Road Object Perception Method Based on Yolov9

This study focuses on deep learning based road vehicle object detection technology, aiming to address the challenges of vehicle detection in the fields of intelligent transportation systems and autonomous driving by combining the advantages of two popular algorithms, YOLOv9 and Faster R-CNN. The objective of this study is to enhance the accuracy and real-time performance of detection by leveraging the robust feature extraction capabilities of Convolutional Neural Networks (CNNs). Through rigorous data preprocessing, including Gaussian filtering and image enhancement techniques, the quality and diversity of the training data are ensured. The training and validation of the model utilized large-scale real-world datasets to enhance its generalization ability and robustness. For YOLOv9, we fully utilize its fast detection characteristics and combine PGI with a GELAN architecture to optimize the network structure and connection methods, ensuring that the model can efficiently extract and utilize image features. The experimental results show that the combination of YOLOv9 and Faster R-CNN performs well in road vehicle detection tasks, verifying its effectiveness and practicality in the fields of intelligent transportation systems and autonomous driving. This study offers a practical and viable solution for vehicle detection within intelligent transportation systems and establishes a foundation for further investigation into multi-sensor fusion and detection technologies in complex environments.