Transportation serves as a vital link for our economy, various types of vehicles have been an integral part of economic growth but have led to an enormous increase in traffic and its attendant problems like accidents and mortality. Hence, there is ample scope for improvement of newer technologies that have aimed to reduce the negative impact of high volumes of high-velocity traffic, easing traffic congestion and making transport safer. The objective of this paper is to review some of the recent advances in the field of lane detection in autonomous vehicles, the problems perceived, and the solutions designed. Image processing technologies used in traditional lane detection systems were observed to have inherent disadvantages. Further advances in lane detection consist of models built to detect lanes in complex environments mainly using two-stage lane feature extraction implemented with the you only look once (YOLO) v3 algorithm for precise and faster robust lane detection in complex real-life scenarios. In the next few years, huge breakthroughs will happen in autonomous vehicles being able to systematically detect lanes without supervision, using new computer vision-based lane detection methods to make complete autonomous vehicular travel a safe reality.
Recent Advances in Computer Vision Technologies for Lane Detection in Autonomous Vehicles
Algorithms for Intelligent Systems
International Conference on Computer Vision and Robotics ; 2023 ; Lucknow, India February 24, 2023 - February 25, 2023
2023-09-28
16 pages
Article/Chapter (Book)
Electronic Resource
English
Lane detection , Convolutional neural networks (CNN) , Recurrent neural network (RNN) , Region-based convolutional neural network (RCNN) , Spatial , Temporal , Lane boundary regression , Deep reinforcement learning , Deep Q learning (DQLL) , Lane localization , T-S YOLO v3 You Only Look Once (YOLO) , Semantic segmentation network , 3D network models , Adaptive learning methods , Random sample consensus algorithm (RANSAC) Engineering , Control, Robotics, Mechatronics , Computer Imaging, Vision, Pattern Recognition and Graphics , Signal, Image and Speech Processing , Cyber-physical systems, IoT