Image identification device, image identification method, and recording medium

1. An image identification device that learns in advance a separating plane used for classifying images into predetermined categories in a feature quantity space using a feature of the image as a coordinate axis and uses the separating plane to classify block images obtained by dividing a target image into the categories, the image identification device comprising:

an input unit that inputs the target image;

a block image generation unit that divides the target image into a plurality of blocks to generate the block images;

a feature quantity computing unit that computes feature quantities of the block images; and

a category determination unit that determines whether the block images are classified into one of the categories or not, using the separating plane and coordinate positions corresponding to magnitudes of feature quantities of the block images in the feature quantity space,

wherein

the feature quantity computing unit uses, as a feature quantity of a given target block image in the block images, local feature quantities calculated according to image information of the target block image and a global feature quantity calculated according to image information of the target image.

2. The image identification device according to claim 1 , wherein the feature quantity computing unit uses, as the global feature quantity, a feature quantity calculated according to image information of four corner areas in the target image.

3. The image identification device according to claim 1 , wherein the feature quantity computing unit uses, as the feature quantity of the target block image, a neighbor feature quantity calculated according to image information of neighboring block images that are block images surrounding the target block image.

4. The image identification device according to claim 2 , wherein the feature quantity computing unit uses, as the feature quantity of the target block image, a neighbor feature quantity calculated according to image information of neighboring block images that are block images surrounding the target block image.

5. The image identification device according to claim 3 , wherein the feature quantity computing unit uses, as the neighbor feature quantity, a feature quantity calculated according to image information of block image selected from the neighboring block images.

6. The image identification device according to claim 4 , wherein the feature quantity computing unit uses, as the neighbor feature quantity, a feature quantity calculated according to image information of block image selected from the neighboring block images.

7. The image identification device according to claim 5 , wherein the feature quantity computing unit selects, among the neighboring block images of the target block image, ones existing symmetrically with respect to the target block image.

8. The image identification device according to claim 6 , wherein the feature quantity computing unit selects, among the neighboring block images of the target block image, ones existing symmetrically with respect to the target block image.

9. The image identification device according to claim 1 , wherein the feature quantity computing unit uses, as the feature quantity of the target block image, a sub-global feature quantity calculated according to image information of a partial area comprising a plurality of the block images including the target block image.

10. The image identification device according to claim 9 , wherein the feature quantity computing unit uses, as the sub-global feature quantity, a feature quantity calculated according to image information of a horizontal long partial area comprising the block images including the target block image, and a feature quantity calculated according to image information of a vertical long partial area comprising the block images including the target block image.

11. An image identification method for learning in advance a separating plane used for classifying images into predetermined categories in a feature quantity space using a feature of the image as a coordinate axis and then using the separating plane to classify block images obtained by dividing a target image into the categories, the image identification method comprising:

an input step of inputting the target image;

a block image generation step of dividing the target image into a plurality of blocks to generate the block images;

a feature quantity computing step of computing feature quantities of the block images; and

a category determination step of determining whether the block images are classified into one of the categories or not, using the separating plane and coordinate positions corresponding to magnitudes of feature quantities of the block images in the feature quantity space,

wherein

at the feature quantity computing step,

as a feature quantity of a given target block image in the block images, local feature quantities calculated according to image information of the target block image and a global feature quantity calculated according to image information of the target image are used.

12. A non-transitory computer-readable recording medium in which an image identification program is stored that causes a computer to operate so as to learn in advance a separating plane used for classifying images into predetermined categories in a feature quantity space using a feature of the image as a coordinate axis and use the separating plane to classify block images obtained by dividing a target image into the categories, the recording medium storing therein an image identification program that causes the computer to operate as: an input unit that inputs the target image; a block image generation unit that divides the target image into a plurality of blocks to generate the block images; a feature quantity computing unit that computes feature quantities of the block images, wherein the feature quantity computing unit uses, as a feature quantity of a given target block image in the block images, local feature quantities calculated according to image information of the target block image and a global feature quantity calculated according to image information of the target image; and a category determination unit that determines whether the target block image is classified into one of the categories or not, using the separating plane and coordinate positions corresponding to magnitudes of feature quantities of the target block image in the feature quantity space.