Point cloud segmentation algorithm based on feature fusion used for understory environments
-
摘要: 针对林下环境几何特征的复杂性,以及基于边检测、表面增长和聚类分割方法存在的效率低、分割不足及过度分割等问题,提出了一种基于特征融合的点云分割方法。采用地面激光扫描仪FARO在北京林业大学选择样本区域进行扫描,对扫描得到的数据进行采样点剔除及滤波,得到由1166302个点组成的林下环境点云数据,主要包括林木、地面、石块、人4类目标。综合利用点云法向量信息和激光反射强度信息可实现点云分割。其中,点云激光反射强度可直接从扫描得到的点云数据中获取;法向量可根据点云数据的三维坐标信息,通过对点云数据建立kd-tree数据结构,执行k-邻域搜索,并基于PlanePCA算法计算得到。将点云法向量和激光反射强度2方面的特征优势进行融合,计算中心点与邻域点的综合相异度,并判断其是否在阈值范围内,最终实现点云分割。比较基于特征融合、法向量和激光反射强度3种聚类分割方法得到的分割结果可知,基于特征融合的聚类分割方法能较好地保留数据特征,且分割完整度明显优于其他2种方法。Abstract: Aimed at the complexity of geometric features of understory environments and the deficiency of edge detection based method, region growing based method and clustering feature based method, we propose a new point cloud segmentation algorithm based on feature fusion. The 3D data set acquired from Beijing Forestry University using a FARO laser scanner consists of 1166302 points after removing outliers and filtering. The data set has four targets, i.e., tree, ground, stone and person. Point cloud segmentation can be achieved via fusing normal vector and laser reflection intensity of each point. The laser reflection intensity values can be obtained from point cloud data set directly, and normal vector should be calculated based on the Plane PCA algorithm. Also, it is necessary to create kd-tree data structure and perform k-NN search during the calculation of normal vector. Segmentation is realized after fusing the advantages of normal vector and laser reflection intensity and calculating synthetical difference degree between query points and neighborhood points. Comparing the segmentation results from point cloud segmentation algorithms based on feature fusion, normal vector and laser reflection intensity, the method based on feature fusion overcomes the problem of data deficiency that the other two methods suffer.
-
-
[1] NOVAK L M, BURL M C. Optimal speckle reduction in polarimetric SAR imagery[J]. Aerospace & Electronic Systems IEEE Transactions, 1990, 26(2):293-305.
[1] ZHANG Q. Research of visual recognition algoriyhm for service robot based on PCL[D].Hefei: University of Science and Technology of China, 2014.
[2] LIU Y F. Theory and method on point cloud data spatial management based on kd-tree[D]. Changsha: Central South University, 2009.
[2] GOZE S, LOPES A. A MMSE speckle filter for full resolution SAR polarimetric data[J]. Journal of Electromagnetic Waves & Applications, 1993, 7(7):717-737.
[3] HOFFMAN R, JAIN A K. Segmentation and classification of range images[J]. IEEE Transactions on Pattern Analysis & Machine Intelligence, 1987, 9(5):608-620.
[3] ZHANG L W. Research on the technology of the three-dimensional surface reconstruction from scattered point cloud[D].Changsha: Graduate School of National University of Defense Technology,2009.
[4] HUANG J, MENQ C H. Automatic data segmentation for geometric feature extraction from unorganized 3-D coordinate points[J]. Robotics & Automation IEEE Transactions, 2001, 17(3):268-279.
[4] ZHU D H. Point cloud library(PCL) learning tutorial[M]. Beijing:Beijing University of Aeronautics and Astronautics Press, 2012.
[5] LI T S, WANG Y M, HU C M. The study of point cloud registration automatically based on laser reflection intensity[J]. Bulletin of Surveying and Mapping, 2014(Suppl. 2):143-145.
[5] JEAN K, MINSOO S, SUCHENDRA M B. A multilayer self-organizing feature map for range image segmentation[J]. Neural Networks, 1995, 8(1): 67-86.
[6] PU S, VOSSELMAN G. Knowledge based reconstruction of building models from terrestrial laser scanning data[J]. Isprs Journal of Photogrammetry & Remote Sensing, 2009, 64(6):575-584.
[7] KOH J, SUK M, BHANDARKAR S M. A self-organizing neural network for hierarchical range image segmentation[C]//IEEE International Conference on Robotics & Automation. Atlanta, GA: IEEE, 1993:758-763.
[8] 张强. 基于点云库的服务机器人视觉识别算法研究[D]. 合肥:中国科学技术大学, 2014. [9] 刘艳丰. 基于kd-tree的点云数据空间管理理论与方法[D]. 长沙:中南大学, 2009. [10] KLASING K, ALTHOFF D, WOLLHERR D, et al. Comparison of surface normal estimation methods for range sensing applications[C]//Proceedings of the IEEE International Conference on Robotics and Automation (ICRA). Kobe, Japan:IEEE, 2009:12-17.
[11] SHAKARJI C M. Least-squares fitting algorithms of the NIST algorithm testing system[J]. Journal of Research of the National Institute of Standards & Technology, 1998, 103(6):633-641.
[12] 张连伟. 散乱点云三维表面重建技术研究[D]. 长沙:国防科学技术大学, 2009. [13] 朱德海.点云库PCL学习教程[M].北京:北京航空航天大学出版社,2012. [14] 李天烁, 王晏民, 胡春梅. 基于激光反射强度的点云自动配准研究[J]. 测绘通报, 2014(增刊2):143-145. -
期刊类型引用(6)
1. 林秀云，孙圆，刘晨曦，姚睿涵，周春国，曹林，曹福亮. 依据地面激光扫描数据的杉木材积建模与造材. 东北林业大学学报. 2022(01): 33-39 . 
百度学术
2. 李沛婷，赵庆展，田文忠，马永建. 结合无人机载LiDAR点云法向量的K-means++聚类精简. 国土资源遥感. 2020(02): 103-110 . 百度学术
3. 程子阳，任国全，张银. 扫描线段特征用于三维点云地面分割. 光电工程. 2019(07): 111-120 . 百度学术
4. 蔡越，徐文兵，梁丹，邓愫愫，李翀. 基于激光回波强度判别毛竹年龄. 中国激光. 2018(01): 272-280 . 百度学术
5. 曾碧，黄文. 一种融合多特征聚类集成的室内点云分割方法. 计算机工程. 2018(03): 281-286 . 百度学术
6. 田青华，白瑞林，李杜. 基于改进欧氏聚类的散乱工件点云分割. 激光与光电子学进展. 2017(12): 316-324 . 百度学术
其他类型引用(7)
计量
- 文章访问数: 1915
- HTML全文浏览量: 292
- PDF下载量: 35
- 被引次数: 13
下载:
