Image processing apparatus and image processing method

An image processing apparatus comprising:

• a unit which acquires N images sensed under N respective types of image sensing conditions;
• a management unit which extracts respective feature amounts from the N sensed images, and manages the extracted feature amounts as a feature amount cluster, in which each of the extracted feature amounts is in association with corresponding one of the image sensing conditions;
• a specification unit which specifies, for a first feature amount cluster of feature amount clusters managed by said management unit, among the feature amount clusters, a second feature amount cluster containing a feature amount similar to a feature amount of interest in the first feature amount cluster;
• and
• a determination unit which determines an image sensing condition for discriminating the first and second feature amount clusters, based on an image sensing condition associated with the feature amount of interest and an image sensing condition associated with the feature amount being similar to the feature amount of interest and being contained in the second feature amount cluster specified by said specification unit.

An image processing method comprising:

• a step of acquiring N images sensed under N respective types of image sensing conditions;
• a management step of extracting respective feature amounts from the N sensed images, and managing the extracted feature amounts as a feature amount cluster, in which each of the extracted feature amounts is in association with corresponding one of the image sensing conditions;
• a first specification step of specifying a second feature amount cluster containing a feature amount similar to a feature amount of interest in a first feature amount cluster among feature amount clusters managed in the management step;
• a second specification step of specifying sets of feature amounts, each of the sets being associated with a corresponding one of the N respective types of image sensing conditions from the first feature amount cluster and the second feature amount cluster;
• a third specification step of specifying a set having a largest distance between feature amounts in the set among the sets specified in the second specification step;
• and
• a determination step of determining an image sensing condition associated with the set specified in the third specification step, as an image sensing condition for discriminating the first feature amount cluster and the second feature amount cluster,
• wherein at least one of said steps is performed with a processor.

An image processing method comprising:

• a step of acquiring N images sensed under N respective types of image sensing conditions;
• a management step of extracting respective feature amounts from the N sensed images, and managing the extracted feature amounts as a feature amount cluster, in which each of the extracted feature amounts is in association with corresponding one of the image sensing conditions;
• a specification step of specifying, for a first feature amount cluster of feature amount clusters managed by said management step, among the feature amount clusters, a second feature amount cluster containing a feature amount similar to a feature amount of interest in the feature amount cluster;
• and
• a determination step of determining an image sensing condition for discriminating the first and second feature amount clusters, based on an image sensing condition associated with the feature amount of interest and an image sensing condition associated with the feature amount being similar to the feature amount of interest and being contained in the second feature amount cluster specified in the specification step,
• wherein at least one of said steps is performed with a processor.