Image processing apparatus, image processing method, and computer-readable recording medium

Patent metadata

Field	Value
Publication number	US-10456009-B2
Application number	US-201715817492-A
Country	US
Kind code	B2
Filing date	Nov 20, 2017
Priority date	May 21, 2015
Publication date	Oct 29, 2019
Grant date	Oct 29, 2019

Abstract

An image processing apparatus includes: a candidate region detection unit configured to detect, from an image acquired by imaging inside a lumen of a living body, a candidate region for a specific region that is a region where a specific part in the lumen has been captured; a candidate region information acquiring unit configured to acquire information related to the candidate region detected by the candidate region detection unit; an identification means determination unit configured to determine an identification means for identification of, based on the information related to the candidate region, whether or not the candidate region is the specific region; and an identification unit configured to identify whether or not the candidate region is the specific region by using the identification means determined by the identification means determination unit.

First claim

What is claimed is:

1. An image processing apparatus, comprising:
a hardware processor configured to:
detect, from an image acquired by imaging inside a lumen of a living body, a candidate region for a specific region that is a region where a specific part in the lumen has been captured;
acquire information related to the candidate region;
determine a difficulty of identification of whether or not the candidate region is the specific region;
select a type of identification for use in an identification of whether or not the candidate region is the specific region, wherein the selection is based on the information related to the candidate region and based on the difficulty of identification, and an identification accuracy of the selected type of identification is based on a type of the difficulty of identification; and
identify whether or not the candidate region is the specific region by using the selected type of identification.

2. The image processing apparatus according to claim 1 , wherein the hardware processor is configured to:
in response to the identification difficulty being a binary high, select a type of identification of a first identification accuracy, wherein the first identification accuracy is associated with a first number of dimensions of feature data and a first amount of calculation; and
in response to the identification difficulty being a binary low, select a type of identification of a second identification accuracy, wherein the second identification accuracy is associated with a second number of dimensions of feature data and a second amount of calculation, the second number of dimensions of feature data is less than the first number, and the second amount of calculation is less than the first amount.

3. The image processing apparatus according to claim 1 , wherein the hardware processor includes an identifier selector configured to select, based on the information related to the candidate region, an identifier to be used in the identification of the candidate region, from one or more identifiers.

4. The image processing apparatus according to claim 3 , wherein the hardware processor is configured to:
estimate a type of specific region that the candidate region belongs to, by using feature data of the image, and
select the identifier, based on the estimated type of specific region.

5. The image processing apparatus according to claim 3 , wherein
the identifier selector is configured to select two or more identifiers from the one or more identifiers,
the hardware processor is configured to:
determine an integration process for integration of two or more identification results acquired respectively by the selected two or more identifiers, and
identify whether or not the candidate region is the specific region by integrating the two or more identification results together using the determined integration process.

6. The image processing apparatus according to claim 5 , wherein the hardware processor is configured to:
estimate a type of specific region that the candidate region belongs to, by using feature data of the image, and
select the two or more identifiers by priority according to the estimated type of specific region.

7. The image processing apparatus according to claim 5 , wherein the hardware processor is configured to determine two or more weights given respectively to the two or more identification results when the two or more identification results are integrated together.

8. The image processing apparatus according to claim 7 , wherein the hardware processor is configured to:
estimate a type of specific region that the candidate region belongs to, by using feature data of the image, and
determine the two or more weights according to the estimated type of specific region.

9. An image processing method, comprising:
detecting, from an image acquired by imaging inside a lumen of a living body, a candidate region for a specific region that is a region where a specific part in the lumen has been captured;
acquiring information related to the detected candidate region;
determining a difficulty of identification of whether or not the candidate region is the specific region;
selecting a type of identification for use in an identification of whether or not the candidate region is the specific region, wherein the selection is based on the information related to the candidate region and based on the difficulty of identification, and the selected type of identification is high in identification accuracy when the difficulty of identification is high; and
identifying whether or not the candidate region is the specific region by using the selected type of identification.

10. A non-transitory computer-readable recording medium recording an image processing program that causes a computer to execute:
detecting, from an image acquired by imaging inside a lumen of a living body, a candidate region for a specific region that is a region where a specific part in the lumen has been captured;
acquiring information related to the detected candidate region;
determining a difficulty of identification of whether or not the candidate region is the specific region;
selecting a type of identification for use in an identification of whether or not the candidate region is the specific region, wherein the selection is based on the information related to the candidate region and based on the difficulty of identification, and the selected type of identification is high in identification accuracy when the difficulty of identification is high; and
identifying whether or not the candidate region is the specific region by using the selected type of identification.

11. An image processing apparatus, comprising:
a hardware processor configured to:
detect, from an image acquired by imaging inside a lumen of a living body, a candidate region for a specific region that is a region where a specific part in the lumen has been captured;
acquire information related to the candidate region;
select, based on the information related to the candidate region, two or more identifiers to be used in the identification of the candidate region, from a plurality identifiers;
determine an integration process for integration of two or more identification results acquired respectively by the selected two or more identifiers; and
identify whether or not the candidate region is the specific region by integrating the two or more identification results together using the determined integration process.

Assignees

• Olympus Corp

Inventors

• Kamiyama Toshiya
• Kitamura Makoto
• Kimura Mitsutaka

Classifications

• G06T2207/30092

  Stomach; Gastric · CPC title

• G06T7/44

  using image operators, e.g. filters, edge density metrics or local histograms · CPC title

• G06T2207/30096

  Tumor; Lesion · CPC title

• G06T7/75

  involving models · CPC title

• G06T7/90

  Determination of colour characteristics · CPC title