Night time Vehicle Detection Algorithm Based on Improved YOLO

Nighttime vehicle detection plays a crucial role in enhancing safety and surveillance systems by enabling accurate identification of vehicles in low-light conditions. This project presents a Nighttime Vehicle Detection Algorithm based on the improved YOLOv9 model, designed to detect vehicles such as buses, cars, motorcycles, and trucks in nighttime environments. The algorithm leverages YOLOv9's advanced object detection capabilities, optimizing the model for detecting vehicles with high precision, recall, and mean average precision (mAP). With the dataset comprising diverse vehicle types, the model achieves a significant performance, particularly excelling in detecting cars with an mAP50 of 0.993. By fine-tuning the YOLOv9 model, the project addresses the challenges of detecting vehicles under low visibility conditions, providing an efficient solution for intelligent transportation systems and surveillance applications. The model's performance metrics, including precision, recall, and mAP, demonstrate its ability to reliably detect various vehicle types, even in complex nighttime scenarios. This work highlights the potential of YOLOv9 as a robust tool for real-time vehicle detection in challenging environments.

Keywords: Nighttime Vehicle Detection, YOLOv9, Object Detection, Vehicle Identification, Real-Time Surveillance, Precision, Recall, Mean Average Precision (mAP), Deep Learning, Intelligent Transportation Systems.