Machine Vision Security System
Project information
- Category: Computer Vision Vision
- Project date: July 2023 - October 2023
- Key Features: Face Detection, Secure Mode, Fire and Animal Detection, Unknown Person Identification
- Technologies Used: Python, OpenCV, TensorFlow, Flask, Firebase
- Team Members: E S Dhanush (Team Lead), Ravivarman K, Sanjay S, Thamilthennarasu, Shyamsunder S
- Project Link : Github Link
The Machine Vision Security System is an advanced security solution utilizing machine learning and computer vision to provide real-time surveillance. It can detect and recognize individuals, identify potential threats such as fire and animals, and trigger alerts via app notifications, emails, and alarms.
Unveiling the Future of Security with Machine Vision Technology
In a world where security is paramount, traditional surveillance systems are becoming outdated. Enter Machine Vision Security System—an advanced solution that leverages machine learning and computer vision to redefine how we monitor and protect our environments. From facial recognition to real-time alerts, this system is designed to offer a comprehensive approach to modern security.
Core Features of the Machine Vision Security System
1. Real-Time Recognition of Known Faces
Using pre-trained models, the Machine Vision Security System can instantly identify and label known faces. The system scans live camera feeds, recognizing familiar individuals from a pre-defined database and labeling them in real time. This functionality is particularly useful in environments where access control and familiarity matter.
Figure 1: Identifying and labeling a known individual in real-time.
2. Detecting and Labeling Unknown Persons
The system takes security a step further by identifying unknown individuals. Any face that does not match the known database is labeled as "Unknown Person." This immediate distinction ensures that strangers are easily detected, enabling swift responses in situations requiring attention.
Figure 1: Identifying and labeling a known individual in real-time.
3. Search for Last Detected Photos of Known Individuals
Tracking individuals is easier with the system’s search functionality. Users can enter a person’s name and retrieve the last five detected photos. This feature enables historical tracking of movements and offers a visual log of where and when a specific individual was last seen.
Figure 1: Identifying and labeling a known individual in real-time.
4. Image-Based Search for Unknown Persons
Ever wanted to track an unknown individual’s movements? With the image-based search, users can upload a photo of an unknown person, and the system will return the locations and times of previous detections. This allows for effective surveillance over time, providing visibility over potential security risks.
Figure 4: Upload an image to trace the last detected appearances of an unknown person.
5. Secure Mode: Real-Time Alerts for Unfamiliar Faces
Secure Mode is a critical feature for high-alert situations. When enabled, the system continuously monitors for unknown individuals. If an unfamiliar face is detected, it triggers an immediate email notification to the user, complete with a photo of the person. This ensures users are promptly alerted to any suspicious activities, even when away from the premises.
Figure 5: The system sends an alert with an image of an unknown person during secure mode