https://github.com/sanshruthr/cctv_sentry_yolo11
Real-time monitoring, object tracking, and line-crossing detection for CCTV camera streams.
https://github.com/sanshruthr/cctv_sentry_yolo11
cctv cctv-surveillance huggingface yolo11 yolov11
Last synced: 8 months ago
JSON representation
Real-time monitoring, object tracking, and line-crossing detection for CCTV camera streams.
- Host: GitHub
- URL: https://github.com/sanshruthr/cctv_sentry_yolo11
- Owner: SanshruthR
- Created: 2025-01-26T07:44:55.000Z (9 months ago)
- Default Branch: main
- Last Pushed: 2025-02-17T08:45:41.000Z (8 months ago)
- Last Synced: 2025-02-17T09:34:32.374Z (8 months ago)
- Topics: cctv, cctv-surveillance, huggingface, yolo11, yolov11
- Language: Python
- Homepage: https://huggingface.co/spaces/Sanshruth/CCTV_SENTRY_YOLO11
- Size: 4.72 MB
- Stars: 2
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# CCTV_SENTRY_YOLO11
## Overview
CCTV_SENTRY_YOLO11 is an advanced object detection system built using YOLOv11 by Ultralytics. It provides real-time monitoring, object tracking, and line-crossing detection for IP camera streams. Hosted on Hugging Face Spaces, it enables users to easily interact with the model via a web interface.
## Features
- **Line-Crossing Detection**: Detects objects crossing user-defined lines.
- **Real-Time Object Detection**: Utilizes YOLOv11 for high-speed object tracking.
- **Interactive Interface**: Powered by Gradio for an intuitive user experience.
- **Customizable Classes**: Filter objects based on specific detection classes.
- **Detailed Visualization**: Annotated frames with bounding boxes, IDs, and counts.
https://github.com/user-attachments/assets/e29ad9df-b810-4308-b6a8-4ff81019edea
## Model Details
The system leverages the YOLOv11 model from Ultralytics for accurate and efficient object detection. Key technologies include:
- **OpenCV**: For video frame processing.
- **Gradio**: For creating an interactive user interface.
- **Ultralytics YOLO**: For state-of-the-art object detection and tracking.
## How It Works
1. Upload or provide the URL of an IP camera stream.
2. Draw a line on the first frame to set the detection boundary.
3. Select the object classes to monitor.
4. Watch real-time detections and line-crossing counts directly on the interface.
## Industrial and Commercial Applications
### **Parking Management**
- Vehicle entry/exit tracking
- Parking space occupancy monitoring
- Unauthorized parking detection
### **Manufacturing**
- Conveyor belt product counting
- Quality control inspections
- Real-time inventory tracking
### **Retail and Logistics**
- Customer movement analysis
- Stock level monitoring
- Theft prevention systems
### **Transportation**
- Vehicle tracking
- Loading dock management
- Traffic flow analysis
### **Security**
- Perimeter surveillance
- Restricted area monitoring
- Crowd density estimation
## Usage
### Requirements
- Python 3.x
- Install dependencies:
```bash
pip install ultralytics opencv-python-headless gradio numpy pillow
```
### Run Locally
1. Clone the repository:
```bash
git clone https://github.com/SanshruthR/CCTV_SENTRY_YOLO11.git
```
2. Navigate to the project directory:
```bash
cd CCTV_SENTRY_YOLO11
pip install -r requirements.txt
```
3. Start the application:
```bash
python app.py
```
### Live Demo
Experience the project live on Hugging Face Spaces:
[CCTV_SENTRY_YOLO11 on Hugging Face](https://huggingface.co/spaces/Sanshruth/CCTV_SENTRY_YOLO11)
## Deployment
* To create a live HLS stream for testing, refer to this GitHub repository:
https://github.com/SanshruthR/mock-hls-server
* Use the sample video file for testing:
https://videos.pexels.com/video-files/1169852/1169852-hd_1920_1080_30fps.mp4
## License
This project is licensed under the MIT License.