rtgamerz009/attendance-verifier

⚡ Overview: The Proxy Problem

Traditional attendance systems rely on vulnerable processes—roll calls, ID cards, or simple PINs. These are easily exploited, leading to proxy attendance, human error, and massive time consumption. Attendance Verifier eliminates these vulnerabilities using multi-layer biometric validation.

🔥 Key Features

🏗️ System Architecture

graph TD;
    A[Client Device / WebCam] -->|Live Video Feed| B(Processing Layer)
    
    subgraph Verification Engine
    B --> C{Face Detection}
    C -->|Detected| D[Liveness Check Challenge]
    D -->|Passed| E[Anti-Spoof Analysis]
    E -->|Cleared| F[Identity Matching]
    end
    
    subgraph Data Layer
    F -->|Match Confirmed| G[(Secure Database)]
    E -->|Spoof Detected| H[Threat Log]
    end
    
    G --> I[Admin Dashboard]
    H --> I

🔄 Working Flow

1. Capture ➔ User positions face within the camera reticle. 2. Challenge ➔ System prompts a random liveness challenge (e.g., subtle head movement). 3. Analysis ➔ Engine checks for depth, edge anomalies, and reflections to prevent screen spoofing. 4. Record ➔ Upon >95% confidence score, attendance is securely recorded. Multiple failures trigger an automatic ban.

🛠️ Technology Stack

🧪 Core Intelligence & Logic

The system utilizes a staged confidence-scoring model to ensure high accuracy and security:

• Bounding Box Tracking: Initial logic ensures a face is centered, properly lit, and clear of occlusions.
• Liveness Validation: Implements randomized physical challenges that a static image cannot replicate.
• Spoof Detection Rules: Analyzes the video feed for flat edges, screen refresh rates (Moiré patterns), and unnatural lighting reflections to block digital proxy attempts.
• Three-Strike Protocol: If a user fails the anti-spoofing checks three times consecutively, the system locks the session and dispatches an alert to administrators.

📊 Impact

• 100% Proxy Elimination: Ensures attendance records represent actual physical presence.
• Frictionless Experience: Reduces the time to mark attendance from minutes to seconds.
• Actionable Analytics: Provides administrators with real-time insight into attendance trends and security anomalies.

🚀 Future Scope

• Mobile App Integration: Native iOS/Android clients utilizing internal FaceID/hardware sensors.
• Cloud Edge Processing: Offloading heavy verification logic to edge servers to support lower-end devices.
• Advanced AI Recognition: Integration with Python-based backend ML models (OpenCV/Dlib) for enterprise-scale face matching.

---