ObjectDetection / README.md
EnYa32's picture
Update README.md
92f6d1f verified
|
Raw History Blame Contribute Delete
1.37 kB
---
title: ObjectDetection
emoji: πŸš—
colorFrom: red
colorTo: red
sdk: streamlit
app_file: src/streamlit_app.py
pinned: false
short_description: Detect and count cars/trucks in images (cvlib + OpenCV).
license: mit
---
# πŸš— Vehicle Counter – Object Detection App
This project uses computer vision to detect and count vehicles in an image.
A pre-trained object detection model from **cvlib** (YOLO) is used to identify cars and trucks.
The results are saved as:
- an output image with bounding boxes (`detected.jpg`)
- a JSON file with the object counts (`results.json`)
---
## πŸ” Features
- Upload an image (JPG / PNG / WEBP)
- Detect vehicles using cvlib and OpenCV
- Count cars and trucks automatically
- Save results as JSON
- Display detected image in a Streamlit app
---
## 🧠 How it works
1. The user uploads an image
2. Objects are detected using `cvlib.detect_common_objects`
3. Bounding boxes are drawn on the image
4. The number of cars and trucks is counted
5. Results are saved to `src/results.json` and `src/detected.jpg`
---
## ▢️ Run locally
```bash
pip install -r requirements.txt
streamlit run app.py
πŸ“ Project Structure
.
β”œβ”€β”€ app.py
β”œβ”€β”€ requirements.txt
β”œβ”€β”€ README.md
└── src/
β”œβ”€β”€ results.json
└── detected.jpg
πŸ“Œ Output Example
{
"cars": 11,
"trucks": 3
}
Author
Enes Yamac