基于深度可分离卷积和残差注意力模块的车道线检测方法  

A lane line detection method based on depthwise separable convolution and residual attention modules

作　　者：崔明义 冯治国[1] 代建琴 赵雪峰[1] 袁森 CUI Mingyi;FENG Zhiguo;DAI Jianqin;ZHAO Xuefeng;YUAN Sen(Guizhou University,Guiyang 550025,China;Guizhou University of Technology,Guiyang 550003,China)

机构地区：[1]贵州大学,贵阳550025 [2]贵州理工学院,贵阳550003

出　　处：《激光杂志》2024年第4期81-87,共7页Laser Journal

基　　金：贵州省科技重大专项(No.黔科合重大专项字ZNWLQC[2019]3012);贵州省交通运输厅科技项目(No.2019-312-020、2021-322-02)。

摘　　要：针对全天候条件下道路车道线视觉检测技术存在的算法结构复杂、参数数量较多等问题,提出一种基于深度可分离卷积和残差注意力模块的车道线检测方法,建立了LPINet网络模型。利用深度可分离卷积减小输入图像尺寸,设计三种不同结构的瓶颈残差单元降低网络参数数量,引入ECANet注意力机制增加重要特征通道权重,提升车道线检测精度。在Tusimple数据集和GZUCDS自建数据集上的实验结果表明:在晴天场景下,LPINet网络车道线检测精度可达96.62%,且模型参数量降至1.64 MB,实现了轻量化设计;在雾天、雨天、夜晚和隧道复杂场景中进行了探索性研究,车道线检测精度达到93.86%,证明了方法的有效性。Aiming at the problems of complex algorithm structure and large number of parameters existing in road lane visual detection technology under all-weather conditions,a lane detection method based on depth-separable con-volution and residual attention module is proposed,and the LPINet network model is established.We use depth-sepa-rable convolution to reduce the size of the input images,design three bottleneck residual units with different structures to reduce the number of network parameters,and introduce the ECANet attention mechanism,which can increase the weight of important feature channels,to improve the lane detection accuracy.The experimental results on Tusimple dataset and GZUCDS self-built dataset show that the LPINet network lane detection accuracy can reach 96.62%in sunny scenarios,and the number of model parameters is reduced to 1.64 MB,which realizes the lightweight design.We carried out exploratory researches in complex scenes such as foggy,rainy,night and tunnel,and the accuracy of lane detection reaches 93.86%,which proves the effectiveness of our method.

关 键 词：车道线检测 深度学习 残差网络 深度可分离卷积 注意力机制 

分 类 号：TN249[电子电信—物理电子学]