Small sample image classification two-stage meta-learning method based on clustering

The invention discloses a clustering-based small sample image classification two-stage meta-learning method, which applies clustering and contrast learning to small sample image classification, and comprises the following steps of: 1, determining a small sample data set, and clustering small sample image data; 2, designing a two-stage meta-learning network structure; 3, designing a loss function; 4, inputting the clustered pictures into a two-stage meta-learning network for comparison training; step 5, utilizing the trained two-stage model to carry out feature extraction on the new class of images; step 6, calculating image similarity based on multiple features and outputting a classification result; clustering and two-stage meta-learning network comparison training are introduced into small sample learning, so that the network can better extract distinguishable features of an image; on the network structure, a common feature extraction network and a personalized feature extraction network are added, triple l