Image processing device, method, and program

An image processing method, comprising:

• acquiring a plurality of input images of the same subject photographed at different points in time;
• setting, within an input image that is photographed at a reference time point out of the plurality of acquired input images, a first region estimated as highly probable to be a specific region and a second region estimated as highly probable to be a background region, which is a region other than the specific region;
• calculating a first histogram, which is a density histogram of respective density values of pixels in the first region, and a second histogram, which is a density histogram of respective density values of pixels in the second region, and extracting the specific region within the input image that is photographed at the reference time point, based on the first histogram and the second histogram;
• and
• extracting, based on the first histogram and the second histogram which are calculated with use of the input image photographed at the reference time point, a specific region corresponding to the specific region within the input image that is photographed at the reference time point from one of the plurality of input images that is photographed at a time point different from the reference time point,
• wherein the extracting includes:
• calculating, for each density value, based on the first histogram and the second histogram, a first evaluation value which indicates a likelihood that the density value represents the specific region;
• calculating, for every two adjacent pixels within the input image, a second evaluation value which indicates a likelihood that the two adjacent pixels represent an outline, based on the respective first evaluation values of the density values of the two adjacent pixels;
• and
• defining a first reference vertex which belongs to the specific region, a second reference vertex which belongs to the background region, and a plurality of vertices respectively representing the pixels within the input image, set a connection cost, which indicates how likely the plurality of vertices are to be connected to each other, based on the second evaluation value, and connect the vertices to each other based on the set connection cost, thereby extracting the specific region.

A non-transitory computer-readable recording medium containing a program for causing a computer to execute the procedures of:

• acquiring a plurality of input images of the same subject photographed at different points in time;
• setting, within an input image that is photographed at a reference time point out of the plurality of acquired input images, a first region estimated as highly probable to be a specific region and a second region estimated as highly probable to be a background region, which is a region other than the specific region;
• calculating a first histogram, which is a density histogram of respective density values of pixels in the first region, and a second histogram, which is a density histogram of respective density values of pixels in the second region, and extracting the specific region within the input image that is photographed at the reference time point, based on the first histogram and the second histogram;
• and
• extracting, based on the first histogram and the second histogram which are calculated with use of the input image photographed at the reference time point, a specific region corresponding to the specific region within the input image that is photographed at the reference time point from one of the plurality of input images that is photographed at a time point different from the reference time point,
• wherein the extracting includes:
• calculating, for each density value, based on the first histogram and the second histogram, a first evaluation value which indicates a likelihood that the density value represents the specific region;
• calculating, for every two adjacent pixels within the input image, a second evaluation value which indicates a likelihood that the two adjacent pixels represent an outline, based on the respective first evaluation values of the density values of the two adjacent pixels;
• and
• defining a first reference vertex which belongs to the specific region, a second reference vertex which belongs to the background region, and a plurality of vertices respectively representing the pixels within the input image, set a connection cost, which indicates how likely the plurality of vertices are to be connected to each other, based on the second evaluation value, and connect the vertices to each other based on the set connection cost, thereby extracting the specific region.