| 摘要 |
Provided is an image matching method which executes a highly accurate matching using arrangement of centers of gravity of coupled regions as characteristic points. Firstly, a geometric transformation parameter required for converting one image (first image) to another image (second image) is obtained. Next, the geometric transformation parameter is used to acquire a projected partial region obtained by projecting a partial region of the first image to the second image. In the projected partial region, a characteristic amount is calculated from the second image and compared to the characteristic amount calculated from the first image. |
