Domestic garbage image classification model training method and device based on comparative learning

The invention discloses a household garbage image classification model training method and device based on comparative learning, and relates to the technical field of garbage classification. Household garbage images in a training set are subjected to two kinds of image preprocessing, and an anchor point data set and a positive and negative sample set are obtained; respectively inputting the anchor point data set and the positive and negative sample sets into a comparative learning encoder and a momentum encoder, extracting multi-scale features of an encoder backbone network through a Hash feature fusion method, and generating corresponding global Hash codes; inputting the global hash code into a linear dependency group query full-connection embedded class decoder to obtain a corresponding class feature vector; calculating parameter center rebalance self-adaptive comparison loss according to category feature vector similarity; household garbage images are classified according to the comparison loss, and a hous