Stranger monitoring method, device, system and storage medium

Abstract

The present invention provides a stranger monitoring method, device, system and storage medium. The method includes: acquiring a stranger base database, which has multiple stranger files; acquiring passerby face images, and extracting passerby face features Compare with the file features of multiple stranger files; establish a correlation between the passer-by face image and the stranger file corresponding to the file feature with the highest similarity to the passer-by face feature and higher than the threshold value, and The face images of passers-by associated with the same stranger file are regarded as a set corresponding to the stranger file; according to the similarity between the set and the stranger file, it is judged whether the set compares with the stranger file, when the set When comparing two or more stranger files, the two or more stranger files are combined. The present invention can use the collection of face images of passers-by to organize the stranger files in the stranger bottom database, improve the quality of the stranger files, and ensure the accuracy of face recognition.

Description

technical field [0001] The present invention relates to the technical field of face recognition, and more particularly relates to a stranger monitoring method, device, system and storage medium. Background technique [0002] With the continuous improvement of people's living standards and quality of life, the safety needs of housing have gradually become the focus of widespread attention. The current community security management usually requires the frequent alarm function of strangers, so as to realize community anti-spotting and identify abnormal persons in the community. At present, the mainstream method in the market is to compare the face images of passers-by with the face images in the blacklist, whitelist and other base databases. The ones in the file are added to the stranger's file, resulting in an excessive amount of data in the stranger's file. Moreover, there is no quality judgment rule set in the existing scheme, adding too many passer-by face images with poo...

AI Technical Summary

Problems solved by technology

At present, the mainstream method in the market is to compare the face images of passers-by with the face images in the blacklist, whitelist and other base databases. The stranger files are added, resulting in an excessive amount of data in the stranger files.

Moreover, there is no quality judgment rule set in the existing scheme, adding too many passer-by face images with poor quality to stranger files will lead to a decrease in comparison accuracy, and further lead to miscellaneous categories (face images that do not belong to the same person into the same file) or file splitting (face images belonging to the same person are divided into two files) and other issues

Images