Face image clustering method and device, server and storage medium

Abstract

The embodiment of the invention discloses a face image clustering method and device, a server and a storage medium. The method comprises the steps of obtaining a target face image collected by a target image collector; determining a target area where the target image collector is located, and obtaining all face centroids belonging to the target area, wherein the target area is any geographic area obtained by performing area division on the image collector based on a division rule in advance, the face centroid is used for representing a clustered face image set; and comparing the target face image with all face centroids belonging to the target area, and determining the category of the target face according to a comparison result. According to the embodiment of the invention, the target face is compared with the face centroid belonging to the target area, the clustering accuracy can be improved, the comparison between the target face image and the full face centroid is avoided during the comparison and clustering, namely, the cardinality of centroid comparison is reduced, and the clustering efficiency is improved.

Description

technical field [0001] The embodiments of the present application relate to the technical field of video surveillance, and in particular to a face image clustering method, device, server and storage medium. Background technique [0002] With the development of Internet technology and computer technology, more and more data are stored in the hard disk, and companies with a large amount of user data are also stepping up to devote energy to the mining and reuse of these data, such as in the face image business , it is necessary to label a large number of face pictures, and obtain a data result set corresponding to one label for multiple snapshots of a person, so as to support the research and analysis of large data volumes, reduce the work pressure of front-line police officers, and further Improve the intelligence of the public security system. [0003] The traditional clustering algorithm belongs to an unsupervised learning algorithm in the field of machine learning, which c...

AI Technical Summary

Problems solved by technology

[0004] The embodiment of the present application provides a method, device, server and storage medium for real-time clustering of face images, so as to solve the problems of accuracy and real-time performance when clustering face data with a large amount of data, and improve the speed of clustering and the clustering results the accuracy of

Images