This project focuses on real-time queue analysis, helping address growing urban traffic challenges through AI-powered vehicle detection.
Team Members :
- Satyam Bhagat
- Utkarsha Bhoir
- Hinal Macchi
- Saini Pagdhare
Project Overview :
Managing traffic congestion efficiently is becoming increasingly important in modern cities. Our solution was designed to detect and analyze vehicle queues in real time using deep learning and computer vision techniques. The system can support smarter traffic management, improve monitoring efficiency, and assist in data-driven urban planning.
Dataset & Training :
The model was trained on a dataset of 2571 annotated images prepared specifically for vehicle queue analysis. Using YOLOv8, we trained and evaluated the system on multiple performance metrics including:
- Precision
- Recall
- mAP@0.5
- mAP@0.5:0.95
- Box, Classification, and DFL Loss
We also analyzed training performance through:
- Confusion Matrices
- Precision-Recall Curves
- F1-Score Analysis
- Training & Validation Visualizations
Model Training and Validation :
We trained the YOLOv8 model for object detection. Key metrics monitored
were Precision, Recall, mAP@0.5, mAP@0.5-0.95, and loss values (box,
class, and DFL loss). During training, we analyzed visual metrics like
confusion matrices, PR curves, and F1-score curves. Our results were logged
in results.csv and visualized in plots like results.png, confusion_matrix.png,
and more.
Model Deployment & Results :
The trained model achieved strong real-time detection performance with reliable accuracy and smooth inference speed. The project demonstrates the practical application of AI in intelligent transportation and traffic analysis systems.
GitHub Repository :
Demo Videos & Prediction Outputs :
Drive Folder – Videos & Photos
Challenges Faced
- Collecting diverse images datasets
- Manual annotation of 1911 images
- Handling crowded parking scenarios
- Reducing false detections
- Real-time deployment optimization
These challenges helped improve our teamwork, debugging, and problem-solving skills throughout the internship.
Conclusion
The project successfully demonstrated how AI and Computer Vision can be used to solve real-world Queue and Parking management problems.
Through this internship, we gained practical experience in:
- Dataset collection
- Manual image annotation
- YOLO model training
- Real-time deployment
- AI-based surveillance systems
This experience strengthened our understanding of Machine Learning, Deep Learning, and Computer Vision workflows while improving our practical industry skills.
Special Thanks
A sincere thanks to Dr. Chandrakant Bothe and the WISERLI Team for their mentorship, guidance, and support throughout the internship journey.
#YOLOv8 #ComputerVision #ArtificialIntelligence #DeepLearning #TrafficMonitoring #VehicleDetection #ObjectDetection #MachineLearning #WISERLI yolovx