Feature detection on point clouds via Gabriel Triangles creation and l1 normal reconstruction

In this paper, we present a robust subneighborhoods selection technique for feature detection on point clouds scattered over a piecewise smooth surface. The proposed method first identifies all potential features using covariance analysis of the local- neighborhoods. To further extract the accurate features from potential features, Gabriel triangles are created in local neighborhoods of each potential feature vertex. These triangles tightly attach to underlying surface and effectively reflect the local geometry struc- ture. Applying a shared nearest neighbor clustering algorithm on ~ 1 reconstructed normals of created triangle set, we classify the lo- cal neighborhoods of the potential feature vertex into multiple subneighborhoods. Each subneighborhood indicates a piecewise smooth surface. The final feature vertex is identified by checking whether it is locating on the intersection of the multiple surfaces. An advantage of this framework is that it is not only robust to noise, but also insensitive to the size of selected neighborhoods. Ex- perimental results on a variety of models are used to illustrate the effectiveness and robustness of our method.

Study on the Universality of the Normal Cloud Model

The distribution function is an important tool in the study of the stochastic variances. The normal distribution is very popular in the nature and our society. The idea of membership functions is the foundation of the fuzzy sets theory. While the fuzzy theory is widely used, the completely certain membership function that has no any fuzziness at all has been the bottleneck of the applications of this theory. Cloud models are the effective tools in transforming between qualitative concepts and their quantitative expressions. It can represent the fuzziness and randomness and their relations of uncertain concepts. Also cloud models can show the concept granularity in multi-scale spaces by the digital characteristic Entropy (En). The normal cloud model not only broadens the form conditions of the normal distribution but also makes the normal membership function be the expectation of the random membership degree. In this paper, the universality of the normal cloud model is proved, which is more superior and easier, and can fit the fuzziness and gentleness of human cognitive processing. It would be more applicable and universal in the representation of uncertain notions.

Road Surface Modeling and Representation from Point Cloud Based on Fuzzy Clustering

A scheme for an automatic road surface modeling from a noisy point cloud is presented. The normal vectors of the point cloud are estimated by distance-weighted fitting of local plane. Then, an automatic recognition of the road surface from noise is performed based on the fuzzy clustering of normal vectors, with which the mean value is calculated and the projecting plane of point cloud is created to obtain the geometric model accordingly. Based on fuzzy clustering of the intensity attributed to each point, different objects on the road surface are assigned different colors for representing abundant appearances. This unsupervised method is demonstrated in the experiment and shows great effectiveness in reconstructing and rendering better road surface.

Comprehensive evaluation of 5G+smart distribution network based on combined weighting method-cloud model

With the large-scale application of 5G technology in smart distribution networks,the operation effects of distribution networks are not clear.Herein,we propose a comprehensive evaluation model of a 5G+smart distribution network based on the combination weighting and cloud model of the improved Fuzzy Analytic Hierarchy-Entropy Weight Method(FAHP-EWM).First,we establish comprehensive evaluation indexes of a 5G+smart distribution network from five dimensions:reliable operation,economic operation,efficient interaction,technological intelligence,and green emission reduction.Second,by introducing the principle of variance minimization,we propose a combined weighting method based on the improved FAHP-EWM to calculate the comprehensive weight,so as to reduce the defects of subjective arbitrariness and promote objectivity.Finally,a comprehensive evaluation model of 5G+smart distribution network based on cloud model is proposed by considering the uncertainty of distribution network node information and equipment status information.The example analysis indicates that the overall operation of the 5G+smart distribution network project is decent,and the weight value calculated by the combined weighting method is more reasonable and accurate than that calculated by the single weighting method,which verifies the effectiveness and rationality of the proposed evaluation method.Moreover,the proposed evaluation method has a certain guiding role for the large-scale application of 5G communication technology in smart distribution networks.

A new approach to retrieve leaf normal distribution using terrestrial laser scanners

Leaf normal distribution is an important structural characteristic of the forest canopy. Although terrestrial laser scanners（TLS） have potential for estimating canopy structural parameters, distinguishing between leaves and nonphotosynthetic structures to retrieve the leaf normal has been challenging. We used here an approach to accurately retrieve the leaf normals of camphorwood（Cinnamomum camphora） using TLS point cloud data.First, nonphotosynthetic structures were filtered by using the curvature threshold of each point. Then, the point cloud data were segmented by a voxel method and clustered by a Gaussian mixture model in each voxel. Finally, the normal vector of each cluster was computed by principal component analysis to obtain the leaf normal distribution. We collected leaf inclination angles and estimated the distribution, which we compared with the retrieved leaf normal distribution. The correlation coefficient between measurements and obtained results was 0.96, indicating a good coincidence.

Risk transmission evaluation for parallel construction of warships based on IFCM and the cloud model

To cope with multi-directional transmission coupling,spreading, amplification, and chain reaction of risks during multiproject parallel construction of warships, a risk transmission evaluation method is proposed, which integrates an intuitionistic cloud model with a fuzzy cognitive map. By virtue of expectancy Ex, entropy En, and hyper entropy He, the risk fuzziness and randomness of the knowledge of experts are organically combined to develop a method for converting bi-linguistic variable decision-making information into the quantitative information of the intuitionistic normal cloud(INC) model. Subsequently, the threshold function and weighted summation operation in the traditional fuzzy cognitive map is converted into the INC ordered weighted averaging operator to create the risk transmission model based on the intuitionistic fuzzy cognitive map(IFCM) and the algorithm for solving it. Subsequently, the risk influence sequencing method based on INC and the risk rating method based on nearness are proposed on the basis of Monte Carlo simulation in order to realize the mutual conversion of the qualitative and quantitative information in the risk evaluation results.Example analysis is presented to verify the effectiveness and practicality of the methods.

Study on the Scope of Small Amplitude of Cloud-to-ground Lightning in Lightning Weather Monitoring

Based on lightning location data in Chongqing region during 1999-2008,the frequency of lightning in various amplitude ranges and its annual variations were analyzed firstly.Afterwards,with the aid of matlab mathematical software,the distribution of the lightning location data was fitted using logarithmic normal distribution function.The results showed that after data of cloud-to-ground lightning with current amplitude from-5 to5 kA were deleted from lightning location data,the statistical characteristics of cloud-to-ground lightning could be reflected well.Meanwhile,lightning with current amplitude from-5 to 5 kA accounted for 1.05%（less than 2%）,which accorded with the detection principle of lightning position indicator（there existed error detection）.Therefore,cloud-to-ground lightning with current amplitude of-5-5 kA in lightning location data of Chongqing region was defined as small amplitude of cloud-to-ground lightning,which could provide scientific references for the processing of lightning location data in Chongqing region as well as analysis and quality control of lightning location data in other regions.

基于组合赋权与灰云模型的综合能源系统需求响应效益评价

考虑到当前综合能源系统研究缺乏完善的需求响应量化评估方案,难以在优化运行层面综合评估需求响应效益这一问题,提出了一种基于改进AHP群决策—CRITIC组合赋权与灰云模型的综合能源系统需求响应效益评价方法。首先,从综合能效性、社会经济性和需求侧互动性三个维度构建了需求响应效益评价指标体系。其次,采用一致性和权重拟合性更优的指数标度法构造判断矩阵,降低赋值误差,并通过AHP群决策法确定主观权重,从而削弱主观极值偏差对权重的影响;在由CRITIC法确定客观权重后,基于最小欧氏距离建立组合权重模型,并通过非线性规划求取最优组合权重;针对评价等级信息的模糊性与隶属等级的随机性,采用正态灰云白化权模型确定指标分类等级与场景评分。最后,以北方某综合能源系统为例,根据用户参与需求响应方式设置了4种运行场景,分析了不同需求响应对系统运行的影响与作用,算例结果表明所提指标体系与评价方法科学有效。

基于法线微分的3维声呐点云自适应简化方法

在不损失原始点云数据质量的前提下,大幅约简点云数据量是减少存储空间、降低后期计算强度的重要预处理步骤。针对这一需求,提出了针对水下3维声呐点云数据的自适应简化方法。首先,定义法线微分算子来识别点云中几何尺度的骤变,从而实现原始点云中边界部分点云和主体部分点云的分割。其次,对于点云的边界部分,应用移动最小二乘法来对边界点进行优化,降低噪点的影响,并保持其曲面的几何一致性;基于体素栅格结构,在边界上使用八叉树进行降采样,并在此基础上实施局部最远点采样,在实现均匀简化的同时保证已简化点云的边界部分具有各向同性,有效保留边界部分点云的几何特征信息。再次,对于点云的主体部分,为保持简化后点云整体的各向同性,使用体素中心采样法来减少数据量。然后,通过高斯滤波平滑点云表面,最后,整合简化后的边界点云和主体点云,得到简化结果。实验结果表明,提出的简化方法计算成本低、处理速度快,在与现行典型算法保持一致简化率的情况下,对水下点云数据的简化速度提高了约32%。另外,通过表面密度对比与几何失真分析,证明了提出方法对水下3维点云边界点及整体分布的优化作用。综上,此方法能提高水下作业目标探测效率,得到保留重要几何特征信息并具有各向同性的水下任务目标点云简化结果。

结合超体素与颜色信息的区域生长点云分割方法

为解决传统区域生长点云分割算法存在的欠分割和过分割现象,提出一种结合超体素与颜色信息的区域生长点云分割方法。在分割过程中加入超体素过分割步骤,避免直接从点云中分割数据,有效消除噪声和异常值对分割的影响,利用一种几何和颜色信息的联合准则合并超体素并进行区域生长。与深度学习方法和其它3种传统分割算法相比,分割效率和精度都得到了较大提升,解决了欠分割和过分割的问题。

基于三维激光点云数据的电力电缆绝缘缺陷识别

针对电力电缆绝缘缺陷识别精准度较低的问题,提出一种基于三维激光点云数据的电力电缆绝缘缺陷识别方法。利用剪裁法增强原始三维激光点云数据,通过区域生长法与最小二乘法完整获取电缆绝缘材料的三维结构面信息。借助Canny边缘检测方法求解电缆绝缘表面缺陷与内部缺陷边缘信息,自动识别出电力电缆绝缘的缺陷位置及缺陷类别。结果表明,所提方法可以精准识别电缆绝缘表面的划痕缺陷、电缆外屏蔽表面起泡和孔洞缺陷,识别耗时短,鲁棒性较优,具有较高实际应用价值。

基于点云数据的大曲率工件表面法向计算方法研究

航空航天构件装配制孔时对垂直度有严格要求,采用机器人等自动设备制孔时,为保证垂直度需要对每一个制孔位置的法向进行在位测量。现有的工件表面法向测量方式难以适应大曲率工件表面的测量需求。通过线激光扫描,可以获得反映工件局部表面详细信息的点云数据,在此基础上提出一种大曲率工件表面法向计算方法。首先研究了通过主成分分析(PCA)对局部点云数据进行平面拟合获得曲面法向的方法,然后选取了若干典型曲率,通过仿真生成点云数据,并分析了法向拟合误差的变化规律。通过试验验证了上述方法的有效性,结果表明,当点云选择范围在12mm以下时,直径为50mm的圆柱试验件的法向拟合误差可以保证在0.219°以下。

基于二维正态分布的无人机激光雷达点云匹配研究

无人机激光雷达点云特征多,一次匹配花时间长,难以进行二次匹配,因此研究基于二维正态分布的无人机激光雷达点云匹配方法。采集无人机激光雷达点云图像,通过旋转平移方法、双边滤波方法对图像预处理,利用二维正态分布算法和动态时间规整算法完成点云特征提取,使用初始变换矩阵估计算法对点云进行粗匹配,再使用近点迭代算法进行点云快速精匹配,通过两次匹配实现无人机激光雷达点云快速匹配。实验结果表明,所提方法的无人机激光雷达点云图像去噪效果好,点云匹配时间短,匹配偏差仅在0.04 m-0.15 m之间,匹配精度达到了相关预期。

基于法向量夹角的果树点云配准与枝叶分割方法研究

在实现果园作业全自动化的过程中,亟需直接构建自然环境下果树枝干三维模型的方法。本文通过对自然环境下以不同角度采集的果树点云进行配准,并针对采样一致性(SAC-IA)+迭代最近点(ICP)配准算法在点云配准中耗时较长以及精度不高的问题,结合点云法向量夹角提取源点云和目标点云的特征点,并通过点云法向量夹角的余弦值在源点云和目标点云的特征点中查找待匹配点对的方法,提出了一种基于果树点云待匹配点对的改进SAC-IA+ICP点云配准算法;借助最小包围盒划分的分块技术对配准后的果树点云进行分块,然后利用点云的几何特征,对划分的子块进行枝叶粗分割,最后使用欧氏聚类完成枝叶的精细分割。对比实验结果显示,改进后的SAC-IA+ICP算法在平均旋转误差上相较于原始SAC-IA+ICP算法减少85.44%,配准均方根误差相较于原始SAC-IA+ICP算法减少71.74%,配准时间相较于原始SAC-IA+ICP算法减少97.99%;同时,改进后的SAC-IA+ICP算法在平均旋转误差上相较于SAC-IA+NDT算法减少90.38%,配准均方根误差相较于SAC-IA+NDT算法减少85.39%,配准时间相较于SAC-IA+NDT算法减少98.04%。另外,本文采用的枝叶分割算法能够完成枝叶分割,且相较于人工分割其分割准确度可达94.77%。

基于三维重建的静爆场破片检测方法

针对靶场静爆试验中破片统计分析工作效率低的问题,提出一种基于多视图三维重建的破片检测方法。引入点云分割中的自适应阈值,有效保留三维点云的细节信息,增强了算法对靶场不同场景应用的泛化能力。针对大数量点云处理效率不佳的问题,采用点云法向量预先验证的方法,再结合颜色的区域生长分割算法,有效降低了检测误差和漏检概率,提升了破片检测的精度。实验结果表明:文中提出的检测方法检测准确率达到95%,破片弹着点误差小于4%,可以有效地分割出破片的着靶参数,为后续破片的统计分析工作提供了实验依据。

点云与图像融合的无监督异常检测算法研究

针对多模态工业检测中高维特征之间存在干扰,导致检测率不理想的问题,提出了一种基于标准化流的多模态工业异常检测方法。首先提取图像的3D点云的深度信息将其作为第4个通道添加到RGB图像中,生成融合后的RGBD图像,然后使用预训练的特征提取网络提取融合后的图像特征,最后使用特征训练得到一个用于异常检测的标准化流模型。实验结果表明,异常检测模型在MVTec 3D-AD数据集的平均Pixel AUROC达到95.8%,平均AUPRO达到86.2%。相较于其他模型分别提升了2.6%和9.1%。

基于差动共焦法向测量配准的自由曲面测量方法

针对自由曲面设计坐标系和测量坐标系基准不同带来的测量问题,提出了一种基于差动共焦法向测量配准的自由曲面测量方法。方法利用差动共焦曲线过零点位置与被测点准确对应的特性,实现了对被测点的精确定焦;利用PSD在传感器焦点位置对被测点倾角进行精确测量;根据被测点倾角,利用ICP Point to Plane方法进行点云配准,提高了点云配准精度,显著提升了面形测量精度。经理论分析和初步实验证明,方法测量面形PV值的精度可以达到65nm,RMS值达到10nm,PV值与zygo干涉仪测量结果相差7nm,RMS值相差1nm,测量结果误差较小。该方法为自由曲面面形的高精度测量提供了一种新的技术途径。

基于大数据挖掘的激光点云边界法向矢量提取方法

激光点云边界法向矢量提取过程中点云边界数据难以精准配准,导致提取精度和效率较差。为了更加准确的对激光扫描信息实施分析,提出了基于大数据挖掘的激光点云边界法向矢量提取方法。该方法首先通过摄像机对激光点云实施标定,获取激光点云图像,通过主成分分析、曲面拟合以及滤波算法对获取图像实施去噪处理,最后对图像实施灰度化以及高斯滤波处理,同时结合大津阈值法提取图像中心点法线方向上的灰度中心,以此实现对激光点云边界法向矢量的提取。经试验验证,所提方法提取的法向矢量夹角特征与理想状态下的夹角特征基本一致,且提取效率高、迭代误差小。

基于熵权-正态云模型的小净距隧道施工风险评估模型

随着我国基础设施建设的大力发展,小净距隧道在我国高速公路建设过程中的占比越来越高。由于地质条件的原因,部分小净距隧道可能会穿越岩堆体,而影响其施工安全的风险因素繁多、关系模糊复杂、风险因子不确定性大,导致风险发生概率存在各种不确定性。针对穿越岩堆体的小净距隧道,提出一种基于熵权-正态云模型的小净距隧道施工风险评估模型,选取工程地质条件、隧道设计、外部因素和工程管控等4个方面的15个评估指标,并以某实际隧道工程为例,论证了该模型在小净距隧道施工风险定量评估中应用的可行性和独特优势。该研究对其他类似工程施工的风险评估有一定的借鉴意义。

多策略融合的改进哈里斯鹰算法及其路径规划应用

针对哈里斯鹰算法(HHO)后期缺乏全局搜索、易陷入局部极值等问题,提出了一种多策略融合的改进哈里斯鹰算法(MFHHO).首先,提出了自适应混沌和核心种群划分策略,平衡算法后期的全局和局部搜索性能,提高算法的多样性;其次,修改位置更新公式,将莱维飞行替换为黄金正弦策略,提高算法寻优性能和效率;最后,融合自适应正态云最优解扰动策略,提高算法跳出局部最优解的能力.在国际测试函数、CEC2014复杂函数的数值实验中,改进算法性能提升明显;在路径规划实验中,改进算法性能最优,相较于对照组算法性能提升了14.75%,且具有较好的稳定性.