Image processing apparatus and computer program

Abstract

A technique for achieving illustrative interpretability in an ensemble learning model comprised of a plurality of machine learning models is provided. An image processing apparatus includes: a plurality of discrimination units 31, 32, and 33 that perform image processing on a target image Di; and an integration unit 34 that performs integration processing based on the outputs of the discrimination units 31, 32, and 33. The discrimination units 31, 32, and 33 respectively use machine learning models M1, M2, and M3 to output discrimination results Ra, Rb, and Rc for the target image Di, and individual heatmaps Ha, Hb, and Hc representing the responsiveness of each pixel of the target image Di during the discrimination process. The integration unit 34 generates an integrated discrimination result Rt for the target image Di based on the plurality of discrimination results Ra, Rb, and Rc, and generates an integrated heatmap Ht representing the responsiveness of each pixel of the target image Di in the integrated discrimination result Rt based on the plurality of individual heatmaps Ha, Hb, and Hc.

Images