This repository contains code and resources for a dustbin detection and classification system using YOLOv8. The system comprises a single YOLOv8 model trained for both dustbin detection and status classification. The model is capable of detecting the presence of dustbins in real-time surveillance footage and classifying their status (full or not full) based on images captured from different angles.
- Detection of dustbins from real-time surveillance footage.
- Classification of dustbin status (full or not full) for efficient waste management.
- Alert system: Sends alerts when a dustbin is detected as full, prompting timely action for emptying.
The model is trained using a diverse dataset of dustbin images, including:
- Images captured from various angles to simulate CCTV camera viewpoints.
- Images with different lighting conditions (bright, low light, and dark).
- Images of dustbins filled to different levels to train the classification model.
- Python 3.x
- TensorFlow
- OpenCV
- YOLOv8 (implementation details and pre-trained weights included)
- Clone this repository to your local machine.
- Install the required dependencies.
- Run the detection and classification scripts.
- Customize the alert system as needed for your environment.
-
Detection Model
Outputs
-
Classification Model
Outputs
Contributions are welcome! If you have any suggestions, bug fixes, or improvements, feel free to open an issue or submit a pull request.
Special thanks to the contributors of the YOLOv8 implementation used in this project.