Research on Optimization of Lane Detection Algorithm under Complex Road Background
Authors:
- 17 November 2023
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 211-216
Abstract
In order to improve the effect of lane detection in complex and changeable road background, this paper proposes a lane detection algorithm that combines attention mechanism and deformable convolution. The algorithm can adaptively adjust the weight of each pixel position in the feature map, so that the shape and position information of lane lines in the image can be captured more quickly and accurately. In addition, the algorithm can better learn the continuous changes of lane lines in the image, thus improving the detection effect. The experimental results show that the detection effect of the algorithm proposed in this paper is better than the traditional lane detection method on Tusimple dataset, and the algorithm can accurately detect lane lines in the face of complex road scenes. Therefore, the optimization and research of this algorithm improve the lane detection performance in the complex and changeable road background.Keywords
This publication has 14 references indexed in Scilit:
- CondLaneNet: a Top-to-down Lane Detection Framework Based on Conditional Convolution2021
- Key Points Estimation and Point Instance Segmentation Approach for Lane DetectionIEEE Transactions on Intelligent Transportation Systems, 2021
- PolyLaneNet: Lane Estimation via Deep Polynomial Regression2021
- Attention deep neural network for lane marking detectionKnowledge-Based Systems, 2020
- Attention-Based Deep Neural Networks for Detection of Cancerous and Precancerous Esophagus Tissue on Histopathological SlidesJAMA Network Open, 2019
- Learning Lightweight Lane Detection CNNs by Self Attention Distillation2019
- FastDraw: Addressing the Long Tail of Lane Detection by Adapting a Sequential Prediction Network2019
- Deformable ConvNets V2: More Deformable, Better Results2019
- CBAM: Convolutional Block Attention Module2018
- Deformable Convolutional Networks2017