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

1. An image processing apparatus comprising:

a region-of-interest setting unit configured to set a region of interest in an image;

a linear convex region extracting unit configured to extract, from the region of interest, a linear region having a predetermined number or more of continuously-arranged pixels whose pixel values are higher than pixel values of neighboring pixels;

an intra-region curvature feature data computing unit configured to compute curvature feature data based on curvatures of one or more arcs along the linear region; and

an abnormality determining unit configured to determine whether there is an abnormal portion in the region of interest, based on a distribution of the curvature feature data.

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

the intra-region curvature feature data computing unit comprises a size feature data computing unit configured to compute distance information and the curvatures of one or more arcs, and

the abnormality determining unit is configured to determine that there is an abnormal portion in the region of interest when the distribution of the curvatures is within a range smaller than a predetermined threshold value, the predetermined threshold value being determined according to the distance information.

3. The image processing apparatus according to claim 2 , wherein

the size feature data computing unit comprises:

a curvature computing unit configured to compute the curvatures of one or more arcs from each of sections of the linear region, the sections being delimited by endpoints of the linear region and/or intersection points of linear regions;

a curvature representative value computing unit configured to compute a representative value from the curvatures of one or more arcs; and

a distance information computing unit configured to compute the distance information from an imaging position of the image to the linear region.

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

the intra-region curvature feature data computing unit comprises a shape feature data computing unit configured to compute a variation in the curvatures of one or more arcs, and

when the variation is greater than a predetermined value, the abnormality determining unit is configured to determine that there is an abnormal portion in the region of interest.

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

the shape feature data computing unit comprises:

a curvature computing unit configured to compute the curvatures of one or more arcs along the linear region from each of sections of the linear region, the sections being delimited by endpoints of the linear region and/or intersection points of linear regions; and

a curvature standard deviation computing unit configured to compute, for each of the sections, a standard deviation of the curvatures of one or more arcs, wherein

the abnormality determining unit is configured to make a determination based on the standard deviation.

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

the intra-region curvature feature data computing unit comprises a gradient direction feature data computing unit configured to compute directions going toward a curvature center from each of the one or more arcs, and directions in which an object is inclined in a depth direction of the image, at positions of each of the one or more arcs, and

when a variance of frequency of the directions going toward the curvature center is less than or equal to a predetermined threshold value, the abnormality determining unit is configured to determine that there is an abnormal portion in the region of interest, the variance of frequency being created for each of the directions in which the object is inclined.

7. The image processing apparatus according to claim 1 , wherein when it is determined that there is no abnormal portion in the set region of interest, the region-of-interest setting unit is configured to reduce a size of the region of interest to re-set the region of interest.

8. An image processing method comprising:

setting a region of interest in an image;

extracting, from the region of interest, a linear region having a predetermined number or more of continuously-arranged pixels whose pixel values are higher than pixel values of neighboring pixels;

computing curvature feature data based on curvatures of one or more arcs along the linear region; and

determining whether there is an abnormal portion in the region of interest, based on a distribution of the curvature feature data.

9. A non-transitory computer-readable recording medium with an executable program stored thereon, the program instructing a processor to execute:

setting a region of interest in an image;

extracting, from the region of interest, a linear region having a predetermined number or more of continuously-arranged pixels whose pixel values are higher than pixel values of neighboring pixels;

computing curvature feature data based on curvatures of one or more arcs along the linear region; and

determining whether there is an abnormal portion in the region of interest, based on a distribution of the curvature feature data.