FreshX is a web application that detects and classifies fruit quality (Fresh vs. Rotten) in real-time. It uses YOLOv8 Object Detection to localize and identify multiple fruits within complex environments, providing confidence scores and detailed analytics.
- 🍎 Multi-Fruit Detection: Detects and displays ALL fruits in an image with individual confidence scores
- 📷 Live Camera with Auto-Scan: Real-time detection from webcam with automatic 1-second interval scanning
- 🎯 Best Result Capture: Auto-Scan saves only the highest confidence result to history
- ⚡ Realtime Capture: Analyzes camera feed every 1 second automatically
- 📊 Live Feedback: Shows capture count and best confidence in real-time
- 📊 Smart Analytics Dashboard: Pie charts and trend graphs of detection history
- 📄 PDF Export: Generate professional reports with FreshX branding, summary statistics, and detection table
- 📥 CSV Export: Download properly formatted data with dd/mm/yy timestamps
- 🔍 Advanced Filtering: Filter by date range, fruit type, and freshness status
- 📝 Detection Notes: Add custom notes (e.g., "Warehouse A, Batch #102") to each scan
- ☁️ Cloud Sync: Automatic save to MongoDB with device-based history
- 🗑️ History Management: View, delete individual items, or clear all history
| Technology | Purpose |
|---|---|
| React 18 (Vite) | UI Framework |
| Tailwind CSS | Styling |
| Recharts | Charts & Analytics |
| Lucide React | Icons |
| jsPDF | PDF Generation |
| Technology | Purpose |
|---|---|
| Flask (Python) | REST API |
| Ultralytics YOLOv8 | AI Object Detection |
| Pillow / NumPy | Image Processing |
| PyMongo | Database Driver |
| Technology | Purpose |
|---|---|
| MongoDB | Data Storage |
flowchart LR
User["User Device"]
Frontend["Frontend (React + Vite)"]
Backend["Backend (Flask API)"]
Model["YOLOv8 Model (.pt)"]
DB["MongoDB"]
User -- "HTTPS" --> Frontend
Frontend -- "REST API (Images)" --> Backend
Backend -- "Object Detection" --> Model
Backend -- "Read/Write" --> DB
- Python 3.10+
- Node.js 20+
- MongoDB (local or Atlas)
git clone https://github.com/Joeliazeers/FreshX.git
cd FreshXcd backend
# Create virtual environment
python -m venv venv
# Activate environment
# Windows:
venv\Scripts\activate
# Mac/Linux:
source venv/bin/activate
# Install dependencies
pip install -r requirements.txt
# Set MongoDB connection (replace with your URI)
# Windows PowerShell:
$env:MONGO_URI="mongodb://localhost:27017/"
# Mac/Linux:
export MONGO_URI="mongodb://localhost:27017/"
# Run the server
python app.pycd frontend
# Install dependencies
npm install
# Create environment file
echo "VITE_API_URL=http://localhost:5000" > .env.local
# Run development server
npm run dev
| Method | Endpoint | Description |
|---|---|---|
| POST | /predict |
Analyze image. Use save=false for temp captures, save=true to persist to history. |
| GET | /history |
Fetch all past predictions for current device. |
| DELETE | /history/<id> |
Delete a single history record by ID. |
| DELETE | /history |
Clear entire history for current device. |
FormData:
- file: Image file (required)
- save: "true" or "false" (default: "true")
- notes: Optional notes string
{
"label": "Fresh Apple",
"confidence": 95.2,
"is_fresh": true,
"model_used": "YOLOv8",
"heatmap_b64": "...",
"detections": [
{"label": "Fresh Apple", "confidence": 95.2, "is_fresh": true, "bbox": [x1, y1, x2, y2]},
{"label": "Rotten Banana", "confidence": 87.5, "is_fresh": false, "bbox": [x1, y1, x2, y2]}
],
"detection_count": 2
}
| Constraint | Value |
|---|---|
| Allowed Types | PNG, JPG, JPEG |
| Max File Size | 5 MB |
This project is created for educational purposes and assignment submission.