Traffic Volume API

FastAPI application for counting vehicles in images using YOLO.

Features

• API Endpoint: Process frames and count vehicles via /process
• Batch Processing: Process local images by date and store results in NPY format

Installation

```bash

Clone repository

git clone https://github.com/blueonline07/capstone
cd capstone

Install dependencies

pip install -r requirements.txt
```

Usage

API Server

Start the server:

```bash

uvicorn main:app --reload

Send POST request to `/process`:

    ```bash
curl -X POST "http://127.0.0.1:8000/process" \
  -H "Content-Type: application/json" \
  -d '{
    "camera_id": "cam_001",
    "slot": "00:00:00",
    "generated_at": "2025-11-15T00:00:00",
    "duration_sec": 300,
      "frames": [
      {"time": "2025-11-15T00:00:00", "image_ref": "path/to/image.jpg"}
    ],
    "speed": null,
    "weather": null
    }'
    ```

### Batch Processing

Process local images by date from folder structure:

```bash
python process_batch.py --date 2025-11-15

Input structure: batch/<node>/batch_<slot>.json

Output: output/<date>/<slot>.npy containing [response_of_node1, response_of_node2, ...]

Each slot file across all nodes is processed and combined into one output file per slot.

Configuration

Environment variables (see config/settings.py):

• APP_NAME: Application name (default: "Traffic Volume API")

Models

The application uses YOLO11n (yolo11n.pt) to detect vehicles with the following classes:

• Class 1: Bicycle
• Class 2: Car
• Class 3: Motorcycle
• Class 5: Bus
• Class 7: Truck

Project Structure

├── main.py                    # FastAPI application
├── process_batch.py           # Batch processing script
├── models/                    # Pydantic models
│   ├── request.py
│   ├── response.py
│   └── common.py
├── config/                    # Configuration
│   └── settings.py
├── batch/                     # Input folder
│   └── <node>/
│       └── batch_<slot>.json
└── output/                    # Batch processing results
    └── <date>/
        └── <slot>.npy