| 摘要 |
PROBLEM TO BE SOLVED: To precisely extract the positions and the number of objects to be detected from photographed images. SOLUTION: This method comprises a first step of extracting boundaries between regions different in brightness from a first image and finding a first edge image, a second step of extracting boundaries between regions different in brightness from a second image and finding second edge image, a third step of finding a difference image from the first edge image and the second edge image, a fourth step of extracting an object existence candidate region to be a candidate for an object existence region and background candidate regions other than the object existence candidate region from the difference image, a fifth step of finding a relation expression showing a brightness relation between corresponding pixels in the background candidate regions for the first and second images, and a sixth step of extracting the object existence region corresponding to a difference between the brightness of the corresponding pixels in the object existence candidate regions for the first and second images and the brightness found from the relational expression. |
