Image recognition device, image recognizing method, storage medium that stores computer program for image recognition

An object identifying device, comprising:na memory; anda processor coupled to the memory and configured to extract a first feature amount from a first image data;generate a first identification result of the first image data based on the first feature amount and a first identifier which identifies a plurality of objects included in a first category, the first identification result indicating any one of the plurality of objects;select, based on the first identification result, a second category from among the second category and a third category, the second category including a second part of the plurality of objects, the third category being different from the second category and including a third part of the plurality of objects;extract a second feature amount from a second image data; andgenerate a second identification result of the second image data based on the second feature amount, the second category, and a second identifier which identifies the second part of the plurality of objects, the second identification result indicating any one of the second part of the plurality of objects.

A non-transitory storage medium storing image identification program that causes a processor to execute operations, the operations comprising:nextracting a first feature amount from a first image data;generating a first identification result of the first image data based on the first feature amount and a first identifier which identifies a plurality of objects included in a first category, the first identification result indicating any one of the plurality of objects;selecting, based on the first identification result, a second category from among the second category and a third category, the second category including a second part of the plurality of objects, the third category being different from the second category and including a third part of the plurality of objects;extracting a second feature amount from a second image data; andgenerating a second identification result of the second image data based on the second feature amount, the second category, and a second identifier which identifies the second part of the plurality of objects, the second identification result indicating any one of the second part of the plurality of objects.

A method of identifying an object comprising:nextracting a first feature amount from a first image data;generating a first identification result of the first image data based on the first feature amount and a first identifier which identifies a plurality of objects included in a first category, the first identification result indicating any one of the plurality of objects;selecting, based on the first identification result, a second category from among the second category and a third category, the second category including a second part of the plurality of objects, the third category being different from the second category and including a third part of the plurality of objects;extracting a second feature amount from a second image data; andgenerating a second identification result of the second image data based on the second feature amount, the second category, and a second identifier which identifies the second part of the plurality of objects, the second identification result indicating any one of the second part of the plurality of objects.