Suspect identification and positioning tracking system and method based on 5G wireless network

Abstract

The invention belongs to the technical field of wireless communication, and discloses a suspect identification and positioning tracking system and method based on a 5G wireless network. The overall structure of the suspect identification and positioning tracking system based on the 5G wireless network comprises a sensing layer, a transmission layer, a computing storage resource layer and an application layer. Based on a 5G wireless high-speed network, real-time return of monitoring video pictures is realized, and suspect recognition and positioning tracking are completed by combining face recognition and binocular vision ranging technologies; high-definition and controllable wireless binocular vision cameras are densely deployed in an urban area, the cameras upload video pictures to nearest mobile edge nodes through a 5G wireless high-speed network, and finally individual suspects are recognized, positioned and tracked. Experiments show that the 5G wireless network time delay providedby the invention is less than 35ms, the calculation time delay is 100ms, the total time delay is equal to 2*network time delay+calculation time delay, and the total time delay is less than 200ms.

Description

technical field [0001] The invention belongs to the technical field of wireless communication, and in particular relates to a system and method for identifying, locating and tracking a suspect based on a 5G wireless network. Background technique [0002] At present, in the field of urban security, high-definition cameras are being iteratively replaced, and face recognition technology is gradually being introduced to solve the problems of low image data quality and manual analysis of massive video source images. In terms of suspect identification and tracking, in a single face recognition technology, image processing technology and deep learning algorithms consume a huge amount of computing resources. In addition, the individual similarities of videos taken in large scenes are extremely high, making it difficult to identify suspects through video methods. It is difficult to solve the problem of real-time positioning, identification, and tracking of suspects by classifying, id...

AI Technical Summary

Problems solved by technology

[0005] (1) In the existing single face recognition technology, the image processing technology and deep learning algorithm consume a huge amount of computing resources. In addition, the individual similarity of the videos taken in large scenes is extremely high, so it is difficult to identify individuals through video methods. Classification identification, positioning and tracking, it is difficult to solve the problem of real-time positioning, identification and tracking of suspects

[0006] (2) There is still a delay problem in the transmission of existing high-definition video streams. Usually, the massive high-definition image data can only be manually analyzed afterwards, or automatically analyzed with the help of image recognition technology to find clues of suspects, which is difficult to solve the problem of urban management and suspects. Tracking strikes and many other issues

Images