SecureTime Pro

Employee time tracking system using facial recognition. Look at the camera for 2 seconds to automatically clock in/out.

What it does

• Face recognition identifies employees automatically
• Auto clock in/out - no buttons to press
• Liveness detection prevents photo spoofing
• Time tracking calculates work hours
• Web interface for management and reports

Installation

Requirements

• Python 3.8+
• Webcam
• Modern browser (Chrome recommended)

Setup

1. Clone repository

   git clone https://github.com/yourusername/securetime-pro.git
   cd securetime-pro

2. Install dependencies

   pip install -r requirements.txt

3. Run application

   python face_recognition_webui.py

4. Open browser

   http://localhost:5000
   

How to use

Add employees

1. Go to "Add Employee"
2. Enter name
3. Take photos from different angles (front, left, right)
4. System trains recognition model

Time tracking

1. Go to "Time Clock"
2. Click "Start Camera"
3. Look at camera for 2 seconds
4. System automatically:
   • Clocks you IN if you're not working
   • Clocks you OUT if you're currently working
5. Camera stops automatically

View reports

1. Go to "Reports"
2. Select date range
3. Generate time reports

How it works

Face Recognition

• Uses dlib and face_recognition libraries
• Adaptive learning improves accuracy over time
• Confidence threshold prevents false matches

Liveness Detection

• Detects real faces vs photos
• Analyzes face movement and quality
• Prevents spoofing with printed photos

Database

• SQLite stores employee data and time records
• Each work session = one record with clock_in/clock_out times
• Automatic hours calculation

Auto Clock Logic

Current Status → Action
Not clocked in → Clock IN
Clocked in → Clock OUT
Previously clocked out → Clock IN (new session)

File Structure

SecureTime-Pro/
├── face_recognition_webui.py          # Main Flask application
├── maml_face_recognition.py           # Face recognition engine  
├── anti_spoofing_system.py            # Liveness detection
├── improved_few_shot.py               # Learning algorithms
├── face_recognition_system.py         # Base recognition system
├── templates/                         # HTML templates
│   ├── base.html                      # Main layout
│   ├── timeclock.html                 # Time clock interface
│   ├── index.html                     # Dashboard
│   ├── employees.html                 # Employee list
│   ├── add_employee.html              # Add new employee
│   └── reports.html                   # Time reports
├── requirements.txt                   # Python dependencies
└── README.md                          # This file

Configuration

Edit these values in face_recognition_webui.py:

RECOGNITION_THRESHOLD = 0.6          # Minimum confidence to recognize (60%)
CLOCK_ACTION_THRESHOLD = 60          # Minimum confidence to clock in/out (60%)
AUTO_CLOCK_DELAY = 2                 # Seconds to hold steady (2 seconds)

API Endpoints

REST API

• GET /api/status - System status
• GET /api/employees - List employees
• GET /api/time_records - Get time records
• DELETE /api/delete_employee/<name> - Remove employee

WebSocket Events

• start_session - Start camera
• video_frame - Process video frame
• auto_clock_result - Clock action result

Troubleshooting

Camera not working:

• Check browser permissions
• Close other apps using camera
• Try different browser

Recognition not accurate:

• Add more training photos
• Ensure good lighting
• Check camera focus

Database errors:

• Check file permissions
• Restart application

Development

Debug Mode

Add debug routes to see employee status:

@app.route('/api/debug/<employee_name>')
def debug_employee(employee_name):
    # Shows current status and all records

Database Schema

employees table:

• id, name, department, created_at

time_records table:

• id, employee_name, clock_in, clock_out, hours_worked, date, confidence

Security Notes

• Employee photos stored locally only
• Database contains sensitive time data
• Use proper access controls in production
• Consider encryption for sensitive environments

Technology Stack

• Backend: Flask, Flask-SocketIO
• Face Recognition: dlib, face_recognition, OpenCV
• Database: SQLite
• Frontend: HTML5, JavaScript, WebSocket
• ML: scikit-learn, numpy