Moving track obtaining method and device, storage medium and terminal

Abstract

The embodiment of the invention discloses a moving track obtaining method and device, a storage medium and a terminal. The method comprises the following steps of obtaining a target image generated for a shot area at a target moment in a selected time period; performing image recognition processing on the target image to obtain a face image set of the target image; recording current position information of each face image in the face image set on the target image at the target moment; and based on the current position information and according to a time sequence, outputting a moving track setof the face image set in the set time period. By adopting the method and the device, the monitoring of the user is realized based on the human face moving track, and the variability, diversity and instability of human body behaviors can be avoided, so that the calculation amount of user monitoring behaviors can be reduced.

Description

technical field [0001] The present invention relates to the field of computer technology, and in particular to a method for obtaining a moving trajectory, a device, a storage medium, and a terminal thereof. Background technique [0002] With the development of security monitoring systems and the trend of digital, networked, and intelligent monitoring is moving forward unswervingly, in the case of a large number of front-end cameras, complex business structures, and high levels of management and integration, the video surveillance management platform is also It has attracted more and more people's attention, and is gradually applied in important security business systems. [0003] The existing security monitoring system is mainly based on the recognition of human behavior, but most of the recognition methods are aimed at the video classification problem of only one behavior in the surveillance video, but the behavior of the monitored person in the real scene is variable and u...

AI Technical Summary

Problems solved by technology

[0003] The existing security monitoring system is mainly based on the recognition of human behavior, but most of the recognition methods are aimed at the video classification problem of only one behavior in the surveillance video, but the behavior of the monitored person in the real scene is variable and unstable Therefore, the existing security monitoring system has the problem of large amount of monitoring behavior calculations.

Images