The present invention provides an image processing device comprising: an input unit 201 that receives an input image; an image generation unit 202 that generates an image for extracting features from the input image; a feature extraction unit 203 that extracts features from the generated image generated by the image generation unit; an identification unit 205 that uses the features outputted by the feature extraction unit to identify an object in the image; an output unit 206 that outputs an identification result outputted by the identification unit; and a feature map generation unit 204 that, on the basis of the features outputted by the feature extraction unit, instructs the image generation unit to generate a new generated image in accordance with necessity, generates a feature map expressing conditions for extracting features from the new generated image, and outputs the feature map to the feature extraction unit. With the above configuration and operations, an object of interest in an image is identified accurately and rapidly.