Computational Intelligence and Machine Learning
RGB Image Detection Model using YOLOv8 for Traffic Targets Detection in Autonomous Vehicles
Authors: Abhishek Kumar Gupta, Shruti Sangam and M. Brindha
Publishing Date: 23-04-2025
ISBN: 978-81-975670-5-6
Abstract
This study uses YOLOv8 to propose an improved traffic target detection model for autonomous cars. Complex problems including identifying cars, pedestrians, and traffic signs in a variety of environments—including dim lighting and occlusions—are handled by the RGB image detection model. By utilizing the KITTI dataset and sophisticated data augmentation methods such as random erasure and normalizing, the model significantly increases the robustness of detection. With a mean average precision (mAP) of 0.84 and a high inference speed, YOLOv8 outperforms baseline models and is appropriate for real-time applications. In order to further increase detection accuracy in harsh environments, future research will concentrate on integrating sensor modalities like LiDAR. These findings demonstrate how YOLOv8 might improve autonomous systems' perceptive abilities, resulting in safer navigation.
Keywords
Traffic Target Detection, YOLOv8, Autonomous Vehicles, KITTI Dataset, Image Processing, Computer Vision.
Cite as
Abhishek Kumar Gupta, Shruti Sangam and M. Brindha, "RGB Image Detection Model using YOLOv8 for Traffic Targets Detection in Autonomous Vehicles", In: Sandeep Kumar and Kavita Sharma (eds), Computational Intelligence and Machine Learning, SCRS, India, 2025, pp. 11-20. https://doi.org/10.56155/978-81-975670-5-6-2