| 摘要 |
The present invention relates to an image processing apparatus and method, and a program that are capable of more easily identifying an area of a subject in an image. A luminance information extraction unit 21 through to a motion information extraction unit 25 extract predetermined information from an input image and create an information map indicating feature quantities of features possessed by an area of a subject of the input image. Furthermore, the luminance information extraction unit 21 through to the motion information extraction unit 25 subtract an average value of the pixel values of the pixels of the entire information map from the pixel value of each pixel of the created information map so as to normalize the information map. As a result, the information map can be normalized, and noise can be removed with a simple process when compared to a case in which the information map is normalized using a DOG filter. A subject map creation unit 26 linearly combines each information map, and creates a subject map indicating the likelihood of an area being a subject in each area of an input image. The present invention can be applied to an image processing apparatus. |
