Face recognition security alarm system and device based on embedded artificial intelligence chip

Abstract

The invention belongs to the technical field of intelligent recognition, and discloses a face recognition security alarm system and device based on an embedded artificial intelligence chip. A video data acquisition module is used for acquiring video information. A video data transfer module is used for obtaining a frame image in the video information. A real-time detection module is used for the development board to carry out on-site face positioning and recognition. A judgment and alarm module is used for the local development board to obtain the recognition result and judge the alarm level,and if the alarm level is too high, alarm information is directly sent. A data unloading module is used for forwarding and storing data. A data acquisition module is used for sending videos to a mobile phone of a user so that the user can obtain and process alarm results in time. The system has the characteristics of capability of directly performing real-time face recognition and real-time monitoring alarm on the transmitted video, evidence saving and low cost, and a user can directly use a mobile phone APP to play and display a real-time picture.

Description

technical field [0001] The invention belongs to the technical field of intelligent recognition, and in particular relates to a face recognition security alarm system and equipment based on an embedded artificial intelligence chip. Background technique [0002] Currently, the closest prior art: [0003] Security issues have always been a prominent issue in life. Whether it is access control or monitoring methods, there are some unsolvable problems, such as: real-time alarm, evidence video preservation, machine active identification, etc. In today's big environment, there are many problems whether it is home security, store security or company security. Although there are endless security products on the market, the real-time monitoring and alarm methods are not ideal. Some can only do For monitoring, some need to touch the red line, and some are for access control, but these are not flexible enough, and it is difficult to achieve security protection when dealing with experie...

AI Technical Summary

Problems solved by technology

[0006] (1) Most of the existing technologies rely on third-party server resources for face recognition, which leads to the fact that the recognition process needs to go through a network channel, so there must be problems such as long delay and data loss, and the real-time performance cannot keep up.

[0007] (2) The existing technology cannot realize intelligent extraction and storage of intrusion evidence, but only the collected surveillance video, which makes it more troublesome to find evidence

[0008] (3) Most of the existing technologies are access control security systems, which cannot completely ensure that no one is in the area

[0009] (4) Existing technology cannot report to the police in time

[0010] (5) The cost of intelligent products with existing technologies is generally higher

[0012] The difficulty of the present invention lies in how to replace the traditional system method of first extracting data for server identification and then feeding it back to the mobile terminal to obtain identification results; how to change the traditional alarm judgment level is single, the alarm effect is not ideal, etc.

Images