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

1. An image processing device comprising:

a processor comprising hardware, wherein the processor is configured to:

extract a specific candidate area that satisfies a predetermined condition from an intraluminal image captured inside a body lumen;

set a reference area that includes at least a part of the specific candidate area extracted;

extract local areas based on the reference area set;

calculate local feature data that is feature data of each of the local areas extracted;

set a weight depending on each of the local areas extracted based on the specific candidate area extracted; and

integrate the local feature data.

2. The image processing device according to claim 1 ,

wherein the processor is configured to:

set a weight for the local feature data; and

integrate the local feature data based on the weight for the local feature data.

3. The image processing device according to claim 1 ,

wherein the processor is configured to:

set, as a weight, extraction density used when the processor extracts each of the local area; and

extract the local areas extracted in accordance with the extraction density set.

4. The image processing device according to claim 1 ,

wherein the processor is configured to perform division setting such that the reference area set is divided into at least a border area and an internal area.

5. The image processing device according to claim 4 ,

wherein the processor is configured to set a higher weight for the border area than for the internal area.

6. The image processing device according to claim 1 ,

wherein the processor is configured to:

calculate a characteristic of each of the local areas; and

set a weight in accordance with the characteristic of each of the local areas.

7. The image processing device according to claim 6 ,

wherein the processor is configured to extract a mucosal area present at a depth position similar to the specific candidate area extracted.

8. The image processing device according to claim 6 ,

wherein the processor is configured to extract a same mucosal area as the specific candidate area extracted.

9. The image processing device according to claim 6 ,

wherein the processor is configured to calculate, as the characteristic, at least one of a color characteristic and a texture characteristic of each of the local areas extracted.

10. The image processing device according to claim 6 ,

wherein the processor is configured to:

determine a type of an organ where the specific candidate area extracted is present; and

set a weight in accordance with the type of the organ.

11. The image processing device according to claim 6 ,

wherein the processor is configured to:

determine a type of the specific candidate area extracted; and

set a weight in accordance with the type of the specific candidate area extracted.

12. The image processing device according to claim 1 ,

wherein the processor is configured to:

calculate feature data in an image of the reference area set; and

extract the local areas extracted based on the feature data in the image of the reference area set.

13. The image processing device according to claim 12 ,

wherein the feature data is color information.

14. The image processing device according to claim 12 ,

wherein the feature data is texture information.

15. The image processing device according to claim 1 ,

wherein the processor is configured to detect the specific candidate area extracted based on the local feature data integrated.

16. An image processing method comprising:

extracting a specific candidate area that satisfies a predetermined condition from an intraluminal image captured inside a body lumen;

setting a reference area that includes at least a part of the specific candidate area extracted;

extracting local areas based on the reference area set;

calculating local feature data that is feature data of each of the local areas extracted;

setting a weight depending on each of the local areas extracted based on the specific candidate area extracted; and

integrating the local feature data.

17. A non-transitory computer-readable recording medium recording an image processing program for causing a computer to execute:

extracting a specific candidate area that satisfies a predetermined condition from an intraluminal image captured inside a body lumen;

setting a reference area that includes at least a part of the specific candidate area extracted;

extracting local areas based on the reference area set;

calculating local feature data that is feature data of each of the local areas extracted;

setting a weight depending on each of the local areas extracted based on the specific candidate area extracted; and

integrating the local feature data.