This repository contains a Python-based program that detects and tracks people in a video, counting the number of individuals entering and exiting a defined area. It uses the YOLOv8 model for object detection and the SORT (Simple Online and Realtime Tracking) algorithm for tracking.
Before running the code, make sure the following Python libraries are installed:
- sahi: For seamless integration with the YOLO model.
- OpenCV (cv2): For video frame manipulation and display.
- imutils: For frame resizing.
- numpy: For numerical operations.
- SORT: For tracking detected objects across video frames.
- Clone the Repository:
git clone https://github.com/yourusername/entry-exit-counter.git cd entry-exit-counter - Create and activate a virtual environment (optional):
python -m venv env
source env/bin/activate # On Windows, use "env\Scripts\activate"
- Install dependencies:
pip install sahi opencv-python imutils numpy
- Run Program:
python entry_exit_counter.py
- Real-time detection and tracking using YOLOv8 and SORT.
- Entry and exit counting with customizable boundaries.
- Graphical display with bounding boxes and unique IDs for detected individuals.
- High contrast video processing for better visibility
- video_path: Set the path to your video file in the script.
- confidence_threshold: Adjust this to change detection confidence.
- iou_threshold: Modify this in the SORT tracker for performance tuning.
- device: Set to 'cuda:0' for GPU processing or 'cpu' for CPU processing.
- Python 3.6+
- Compatible with both CPU and GPU execution (GPU recommended for better performance).
This project is licensed under the MIT License. See the LICENSE file for details.
Contributions are welcome! Please open an issue or submit a pull request for any changes or additions.
- Ultralytics for the YOLOv8 model.
- SORT algorithm by Alex Bewley et al.
- SAHI