Image feature detection apparatus, image feature detection method, and storage medium

An image processing apparatus comprising:

• an amount-of-feature detection section configured to determine an amount of features of a first image and an amount of features of a second image, the first image and the second image containing the same subject and one of the first image and the second image corresponding to a first scene and the other of the first image and the second image corresponding to a second scene within a smaller area of the first scene;
• a memory configured to store the first image and the second image containing the same subject;
• an elimination-area detection section configured to eliminate, from the second image, an area in which the amount of features extracted by the amount-of-feature detection section is larger than a predetermined threshold and to provide the second image in which the eliminated area is transparent;
• and
• an image combination section configured to superimpose the second image, in which the eliminated area is transparent, on the first image, wherein
• the amount-of-feature detection section detects, as the amount of features, the amount of motion of the subject contained in the first image and the second image and determines a frequently detected amount of motion that is common to the first image and the second image,
• the elimination-area detection section eliminates a background area of a subject object having the amount of motion detected by the amount-of-feature detection section, and
• a position of the second image relative to the first image, the second image being superimposed on the first image by the image combination section, is determined on a basis of the motion of the subject contained in the first image.

An image processing method of an image processing apparatus, the method comprising the steps of:

• determining, by a processor of the image processing apparatus, an amount of features of a first image and an amount of features of a second image, the first image and the second image containing the same subject and one of the first image and the second image corresponding to a first scene and the other of the first image and the second image corresponding to a second scene within a smaller area of the first scene;
• eliminating from the second image, an area in which the amount of features is larger than a predetermined threshold and providing the second image in which the eliminated area is transparent;
• and
• superimposing the second image in which the eliminated area is transparent on the first image, wherein
• the determining includes:
• detecting, as the amount of features, color gamuts contained in the first image and the second image and represented by a predetermined color coordinate system;
• determining a frequently detected color gamut that is common to the first image and the second image;
• and
• extracting, as feature portions in the first image and the second image, color gamuts in which values obtained by accumulating, for the respective color gamuts of the first image and the second image, color-gamut chromaticities detected for respective pixels included in the first image and the second image are larger than corresponding predetermined thresholds;
• and
• the eliminating includes eliminating, in the second image, an area corresponding to the color gamut that is common to the feature portions of the first image and the second image.