激光加工技术参考资料激光视觉融合下的运动检测与失配矫正张强.pdfVIP

Opto-Electronic Engineering Article 光 电 工 程 2017 年，第44 卷，第 11 期   激光视觉融合下的运动检测与 失配矫正 1 1,2 1 2 张 强 ，赵江海 *，袁雅薇 ，方世辉 1 中国科学技术大学自动化系，合肥 230026 ； 2 中国科学院合肥物质科学研究院先进制造技术研究所，江苏 常州 213000 摘要：针对单一传感器在动态场景感知问题上的局限性，设计了一种融合激光与视觉的实现系统，并对运动检测中的 背景显露区误判问题和融合中不同传感器间点云的失配问题分别提出了改进算法。在运动检测上，首先基于视觉的背 景差分算法对激光进行前景点分拣，再以激光前景点为启发信息进行视觉前景聚类。在融合失配问题上，首先基于栅 格失配度分别对激光和视觉点云进行聚类分割，再以激光为基准，逐一将对应的视觉点云与之配准，滤除噪声后所得 到的矫正点云可用于场景重建进行进一步验证。实验结果表明，改进算法所获得的融合前景对“影子”有更好的鲁棒 性；较之整体配准的矫正，改进算法在平均失配度上降低了约75%，在y 和z 方向上的偏移比收敛了至少5%。 关键词：激光与视觉融合；运动检测；背景显露区；失配矫正；场景重建 中图分类号： TP391.41 文献标志码：A Motion detection and mismatching correction by fusing laser and vision 1 1, 2 1 2 Qiang Zhang, Jianghai Zhao *, Yawei Yuan and Shihui Fang 1Department of Automation, University of Science and Technology of China, Hefei 230026, China; 2Advanced Manufacturing Technology Institute, Hefei Institute of Physical Science, Chinese Academy of Sciences, Changzhou 213000, China Abstract: Aiming at the limitations of single sensor in dynamic scene perception issue, an implementation system for fusing laser and vision was designed. In addition, two improved algorithms were proposed to solve the problems of the error foreground detection in the motion detection and the mismatching between the point clouds of different sensors. As for motion detection, the laser foreground points were firstly detected based on visual background sub- traction algorithm. Then, the visual foreground was clustered regarding laser foreground points as the heuristic in- formation. To solve the mismatching of fusion, the las