+44 3308180992Harman Kaur, Kumbam Sathwik Reddy*, Singari Jashwanth, Sankuri Samith, Shubham Nautiyal, Amit Kumar Sharma
The rapid advancement of Autonomous Vehicles (AVs) is transforming transportation by integrating cutting-edge technologies, with Computer Vision (CV) playing a pivotal role. This research explores the critical applications of computer vision in enabling autonomous vehicles to navigate complex environments, detect objects, and make real-time decisions. Computer vision techniques, including object detection, lane tracking, pedestrian recognition, and traffic signal detection, provide AVs with a comprehensive understanding of their surroundings. We examine how deep learning models, particularly Convolutional Neural Networks (CNNs), enhance the accuracy of these tasks by processing vast amounts of visual data. Additionally, sensor fusion, integrating camera data with LiDAR and radar, is discussed to highlight its importance in creating robust perception systems for AVs. We also address the challenges in adverse weather conditions, dynamic environments, and real-time processing limitations that impede the full potential of computer vision in AVs. This paper aims to contribute to the ongoing development of safer, more efficient autonomous driving systems by proposing advancements in computer vision algorithms and techniques. By analysing current state-of-the-art approaches, we suggest future research directions to overcome existing limitations and improve the reliability of autonomous vehicles.
Share this article