threshold value, and extracts a ruled line based on a result of the comparison; and
an identifying unit that identifies a noise component in the ruled line extracted by the ruled line extracting unit based on thickness of the ruled line extracted by the ruled line extracting unit and the threshold value the identifying unit generating one of:
(1) a filter having a horizontal width equal to the threshold value multiplied by α, and has a vertical length equal to the ruled line width multiplied by β; and
(2) a filter having a horizontal width equal to the ruled line width multiplied by β, and has a vertical length equal to the threshold value multiplied by α,
wherein α<1, and β≧1.

### Claim 2 (depends on 1)

The image processing apparatus according to claim 1, further comprising:
a solid-lining unit that converts the ruled line extracted by the ruled line extracting unit into a solid line,
wherein the identifying unit identifies the noise component in the ruled line converted into the solid line by the solid-lining unit.

### Claim 3 (depends on 2)

The image processing apparatus according to claim 2,
wherein the filter generated by the identifying unit has a rectangular form and the length of the rectangular form is determined based on the thickness of the ruled line extracted by the ruled line extracting unit.

### Claim 4 (depends on 2)

The image processing apparatus according to claim 2,
wherein the filter generated by the identifying unit has a rectangular form and the length of the rectangular form is determined based on the threshold value.

### Claim 5 (depends on 1)

The image processing apparatus according to claim 1, further comprising:
a solid-lining unit that converts the ruled line extracted by the ruled line extracting unit into a solid line,
wherein the identifying unit identifies the noise component in the ruled line converted into the solid line by the solid-lining unit.

### Claim 6

A non-transitory computer readable medium storing a program causing a computer to execute a process for image processing, the process comprising:
counting the number of pixels within an image;
comparing the counted number of pixels with a threshold value;
extracting a ruled line based on a result of the comparison;
identifying a noise component in the extracted ruled line based on thickness of the extracted ruled line extracted and the threshold value; and
generating one of:
(1) a filter having a horizontal width equal to the threshold value multiplied by α, and has a vertical length equal to the ruled line width multiplied by β; and
(2) a filter having a horizontal width equal to the ruled line width multiplied by β, and has a vertical length equal to the threshold value multiplied by α,
wherein α<1, and β≧1.

### Claim 7 (depends on 6)

The non-transitory computer readable medium according to claim 6, further comprising:
converting the extracted ruled line into a solid line,
wherein identifying the noise component includes identifying the noise component in the ruled line converted into the solid line.

### Claim 8 (depends on 7)

The non-transitory computer readable medium according to claim 7,
wherein the generated filter has a rectangular form and the length of the rectangular form is determined based on the thickness of the extracted ruled line.

### Claim 9 (depends on 6)

The non-transitory computer readable medium according to claim 6, the process further comprising:
converting the extracted ruled line into a solid line,
wherein identifying the noise component includes identifying the noise component in the ruled line converted into the solid line.

### Claim 10

An image processing apparatus comprising at least one processor: and at least one memory, the memory storing instructions which when executed by the at least one processor cause the at least one processor to function as:
a horizontal ruled line extracting unit that counts the number of pixels within a horizontal direction at each position of a vertical direction of an image, compares the counted number of pixels with a horizontal threshold value, and extracts a horizontal ruled line based on a result of the comparison;
a vertical ruled line extracting unit that counts the number of pixels within the vertical direction at each position of the horizontal direction of an image, compares the counted number of pixels with a vertical threshold value, and extracts a vertical ruled line based on a result of the comparison;
a horizontal identifying unit that identifies a noise component in the horizontal ruled line extracted by the horizontal ruled line extracting unit based on thickness of the horizontal ruled line extracted by the horizontal ruled line extracting unit and the horizontal threshold value; and
a vertical identifying unit that identifies a noise component in the vertical ruled line extracted by the vertical ruled line extracting unit based on thickness of the vertical ruled line extracted by the vertical ruled line extracting unit and the vertical threshold value,
wherein the horizontal identifying unit and the vertical identifying unit generate one of:
(1) a filter having a horizontal width equal to the threshold value multiplied by α, and has a vertical length equal to the ruled line width multiplied by β; and
(2) a filter having a horizontal width equal to the ruled line width multiplied by β, and has a vertical length equal to the threshold value multiplied by α, wherein α<1, and β≧1.