| 黄卫华* **,张泽宇**,章政**.基于语义分割的动态VSLAM建图线程实现[J].高技术通讯(中文),2026,36(6):586~592 |
| 基于语义分割的动态VSLAM建图线程实现 |
| Realization of dynamic V-SLAM mapping thread based on semantic segmentation |
| |
| DOI：10. 3772 / j. issn. 1002 - 0470. 2026. 06. 004 |
| 中文关键词: 视觉即时定位与地图构建；语义分割；DeepLabV3+网络；动态环境 |
| 英文关键词: visual simultaneous localization and mapping, semantic segmentation, DeepLabV3+ network, dynamic environment |
| 基金项目: |
| 作者 | 单位 | | 黄卫华* ** | (*武汉城市学院机电工程学部武汉 430083)
(**武汉科技大学人工智能与自动化学院武汉 430081) | | 张泽宇** | | | 章政** | |
|
| 摘要点击次数: 80 |
| 全文下载次数: 77 |
| 中文摘要: |
| 针对动态环境下传统视觉即时定位与地图构建(visual simultaneous localization and mapping，VSLAM)建图存在信息冗余与地图拖影的问题，在ORB-SLAM3框架下，本文设计了基于DeepLabV3+网络的动态VSLAM建图线程。首先，通过轻量级DeepLabV3+网络对建图关键帧进行语义分割，实现剔除VSLAM地图关键帧中的动态区域；然后，将剔除动态特征后的位姿数据与深度学习信息相结合，构建了基于MobileNetV3的轻量级DeepLabV3+网络，生成排除动态特征的三维静态稠密点云地图及八叉树。实验结果表明，本文所设计方法有效减少了动态环境下VSLAM地图信息冗余量和地图拖影，降低了地图空间占用比。 |
| 英文摘要: |
| In order to solve the problems of redundant information and ghosting in map construction of traditional VSLAM in dynamic environments, a dynamic VSLAM mapping thread method based on DeepLabV3+ is designed and applied on ORB-SLAM3 architecture. Firstly, Key mapping frames are semantically segmented with a lightweight DeepLabV3+ network to exclude dynamic regions. Secondly, the pose data after eliminating dynamic features is integrated with deep learning information, and then the point cloud map and octree map with semantic information are generated. Finally, the experimental results demonstrate that the maps constructed by the improved mapping method are with less redundant information and fewer ghosts which decrease the map space occupancy. |
|
查看全文
查看/发表评论 下载PDF阅读器 |
| 关闭 |
