Smart-Parking-using-YOLO-models

Leveraging Object Detection for Efficient Parking Space Monitoring: A Benchmark of YOLO Models.

Introduction

This project explores the use of YOLO (You Only Look Once) object detection models for efficient parking space monitoring. The aim is to evaluate various YOLO models (YOLOv5, YOLOv6, YOLOv7, YOLOv8, YOLOv9) for parking vacancy detection and assess them based on mean Average Precision (mAP), inference speed, and complexity.

Objective

The primary objective of this project is to assess the performance of different YOLO models for detecting parking vacancy in real-time using computer vision techniques.

Key Features

• Evaluate YOLOv5, YOLOv6, YOLOv7, YOLOv8, YOLOv9, and YOLO NAS for parking vacancy detection.
• Measure performance metrics including mean Average Precision (mAP), inference speed, and model complexity.
• Implement real-time monitoring of parking spaces using the selected YOLO model.

Methodology

• Utilize publicly available datasets for training and evaluation.
• Implement YOLO models using popular deep learning frameworks like PyTorch or TensorFlow.
• Conduct experiments to compare and analyze the effectiveness of each YOLO variant for parking space detection.

Usage

1. Installation:

   • Clone the repository:

     git clone https: https://github.com/umeshgjh/Smart-Parking-using-YOLO-models

2. Evaluation:

   • Run scripts to evaluate different YOLO models on parking vacancy detection.

3. Real-time Monitoring:

   • Implement real-time monitoring using the selected YOLO model.

Results

The results of this project will provide insights into the best YOLO model for parking space monitoring applications based on performance and efficiency metrics.

Future Work

• Explore additional data augmentation techniques to improve model robustness.
• Investigate deployment strategies for real-world parking space monitoring systems.

Contributing

Contributions are welcome!

License

This project is licensed under the MIT License.