Video image clustering method and system

Abstract

The invention discloses a video image clustering method which includes that a plurality of video images in the video frames are acquired; the similarities of the acquired video images in the video frames are compared; the video image with the first similarity value less than or equal to the first threshold is divided into different video image sub-clusters; the video image with the second similarity value of the first video image less than or equal to the second threshold is used as the video image typical sample, so as to generate the video image typical sample set. The invention embodiment is implemented to realize the rapid indexing of the video images in the monitoring videos.

Description

technical field [0001] The present invention relates to the technical field of image processing, in particular to a video image clustering method and system. Background technique [0002] With the development of society and the advancement of science and technology, the people's demand for safety is also getting higher and higher. Especially in the past two decades, security threats have occurred frequently: the 911 incident in the United States in 2001, the Russian terrorist bombing and hijacking in 2004, the Spanish railway bombing in 2004, and the British serial bombing in 2005. The occurrence of these incidents has made the governments of various countries attach great importance to cross-regional and cross-country social security prevention and anti-terrorism work. For our country, although no large-scale social security incidents have occurred, it is also facing the threat of terrorist activities. And with the increase of social mobility, the potential possibility of...

AI Technical Summary

Problems solved by technology

In the process of research and practice of the prior art, the inventor of the present invention found that the text information and voice information contained in the current surveillance video have little correlation with the image information, and it is not suitable for integrating text and human information contained in the surveillance video. Face image information for face image indexing

Images