重庆大学学报2025，Vol.48Issue(8)：78-85,8.DOI:10.11835/j.issn.1000-582X.2025.08.007
基于多尺度特征融合和边缘增强的多传感器融合3D目标检测算法
Multi-sensor fusion 3D target detection algorithm based on multi-scale feature fusion and edge enhancement
摘要
Abstract
BEV(bird's eye view)-based multi-sensor fusion perception algorithms for autonomous driving have made significant progress in recent years and continue to contribute to the development of autonomous driving.In the research of multi-sensor fusion perception algorithms,multi-view image-to-BEV conversion and multi-modal feature fusion have been the key challenges in BEV perception algorithms.In this paper,we propose MSEPE-CRN,a fusion sensing algorithm of camera and millimeter-wave radar for 3D target detection,which utilizes edge features and point clouds to improve the accuracy of depth prediction,and then realizes the accurate conversion of multi-view images to BEV features.Meanwhile,a multi-scale deformable large kernel attention mechanism is introduced for modal fusion to solve the misalignment problem due to the excessive difference of features from different sensors.Experimental results on the nuScenes open-source dataset show that compared to the baseline network,the proposed algorithm achieves improvements of 2.17%in mAP,1.93%in NDS,2.58%in mATE,8.08%in mAOE,and 2.13%in mAVE.This algorithm can effectively improve the vehicle's ability to perceive moving obstacles on the road,and has practical value.关键词
3D目标检测/Bird's eye view/多模态融合/深度预测Key words
3D target detection/bird's eye view/multi-modal fusion/depth prediction分类
交通工程引用本文复制引用
刘建国,陈文,赵奕凡,周琪,颜伏伍,尹智帅,郑灏,吴友华..基于多尺度特征融合和边缘增强的多传感器融合3D目标检测算法[J].重庆大学学报,2025,48(8):78-85,8.基金项目
佛山仙湖实验室先进能源科学与技术广东开放基金(XHD2020-003) （XHD2020-003）
广西科技尖峰计划(AA23062030).Supported by Guangdong Open Fund Project of Advanced Energy Science and Technology of Foshan Xianhu Laboratory(XHD2020-003)and Guangxi Key Science and Technology R&D Program(AA23062030). （AA23062030）
