IP Library Granted Patent US 10,771,636
Granted Patent B2
US 10,771,636 · App. 16/124,098 · Granted Sep 8, 2020

Image processing apparatus for detecting a noise line based on noise pixels in the shadow region

Inventors: Takeshi Yokogawa (Kahoku, JP); Kunihiko Ohmi (Kahoku, JP); Kodai Yamane (Kahoku, JP)
Assignee: PFU LIMITED
H04N1/00029G06K9/346H04N5/2171H04N5/3675G06T2207/10008
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 10,771,636
App. No.
16/124,098
Granted
Sep 8, 2020
Kind
B2
Abstract

An image processing apparatus includes a processor for acquiring an input image of a document and a periphery of the document, extracting first edge pixels from the input image, detecting a document region based on the first edge pixels, detecting a shadow region from within a predetermined range outside of the document region, extracting noise pixels in the shadow region, detecting the noise pixels, a number of which connected with one another is equal to or more than a predetermined number, as a noise line, and correcting the document region of the input image based on a detection result of the noise line.

Claims (46)

1. An image processing apparatus comprising:

a processor for

acquiring an input image of a document and a periphery of the document,

extracting first edge pixels from the input image,

detecting a document region based on the first edge pixels,

detecting a shadow region from within a predetermined range outside of the document region,

extracting noise pixels in the shadow region,

detecting the noise pixels, a number of which connected with one another is equal to or more than a predetermined number, as a noise line, and

correcting the document region of the input image based on a detection result of the noise line.

2. The image processing apparatus according to claim 1 , wherein the processor extracts, as the noise pixels, pixels where a difference between a gradient value of the shadow region and a gradient value of the noise pixels is equal to or more than a threshold.

3. The image processing apparatus according to claim 1 , wherein

the processor

acquires device information that indicates arrangement of an imaging device and a light source in a device that generates the input image, and

determines the predetermined range based on the device information.

4. The image processing apparatus according to claim 1 , wherein

the processor

extracts second edge pixels from within the predetermined range,

detects a plurality of line segments based on the second edge pixels, and

detects, as the shadow region, a region within a second predetermined range from each of the plurality of line segments.

5. The image processing apparatus according to claim 1 , wherein

the processor

specifies, in the document region, a noise line region corresponding to a noise line detected in the shadow region,

determines whether the noise line region overlaps a content based on a difference between a gradient value of the noise line region and a gradient value of peripheral pixels of the noise line region, and

changes a method of correcting the noise line region according to a result of the determining whether the noise line region overlaps a content.

6. The image processing apparatus according to claim 5 , wherein the processor corrects the noise line region based on the peripheral pixels of the noise line region when the processor determines that the noise line region does not overlap a content.

7. The image processing apparatus according to claim 5 , wherein

when the processor determines that the noise line region overlaps a content, the processor further determines whether the content is a character,

when the processor determines that the content is not a character, the processor does not correct the noise line region, and

when the processor determines that the content is a character, the processor corrects the noise line region.

8. The image processing apparatus according to claim 7 , wherein when the processor determines that the content is a character, the processor estimates a character overlapping the noise line region from a context of the content and corrects the noise line region based on the estimated character.

9. A control method of an image processing apparatus, the method comprising:

acquiring an input image of a document and a periphery of the document;

extracting first edge pixels from the input image;

detecting a document region based on the first edge pixels;

detecting a shadow region from within a predetermined range outside of the document region;

extracting noise pixels in the shadow region;

detecting the noise pixels, a number of which connected with one another is equal to or more than a predetermined number, as a noise line; and

correcting the document region of the input image based on a detection result of the noise line.

10. A computer-readable, non-transitory medium storing a computer program, wherein the computer program causes an image processing apparatus to execute a process, the process comprising:

acquiring an input image of a document and a periphery of the document;

extracting first edge pixels from the input image;

detecting a document region based on the first edge pixels;

detecting a shadow region from within a predetermined range outside of the document region;

extracting noise pixels in the shadow region;

detecting the noise pixels, a number of which connected with one another is equal to or more than a predetermined number, as a noise line; and

correcting the document region of the input image based on a detection result of the noise line.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Sep 17, 2018
From: YOKOGAWA, TAKESHI; OHMI, KUNIHIKO; YAMANE, KODAI
To: PFU LIMITED
Reel/Frame 046886/0900 →
Priority Claims (1)
JP 2018-044769 · Mar 12, 2018 · national
Continuity (1)
Related Publication 20190281170A1 · Sep 12, 2019