Data incremental clustering method, system and device based on reduction, and medium

Abstract

The invention belongs to the technical field of image data processing, and particularly relates to a reduction-based data incremental clustering method and system, a medium and a device. The invention aims to solve the problem of low efficiency of image incremental clustering caused by high requirement on stand-alone capacity and long data processing time when the steps of searching and traversing are realized when the image data scale is very large. In order to achieve the purpose, in the process of clustering the newly-added data and the historical data, matching or clustering is carried out only through a reduction new cluster and a reduction old cluster instead of directly clustering all the newly-added data and the historical data. Therefore, reduction processing is carried out on the respective clusters before clustering of the newly added data and the historical data, the calculation amount during clustering of the newly added data and the historical data is reduced, the requirement for a running memory is lowered, and the time cost and the economic cost of clustering are saved.

Description

technical field [0001] The invention belongs to the technical field of image data processing, and in particular relates to a reduction-based incremental data clustering method, system, medium and device. Background technique [0002] The number of face images collected by personnel management systems, video surveillance systems, etc. generally increases exponentially, forming a massive face database. In order to realize the management, retrieval and query of face images, it is often necessary to cluster the face images. That is, when a newly added face image is collected, it is necessary to perform feature recognition on the newly added face image, and then classify and store it with the old face image belonging to the same category. This process involves Image Incremental Clustering Techniques. [0003] In the existing image incremental clustering technology, the connection transfer clustering algorithm is mainly used, such as DBSCAN algorithm and so on. In this type of ...

AI Technical Summary

Problems solved by technology

[0005] In order to solve or at least partially solve: when the scale of image data is very large, there are high requirements for stand-alone capacity and long data processing time when implementing the search and traversal steps, resulting in low efficiency of image incremental clustering

Images