This project focuses on developing a real-time Billing Queue Detection System using the YOLOv9 object detection framework to monitor and analyze customer queues in retail and supermarket environments. The main objective of this system is to improve customer management, reduce waiting time, and assist store staff in handling crowd congestion efficiently through AI-powered surveillance and detection.
The model is trained on a custom dataset containing multiple queue scenarios collected under different lighting conditions, camera angles, crowd densities, and real-world retail situations. Using YOLOv9, the system is capable of detecting people standing in billing queues with high accuracy and speed. The detection process can be further extended for queue counting, overcrowding alerts, queue length estimation, and smart analytics for store optimization.
This project demonstrates the practical application of Computer Vision and Deep Learning in solving real-world operational challenges. The workflow includes dataset collection, image annotation, preprocessing, model training, validation, testing, and deployment. The implementation aims to provide efficient real-time performance while maintaining accuracy in dynamic environments.
Key Features: • Real-time queue detection using YOLOv9
• Custom-trained dataset for retail queue scenarios
• Fast and accurate object detection
• Crowd and queue monitoring support
• Scalable for malls, supermarkets, and billing counters
• Potential integration with smart surveillance systems
Technologies Used: • Python
• YOLOv9
• OpenCV
• Roboflow
• Deep Learning & Computer Vision Techniques
Special thanks to my amazing teammates for their valuable support, teamwork, and contributions throughout the development of this project:
• Pratishtha Upadhyay
• Ishita Patil
• Faiz Shaikh
Their collaboration, ideas, and continuous efforts played an important role in making this project successful. 
Demo video: demo