计算机与数字工程2024，Vol.52Issue(5)：1529-1535,7.DOI:10.3969/j.issn.1672-9722.2024.05.046
动态环境下基于深度学习的视觉SLAM
Visual SLAM Based on Deep Learning in Dynamic Environment
摘要
Abstract
Traditional simultaneous visual localization and mapping(SLAM)technology is designed based on the assumption of a static environment.In a dynamic environment,the movement of a moving target will cause feature matching failure,which will affect the estimation of pose.A visual SLAM system combined with convolutional neural network is proposed.By adding a dynamic target detection thread of convolutional neural network combined with attention mechanism to the front end of the RGB-D mode of ORB-SLAM2 system,the dynamic target area is eliminated when extracting image feature points.Static feature points are used to complete accurate estimation of camera pose.The simulation experiments are tested under the TUM dynamic dataset,and the results of the multiple tests show that the improved algorithm improves the positional accuracy by more than 90%compared with the original algorithm,and the algorithm can meet the real-time requirements.关键词
同时定位与建图/深度学习/位姿估计/动态场景/目标检测Key words
simultaneous localization and mapping/deep learning/pose estimation/dynamic scene/target detection分类
信息技术与安全科学引用本文复制引用
陈明强,李奇峰,冯树娟,徐开俊..动态环境下基于深度学习的视觉SLAM[J].计算机与数字工程,2024,52(5):1529-1535,7.基金项目
民航飞行技术与飞行安全重点实验室自主研究项目(编号:FZ2021ZZ06) （编号:FZ2021ZZ06）
高质量民航特色"交通运输"硕士专业学位平台体系建设项目(编号:MHJY2022001)资助. （编号:MHJY2022001）
