Image incremental clustering method, system, medium and device

The invention belongs to the technical field of image processing, and particularly relates to an image incremental clustering method, system, medium and device. The invention aims to solve the technical problem of how to quickly process massive complex face image data samples by using limited computing resources. In order to achieve the purpose, a cluster representation is generated for each new cluster obtained by full-amount clustering, vector similarity retrieval is carried out on each new cluster in all old clusters, one or more optimal old clusters are selected from the similar old clusters, the new cluster is combined with the selected optimal old cluster, and thus taking the output combined cluster representation as the cluster representation of the optimal old cluster. In this way, the feature vectors of all image data samples in one cluster are represented through cluster representation, and the requirement for a system memory is reduced; the optimal old cluster is obtained through vector similarity