欢迎访问《汽车安全与节能学报》，
汽车安全与节能学报

汽车安全与节能学报 ›› 2024, Vol. 15 ›› Issue (3): 433-442.DOI: 10.3969/j.issn.1674-8484.2024.03.017

• 智能驾驶与智慧交通 • 上一篇    

基于MSFA-Net的车辆及车道线检测算法

文斌1,2(), 丁弈夫2, 胡一鸣2, 彭顺2, 胡晖2   

  1. 1.三峡大学 湖北省输电线路工程技术研究中心，宜昌 443002，中国
    2.三峡大学 电气与新能源学院，宜昌 443002，中国
  • 收稿日期:2023-09-02 修回日期:2023-12-04 出版日期:2024-06-30 发布日期:2024-07-01
  • 作者简介:文斌（1985—），男（汉），湖北，讲师。E-mail：wenbin_08@126.com。
  • 基金资助:
    国家自然科学基金资助项目(62273200);国家自然科学基金资助项目(61876097);湖北省输电线路工程技术研究中心研究基金资助项目(2022KXL03);湖北省自然科学基金联合基金类项目(2024AFD409)

Vehicle and lane detection algorithm based on MSFA-Net

WEN Bin1,2(), DING Yifu2, HU Yiming2, PENG Shun2, HU Hui2   

  1. 1. Hubei Provincial Engineering Technology Research Center for Power Transmission Line, China Three Gorges University, Yichang 443002, China
    2. China Three Gorges University, College of Electrical Engineering & New Energy, Yichang 443002, China
  • Received:2023-09-02 Revised:2023-12-04 Online:2024-06-30 Published:2024-07-01

RichHTML

5

PDF

76

摘要：

车辆检测与车道线分割是自动驾驶感知系统的重要组成部分，其基本要求是具有高精度和实时性。鉴此提出一种双任务多尺度特征聚合网络（MSFA-Net），该网络由1个特征提取网络和2个检测分支网络构成，实现了车辆和车道线同时检测。首先使用E-ELAN网络构造共享主干特征网络；在车辆检测分支网络设计增强卷积模块（CBS+）进行自下而上的特征融合以提升精度；在车道线检测分支网络使用特征融合模块（FeatFuse）对多分辨率特征进行自适应加权融合，配合空洞卷积语义感知模块（CDBS）使用梯形结构的多空洞值卷积对融合特征进行采样，以提升不连续车道线及其他非线性车道的分割精度。结果表明：在BDD100K数据集上，该文网络MSFA-Net其平均精度均值、召回率、像素准确率分别达到了81.3%、90.1%和80.1%，检测帧率达到了41.6 帧/s，能较好适应真实行车环境的需求。

关键词: 车辆检测, 交通图像, 深度学习, 车道线分割, 双任务多尺度特征聚合网络（MSFA-Net）

Abstract:

Vehicle detection and lane segmentation are important components of automatic driving sensing system, and their basic requirements are high precision and real-time. Therefore, a dual-task multi-scale feature aggregation network (MSFA-Net) was proposed, which was composed of one feature extraction network and two detection branch networks, and realized the simultaneous detection of vehicles and lane lines. First, E-ELAN network was used to construct the shared backbone feature network. Convolutional basic structure plus (CBS+) was designed for bottom-up feature fusion to improve accuracy in vehicle detection branch. To enhance the accuracy of discontinuous and nonlinear lane segmentation in lane segmentation branch, FeatFuse module was proposed for adaptive weight fusion of multi-features and context dilated convolutional basic structure (CDBS) for sampling fusion features through multi-dilation convolution of trapezoidal structure. The results show that on the BDD100K dataset, the average accuracy, recall rate and pixel accuracy of MSFA-Net reach 81.3%, 90.1% and 80.1% respectively, and the detection frame rate reaches 41.6 frames /s, which can better adapt to the needs of real-life driving scenarios.

Key words: vehicle detection, traffic image, deep learning, lane segmentation, dual task multi-scale feature aggregation network (MSFA-Net)

中图分类号: