当前位置: X-MOL 学术IEEE Trans. Geosci. Remote Sens. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Accurate Rock-Mass Extraction From Terrestrial Laser Point Clouds via Multiscale and Multiview Convolutional Feature Representation
IEEE Transactions on Geoscience and Remote Sensing ( IF 8.2 ) Pub Date : 2020-09-24 , DOI: 10.1109/tgrs.2020.3023119
Yunbiao Wang , Shibiao Xu , Jun Xiao , Feipeng Wang , Ying Wang , Lupeng Liu

Existing 3-D object extraction methods on terrestrial laser point clouds are further developed through filtering and labeling. However, such predefined features are heuristically designed to process generic object point clouds. Thus, existing abilities are insufficient to handle specific rock-mass point clouds. Given the complexity and diversity of terrestrial environments, the effective removal of vegetation points from rock-mass point clouds is particularly challenging. To address such problems, this study presents a novel approach for 3-D rock-mass point clouds labeling by using convolutional feature learning based on distribution priors with multiple scales and views. First, to extract discriminative features of each point for classification, we propose novel multiview supporting planes to analyze the spatial distribution and structure of its neighboring points for each category. Second, we define the multiscale spatial distribution matrix on a grid representation (e.g., the number of points projected into each cell). Last, the statistical information of points is nonlinearly combined and hierarchically compressed to generate a compact and effective convolutional feature representation for classification. The effectiveness of the proposed method is evaluated via experiments on rock-mass point clouds from different scenes. Compared with existing extraction approaches, experimental results indicate the superiority of the proposed method in terms of the precision and recall.

中文翻译:

通过多尺度和多视图卷积特征表示从地面激光点云中精确提取岩质

通过过滤和标记,可以进一步开发现有的地面激光点云上的3-D对象提取方法。但是,此类预定义功能经过启发式设计,可以处理通用对象点云。因此,现有能力不足以处理特定的岩质点云。考虑到陆地环境的复杂性和多样性,从岩质点云中有效去除植被点尤其具有挑战性。为了解决此类问题,本研究提出了一种基于卷积特征学习的3D岩石质点云标记的新方法,该方法基于具有多个尺度和视图的先验分布。首先,要提取每个点的区分特征以进行分类,我们提出了新颖的多视图支撑平面,以分析每个类别的相邻点的空间分布和结构。其次,我们在网格表示上定义多尺度空间分布矩阵(例如,投影到每个像元中的点数)。最后,将点的统计信息进行非线性组合和分层压缩,以生成紧凑有效的卷积特征表示法进行分类。通过对不同场景的岩体点云进行实验,评价了该方法的有效性。与现有的提取方法相比,实验结果表明了该方法在精度和查全率方面的优越性。我们在网格表示上定义多尺度空间分布矩阵(例如,投影到每个像元中的点数)。最后,将点的统计信息进行非线性组合和分层压缩,以生成紧凑有效的卷积特征表示法进行分类。通过对不同场景的岩体点云进行实验,评价了该方法的有效性。与现有的提取方法相比,实验结果表明了该方法在精度和查全率方面的优越性。我们在网格表示上定义多尺度空间分布矩阵(例如,投影到每个像元中的点数)。最后,将点的统计信息进行非线性组合和分层压缩,以生成紧凑有效的卷积特征表示法进行分类。通过对不同场景的岩体点云进行实验,评价了该方法的有效性。与现有的提取方法相比,实验结果表明了该方法在精度和查全率方面的优越性。通过对不同场景的岩体点云进行实验,评价了该方法的有效性。与现有的提取方法相比,实验结果表明了该方法在精度和查全率方面的优越性。通过对不同场景的岩体点云进行实验,评价了该方法的有效性。与现有的提取方法相比,实验结果表明了该方法在精度和查全率方面的优越性。
更新日期:2020-09-24
down
wechat
bug