Clean deployment

README.md

@@ -9,629 +9,3 @@ short_description: AI-powered video surveillance system
 ---
 
 Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
-
-# 🎯 NETRA - AI Video Surveillance Web Application
-
-<div align="center">
-
-### ⚡ **AI-Powered Video Surveillance & Real-Time Threat Detection**
-
-![Status](https://img.shields.io/badge/Status-Active-brightgreen?style=flat-square)
-![Python](https://img.shields.io/badge/Python-3.8%2B-blue?style=flat-square)
-![Flask](https://img.shields.io/badge/Flask-Web-orange?style=flat-square)
-![AI Models](https://img.shields.io/badge/AI%20Models-8%2F8-success?style=flat-square)
-
-</div>
-
----
-
-![1779175689811](image/README/1779175689811.png)
-
-## 🚀 Overview
-
-**NETRA** is a comprehensive web-based video surveillance system powered by advanced AI models that provides:
-
-✅ **User Authentication** - Secure login/registration with SQLite database  
-✅ **Live Camera Monitoring** - Real-time video analysis with AI detection  
-✅ **Video Analysis** - Upload and analyze pre-recorded videos  
-✅ **Multi-Model Support** - 8 integrated AI models for comprehensive detection  
-✅ **Extensible Architecture** - Easy to add new AI models  
-✅ **100% Portable** - Works on any device, any path, any OS
-
-## 📋 Key Features
-
-<table>
-<tr>
-<td width="50%">
-
-### 🔐 **Authentication System**
-
-```
-✓ User registration & login
-✓ Secure password hashing
-✓ Session management
-✓ SQLite database
-```
-
-</td>
-<td width="50%">
-
-### 🎥 **Live Camera Monitoring**
-
-```
-✓ Real-time webcam feed
-✓ YOLOv8 object detection
-✓ Violence recognition
-✓ Instant alerts
-✓ Live statistics
-```
-
-</td>
-</tr>
-<tr>
-<td>
-
-### 📹 **Video Analysis**
-
-```
-✓ Upload videos (500MB max)
-✓ Batch frame processing
-✓ Detailed reports
-✓ Alert classification
-```
-
-</td>
-<td>
-
-### 🤖 **8 AI Models**
-
-```
-✓ Violence detection
-✓ Weapon detection
-✓ Pose analysis
-✓ Object detection
-✓ Activity recognition
-```
-
-</td>
-</tr>
-</table>
-
-## � AI Models
-
-### Model Repository
-
-All AI models are hosted on **Hugging Face Hub** for free:
-
-🔗 **[NETRA-Models on Hugging Face](https://huggingface.co/itsluckysharma01/NETRA-Models)**
-
-### Models Available
-
-- 🎬 **Activity Recognition**: Violence/fight detection model (`violence_model.h5`)
-- 🔫 **Weapon Detection**: Detects guns, knives, and weapons (`best.pt`)
-- 👤 **Pose Detection**: Human body pose and movement analysis (`yolo11n-pose.pt`)
-- 📊 **Object Detection**: General object/person detection (`yolov8n.pt`)
-- 📈 **Analysis Models**: Multiple CNN models for advanced behavior analysis
-
-### Auto-Download on Startup
-
-Models automatically download and cache locally on first run:
-
-```bash
-python app.py  # Models download automatically
-```
-
-### Storage Details
-
-- **Storage Location**: Hugging Face Hub (free)
-- **Total Size**: ~800MB
-- **First Run Download**: ~5-10 minutes (depending on internet speed)
-- **Subsequent Runs**: Instant (uses cached models)
-- **Local Cache**: `ai_models/` folder
-
-## �🛠️ Installation
-
-### Prerequisites
-
-- Python 3.8 or higher
-- Webcam (for live monitoring)
-- Windows/Linux/Mac OS
-
-### Step 1: Clone or Navigate to Project
-
-```bash
-cd f:/PROGRAMMING/MY PROJECTS/ALL_PROJECTS/Collabration_Projects/NETRA/NETRA_Project/webapp
-```
-
-### Step 2: Create Virtual Environment
-
-```bash
-# Step 1: Create virtual environment
-python -m venv venv
-
-# Step 2: Activate it
-# Windows:
-venv\Scripts\activate
-# Linux/Mac:
-source venv/bin/activate
-
-# Step 3: Install dependencies
-pip install -r requirements.txt
-
-# Step 4: Verify setup
-python verify_setup.py
-
-# Step 5: Run application
-python app.py
-```
-
-```bash
-python -m venv venv
-
-# Windows
-venv\Scripts\activate
-
-# Linux/Mac
-source venv/bin/activate
-```
-
-### Step 3: Install Dependencies
-
-```bash
-pip install -r requirements.txt
-```
-
-### Step 4: Setup Directory Structure
-
-The application expects the following structure:
-
-```
-NETRA_Project/
-├── webapp/
-│   ├── app.py
-│   ├── requirements.txt
-│   ├── static/
-│   │   ├── css/
-│   │   ├── js/
-│   │   └── images/
-│   ├── templates/
-│   └── uploads/
-├── NETRA/
-│   ├── violence_detector.py
-│   ├── yolo_detector.py
-│   ├── yolov8n.pt
-│   └── model/
-└── ai_models/
-    ├── activity_recognition/
-    │   └── violence_model.h5
-    └── object_detection/
-        └── yolov8n.pt
-```
-
-## 🚀 Running the Application
-
-### ⚡ Quick Start (3 Steps)
-
-#### Step 1️⃣ Navigate to Project
-
-```bash
-cd NETRA-Project/webapp
-```
-
-#### Step 2️⃣ Activate Virtual Environment
-
-```bash
-# Windows
-venv\Scripts\activate
-
-# Linux/Mac
-source venv/bin/activate
-```
-
-#### Step 3️⃣ Start the Server
-
-```bash
-python app.py
-```
-
-### 🎉 Success! You'll See
-
-```
-╔═════════════���════════════════════════════════════════════════════╗
-║              NETRA - AI Video Surveillance System                 ║
-╚══════════════════════════════════════════════════════════════════╝
-
-📊 Available Models: [violence, yolo, weapon, pose, ...]
-
-🌐 Access Application:
-   → http://localhost:5001
-   → http://127.0.0.1:5001
-
-⚙️  Configuration:
-   → Debug Mode: ON
-   → Port: 5001
-   → Host: 0.0.0.0
-
-✅ Server is RUNNING
-   Press CTRL+C to stop
-```
-
-### 🌐 Open in Browser
-
-| Step | Action                                      |
-| ---- | ------------------------------------------- |
-| 1️⃣   | Open **Chrome**, **Firefox**, or **Safari** |
-| 2️⃣   | Navigate to **http://localhost:5001**       |
-| 3️⃣   | Click **"Sign Up"** to create account       |
-| 4️⃣   | Enter username, email, password (6+ chars)  |
-| 5️⃣   | Click **"Login"** with your credentials     |
-| 6️⃣   | Access **Dashboard** and start monitoring!  |
-
-## 📖 Usage Guide - Interactive Walkthrough
-
-### 🔐 Step 1: Registration
-
-```
-Dashboard → Sign Up Tab
-                ↓
-    Enter Username (any name)
-    Enter Email (valid email)
-    Enter Password (6+ characters)
-                ↓
-    Click "Sign Up"
-                ↓
-    ✅ Redirected to Login
-```
-
-**Requirements:**
-
-- ✓ Username: Any name (5+ chars recommended)
-- ✓ Email: Valid email address
-- ✓ Password: Minimum 6 characters
-
-### 🔑 Step 2: Login
-
-```
-Login Page
-    ↓
-Enter Credentials
-    ↓
-Click "Login"
-    ↓
-✅ Dashboard Loaded
-```
-
-### 📊 Step 3: Dashboard Overview
-
-After login, you'll see:
-
-| Feature            | Description                          | Icon |
-| ------------------ | ------------------------------------ | ---- |
-| **Live Camera**    | Real-time monitoring from webcam     | 🎥   |
-| **Video Analysis** | Upload & analyze pre-recorded videos | 📹   |
-| **Results**        | View detection results & alerts      | 📊   |
-
-### 🎥 Step 4: Live Camera Monitoring
-
-```
-Dashboard
-    ↓
-Click "Start Live Monitoring"
-    ↓
-Allow Camera Permission (browser prompt)
-    ↓
-Real-Time Detection Begins
-    ├── Object Detection (YOLOv8)
-    ├── Violence Detection
-    ├── Weapon Detection
-    └── Pose Analysis
-    ↓
-View Results in Right Panel
-    ├── Detections count
-    ├── Alerts
-    └── Confidence scores
-    ↓
-Click "Stop Camera" when done
-```
-
-### 📹 Step 5: Video Analysis
-
-```
-Dashboard
-    ↓
-Click "Analyze Video"
-    ↓
-Upload Video File (drag & drop or click)
-    ├── Supported formats: MP4, AVI, MOV, MKV
-    └── Max size: 500MB
-    ↓
-Click "Analyze Video"
-    ↓
-⏳ Processing (shows progress bar)
-    ↓
-✅ Results Ready
-    ├── Total Frames Processed
-    ├── Detection by Category
-    ├── Alerts & Severity
-    └── Summary Statistics
-```
-
-### 📈 Step 6: Understanding Results
-
-**Detection Results Show:**
-
-- 🎯 **Objects Detected**: People, vehicles, weapons
-- ⚠️ **Alerts**: Unusual activities flagged
-- 📊 **Statistics**: Frame analysis summary
-- 🔔 **Severity Levels**: High, Medium, Low
-
-## 🔧 Adding New Models
-
-The application is designed to be extensible. To add a new AI model:
-
-### 1. Create Your Model Class
-
-```python
-# my_custom_model.py
-class CustomDetector:
-    def __init__(self, model_path):
-        # Load your model
-        pass
-
-    def detect(self, frame):
-        # Process frame and return detections
-        pass
-```
-
-### 2. Register in ModelManager
-
-Edit `app.py` and add to the `ModelManager.load_models()` method:
-
-```python
-def load_models(self):
-    # Existing models...
-
-    # Add your custom model
-    try:
-        from my_custom_model import CustomDetector
-        custom_model_path = Path(__file__).parent.parent / "models" / "custom_model.h5"
-        if custom_model_path.exists():
-            self.models['custom'] = CustomDetector(str(custom_model_path))
-            print("✓ Custom model loaded")
-    except Exception as e:
-        print(f"✗ Error loading custom model: {e}")
-```
-
-### 3. Integrate in VideoProcessor
-
-Add detection logic in `VideoProcessor.process_frame()`:
-
-```python
-# Custom Model Detection
-if 'custom' in model_manager.models:
-    try:
-        custom_model = model_manager.get_model('custom')
-        custom_results = custom_model.detect(frame)
-        # Process results...
-    except Exception as e:
-        print(f"Custom detection error: {e}")
-```
-
-## 📊 Database Schema
-
-### Users Table
-
-```sql
-CREATE TABLE user (
-    id INTEGER PRIMARY KEY,
-    username VARCHAR(80) UNIQUE NOT NULL,
-    email VARCHAR(120) UNIQUE NOT NULL,
-    password_hash VARCHAR(200) NOT NULL,
-    created_at DATETIME DEFAULT CURRENT_TIMESTAMP
-);
-```
-
-### Analysis History Table
-
-```sql
-CREATE TABLE analysis_history (
-    id INTEGER PRIMARY KEY,
-    user_id INTEGER NOT NULL,
-    analysis_type VARCHAR(50) NOT NULL,
-    filename VARCHAR(200),
-    detections TEXT,
-    alert_count INTEGER DEFAULT 0,
-    created_at DATETIME DEFAULT CURRENT_TIMESTAMP,
-    FOREIGN KEY (user_id) REFERENCES user(id)
-);
-```
-
-## 🎨 Customization
-
-### Logo
-
-Replace the logo by adding your image file:
-
-```
-webapp/static/images/logo.png
-```
-
-### Colors and Styling
-
-Edit `webapp/static/css/style.css` to customize:
-
-- Primary color: `--primary-color`
-- Background gradients
-- Card styles
-- Animations
-
-### Alert Thresholds
-
-Modify detection thresholds in `app.py`:
-
-```python
-# Example: Adjust confidence threshold
-if det.confidence > 0.75:  # Changed from default
-    # Process detection
-```
-
-## 🔒 Security Notes
-
-### Current Implementation
-
-- Password hashing using Werkzeug
-- Session-based authentication
-- CSRF protection recommended for production
-
-### Production Recommendations
-
-1. Change the SECRET_KEY in app.py
-2. Use environment variables for sensitive data
-3. Enable HTTPS
-4. Add rate limiting
-5. Implement CSRF tokens
-6. Use stronger session management
-
-## 🐛 Troubleshooting
-
-### Camera Not Working
-
-- Check if camera is connected
-- Grant browser camera permissions
-- Ensure no other application is using the camera
-- Try restarting the application
-
-### Models Not Loading
-
-- Verify model files exist in correct paths
-- Check file permissions
-- Review console output for error messages
-- Ensure all dependencies are installed
-
-### Upload Fails
-
-- Check file size (max 500MB)
-- Verify file format (MP4, AVI, MOV, MKV)
-- Ensure `uploads/` folder exists
-- Check disk space
-
-### Database Errors
-
-- Delete `netra.db` and restart to recreate
-- Check file permissions
-- Ensure SQLite is properly installed
-
-## 📝 API Endpoints
-
-| Endpoint          | Method | Description               |
-| ----------------- | ------ | ------------------------- |
-| `/`               | GET    | Landing page              |
-| `/register`       | POST   | User registration         |
-| `/login`          | POST   | User login                |
-| `/logout`         | GET    | User logout               |
-| `/dashboard`      | GET    | Main dashboard            |
-| `/live-camera`    | GET    | Live camera page          |
-| `/video-analysis` | GET    | Video analysis page       |
-| `/camera_feed`    | GET    | Camera stream             |
-| `/upload_video`   | POST   | Upload video for analysis |
-| `/api/models`     | GET    | List available models     |
-| `/stop_camera`    | POST   | Stop camera stream        |
-
-## 🤝 Contributing
-
-To contribute to this project:
-
-1. Fork the repository
-2. Create a feature branch
-3. Make your changes
-4. Test thoroughly
-5. Submit a pull request
-
-## 📄 License
-
-See [LICENSE](LICENSE) file for details.
-
-## 👥 Support & Help
-
-### 🆘 Getting Help
-
-| Problem                      | Solution                                                                   |
-| ---------------------------- | -------------------------------------------------------------------------- |
-| **Server won't start**       | Check Python version (3.8+), virtual env activated, dependencies installed |
-| **Models not downloading**   | Check internet connection, HuggingFace Hub access                          |
-| **Camera not working**       | Grant browser camera permissions, check webcam compatibility               |
-| **Port 5001 already in use** | Change port in app.py or close other applications                          |
-
-### 🐛 Reporting Issues
-
-1. **Check existing issues** on GitHub
-2. **Provide details:**
-   - Your OS (Windows/Linux/Mac)
-   - Python version: `python --version`
-   - Error message from terminal
-   - Steps to reproduce
-
-3. **Create issue** with clear title and description
-
-### 📖 Documentation
-
-- 📁 [PORTABLE_SETUP.md](PORTABLE_SETUP.md) - How portability works
-- 🏗️ [ARCHITECTURE.md](docs/ARCHITECTURE.md) - System architecture
-- 🚀 [Quick Start Guide](docs/QUICK_START_BOUNDING_BOXES.md)
-
-## 🎯 Future Enhancements
-
-**Planned Features:**
-
-| Feature                      | Status     | Priority |
-| ---------------------------- | ---------- | -------- |
-| Multi-camera support         | 🔄 Planned | High     |
-| Cloud storage integration    | 🔄 Planned | High     |
-| Email/SMS alerts             | 🔄 Planned | Medium   |
-| Advanced analytics dashboard | 🔄 Planned | Medium   |
-| Mobile app (iOS/Android)     | 📋 Backlog | Low      |
-| Face recognition             | 📋 Backlog | Low      |
-| License plate detection      | 📋 Backlog | Low      |
-| PDF report export            | 📋 Backlog | Low      |
-| Scheduled recording          | 📋 Backlog | Low      |
-| Motion detection zones       | 📋 Backlog | Low      |
-
----
-
-## 🔗 Quick Links
-
-<div align="center">
-
-| 🤖                                                                | 📚                     | 💬                     | 🎯                               |
-| ----------------------------------------------------------------- | ---------------------- | ---------------------- | -------------------------------- |
-| [HF Models](https://huggingface.co/itsluckysharma01/NETRA-Models) | [Documentation](docs/) | [Issues](../../issues) | [Roadmap](#-future-enhancements) |
-
-</div>
-
----
-
-## 📊 Project Statistics
-
-```
-📁 Files:          50+
-📦 Dependencies:   15+
-🤖 AI Models:      8
-💾 Database:       SQLite
-📊 Framework:      Flask
-🐍 Language:       Python 3.8+
-```
-
----
-
-<div align="center">
-
-### ⭐ If you find NETRA useful, please give it a star!
-
-**NETRA** - AI-Powered Video Surveillance System  
-**Version:** 1.0.0  
-**Status:** 🟢 Active & Maintained  
-**Last Updated:** 2026
-
-Made with ❤️ for security & surveillance
-
-</div>