Publication Type : Book
Publisher : Springer, Singapore
Source : Machine Learning and Autonomous Systems
Url : https://link.springer.com/chapter/10.1007/978-981-16-7996-4_14
Keywords : Lane detection,OpenCV,CNN (Convolutional Neural Network),FCN (Fully Convolutional Neural Network),SegNet
Campus : Bengaluru
School : School of Engineering
Department : Electronics and Communication
Year : 2022
Abstract : Advancements in lane detection algorithms lead to realizing autonomous driving technology and improving the real-time use of deep learning algorithms that are currently being studied for their various vision-based applications. In this paper, we propose an efficient and accurate algorithm to detect lanes and lane lines for autonomous vehicles. First, an OpenCV lane detection model was built to improve the understanding of lane detection algorithms. Using this knowledge, a CNN solution was adopted for the lane detection algorithm to train more data, work on curved lanes, and improve accuracy. The shortcomings of this model were identified and to solve these shortcomings, a Fully Convolutional Neural Network approach based on SegNet’s architecture was adopted. The accuracy of the models implemented was found to be as: OpenCV: 91%, CNN: 94.49%, and FCN: 96.12%.
Cite this Research Publication : Savant K.V., Meghana G., Potnuru G., Bhavana V, Lane Detection for Autonomous Cars Using Neural Networks. In: Chen J.IZ., Wang H., Du KL., Suma V. (eds) Machine Learning and Autonomous Systems. Smart Innovation, Systems and Technologies, vol 269. Springer, Singapore. https://doi.org/10.1007/978-981-16-7996-4_14