An Intelligent Monitoring and Security System built using OpenCV, YOLO, face_recognition, PySide6.
- Data Collection and Preprocessing: Utilizes the OpenCV library to capture and preprocess data from the camera feed.
- Face Detection: Employs the YOLOv8 model to detect faces and other objects in the real-time video stream.
- Face Recognition and Analysis: Integrates the face_recognition (dlib) library to detect, recognize, and analyze facial features. This includes silent liveness detection and facial feature analysis using the APIs provided by iFLYTEK and Tencent Cloud.
- Behavior Detection: Monitors and detects the behavior of individuals in the camera feed.
- Object Detection and Recognition: Leverages the YOLOv8 model to detect and recognize various objects in the video stream.
- Security Alerts: Sends timely email alerts using the smtplib and email libraries when suspicious activities or unknown faces are detected.
It is recommended to configure the required dependencies in a virtual Python environment
pip install -r requirements.txtThe version described in requirements.txt is exported by the pipreq command. In theory, similar versions can also run smoothly.
After preparing the dependent environment, you need to fill in example-config.json in resource directory and rename it to config.json before it can run normally.
The face login function relies on iFlytek's silent liveness detection API, the face feature analysis relies on Tencent SDK, and the security alert email function requires the SMTP service provided by the email service provider. If you need to use all the functions of the system, please fill in all items in the config.json file.