Design And Implementation Of Intelligent Video Surveillance System Based On Face Recognition

With the rapid development of computer technology,microelectronics technology,digital image processing technology,digital video compression technology,network communication technology and neural network technology,the development trend of video surveillance systems is shifting to networked,intelligent,and mobile.The development of society has an increasing demand for unattended intelligent video surveillance systems.The design goal of this type of intelligent video surveillance system is to reduce the manual operation in the monitoring process and continuously reduce the surveillance system's The system can use operations between software to automatically complete the real-time detection and tracking operation processing for core targets or people in complex environments.In this paper,after researching the development status,future development trends and related technologies of video surveillance systems,this paper designs and implements an intelligent video surveillance system based on face recognition technology.Intelligent video surveillance systems include video acquisition,face detection,and face detection.Identification and intelligent early warning four modules.The video acquisition module uses the IP camera as the video source,and periodically captures the pictures of the IP camera through the RTSP protocol,and transmits the pictures to the face detection module.Face detection uses the MTCNN algorithm to locate the face and key points through a three-layer cascade of P-net,R-net,and O-net networks,completes facial image correction based on the key point positions,and then crops the face image And normalized to a standard face image with an image size of 112 × 112.The recognition module uses the Insightface algorithm to recognize the cropped standard face image.Based on the recognition result,the intelligent early warning module will use Twilio to make different warnings.If it is a stranger,it will send an alarm SMS to the manager and save the picture of the stranger.If it is a person in the database,record that the person has entered the monitoring area and send a reminder message to the management.At the end of this article,the overall function test and test result analysis of the intelligent video detection system are performed.The experimental results show that the intelligent video monitoring system can meet the monitoring needs in terms of recognition accuracy and early warning response speed,and the system can well complete the monitoring task.This system has certain practical value.