基于谱图理论的变压器区域大规模点云轻量化方法

doi:10.19595/j.cnki.1000-6753.tces.231987

摘要
图/表
参考文献
相关文章 (3)

全文: PDF (4780 KB) HTML
输出: BibTeX | EndNote (RIS)

摘要点云在电网数字化转型中有重要应用价值。现有方法在处理变压器区域大规模点云时难以兼顾点云轻量化后的视觉效果,导致轻量化点云存在较大视觉失真。该文在点云轻量化时引入视觉失真度,以其最小作为约束,实现点云轻量化后视觉效果最优。首先基于谱图理论将原始点云转换为图信号,建立轻量化点云视觉信息损失与重采样矩阵间的函数关系;然后以视觉信息损失最小为目标函数,获得一组满足视觉信息损失最小但特征信息和均匀信息比例不同的轻量化点云;进一步,将原始点云和轻量化点云投影到几何和颜色特征域,用一个低维向量表征点云视觉效果,从而选择出视觉失真度最小的轻量化点云;最后,使用基准数据集和包含8 000余万个点的变压器区域大规模点云进行了验证。结果表明：所提方法与主流的随机降采样法、体素平均法、非均匀网格法、曲率采样法相比,在轻量化点云的视觉效果方面,分别提升了57.4%、69.2%、62.2%、75.6%。

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	杨帆
	吴涛
	郝翰学
	刁冠勋
	李勇

关键词 ：谱图理论, 点云, 轻量化, 视觉失真, 变压器区域

Abstract：The three-dimensional visualization based on point clouds has important application value in the digital and intelligent transformation of the power industry. For transformers, the actual production often obtains the point cloud of their entire operating area, and the huge data scale brings difficulties to use. Existing lightweight methods for processing large-scale point clouds always lead to significant visual distortion. Therefore, this paper proposes anew point cloud lightweight method that can take visual effects into account.
Firstly, the K-nearest neighbor (KNN) algorithm establishes an edge set for the original point cloud and converts it into a graph signal. Then, the graph signal is transformed into the frequency domain through the graph Fourier transform. Further, the feature and uniformity loss expressions during point cloud lightweight are derived. After quantifying the visual information loss, an objective function was established to minimize the loss. Then, the grid search was performed on parameters k and ρ, which affect the proportion of features and uniformity in the lightweight point cloud. A set of lightweight point clouds that meet the minimum visual information loss but have different proportions of features and uniformity can be obtained. A visual distortion quantification criterion was established to select the lightweight point cloud with the lowest visual distortion. The criteria project the 3D point cloud onto a feature domain composed of geometric and color features closely related to visual effects, and convert the visual effects of the point cloud into multiple histograms. Furthermore, statistical parameters quantify each feature's histograms, and the point cloud's visual effects are transformed into vectors. The visual distortion was calculated using the difference between vectors, and the lightweight point cloud with the lowest visual distortion was selected. Finally, the effectiveness of the proposed method was validated using a benchmark dataset and a large-scale point cloud of transformer areas containing over 80 million points. Based on the visual scores calibrated on the benchmark dataset, the correctness of the established visual distortion quantification criteria was first verified, further validating the effectiveness of the proposed method. For a large-scale colored point cloud in a specific transformer area, the proposed method has improved the visual effect of the lightweight point cloud by 57.4%, 69.2%, 62.2%, and 75.6% compared to mainstream random down-sampling, voxel averaging, non-uniform grid, and curvature sampling methods, respectively.
The following conclusion can be drawn: the visual effect of lightweight point clouds is not only affected by visual information loss but also by the proportion of feature information and uniformity in the lightweight point cloud. It is necessary to further search for the optimal proportion while minimizing visual information loss to obtain the lightweight point cloud with the lowest visual distortion. Projection of point clouds onto geometric and color feature domains related to visual effects can quantitatively describe the visual distortion of lightweight point clouds and thus select the lightest point cloud with the best visual effect. Compared with mainstream methods, the lightweight point cloud obtained by the proposed method has better visual effects while retaining the same number of points.

Key words： Spectral graph theory point cloud lightweight visual distortion power transformer area

收稿日期: 2023-11-27

PACS:

TM507

基金资助:国家电网公司总部科技项目资助（5500-202017468A-0-0-00）

通讯作者: 吴涛男, 1993年生, 博士研究生, 研究方向为输变电设备多物理场计算、电力装备数字孪生。E-mail：Bryant_Mentality@163.com

作者简介: 杨帆男, 1980年生, 教授,博士生导师,研究方向为输变电设备多物理场计算、电力装备数字孪生。E-mail：yangfan@cqu.edu.cn

引用本文:

杨帆, 吴涛, 郝翰学, 刁冠勋, 李勇. 基于谱图理论的变压器区域大规模点云轻量化方法[J]. 电工技术学报, 2024, 39(23): 7528-7541. Yang Fan, Wu Tao, Hao Hanxue, Diao Guanxun, Li Yong. Large Scale Point Cloud Lightweight Method for Power Transformer Area Based on Spectral Graph Theory. Transactions of China Electrotechnical Society, 2024, 39(23): 7528-7541.

链接本文:

https://dgjsxb.ces-transaction.com/CN/10.19595/j.cnki.1000-6753.tces.231987 https://dgjsxb.ces-transaction.com/CN/Y2024/V39/I23/7528

[1] 朱继忠, 骆腾燕, 吴皖莉, 等. 综合能源系统运行可靠性评估评述Ⅰ:模型驱动法[J]. 电工技术学报, 2022, 37(11): 2761-2776.
Zhu Jizhong, Luo Tengyan, Wu Wanli, et al.A review of operational reliability assessment of integrated energy systemsⅠ: model-driven method[J] Transactions of China Electrotechnical Society, 2022, 37(11):2761-2776.
[2] 江悦, 沈小军, 吕洪, 等.碱性电解槽运行特性数字孪生模型构建及仿真[J]. 电工技术学报, 2022, 37(11): 2897-2908.
Jiang Yue, Shen Xiaojun, Lü Hong, et al.Construction and simulation of operation digital twin model for alkaline water electrolyzer[J]. Transactions of China Electrotechnical Society, 2022, 37(11): 2897-2908.
[3] 高树国, 汲胜昌, 孟令明, 等. 基于在线监测系统与声振特征预测模型的高压并联电抗器运行状态评估方法[J]. 电工技术学报, 2022, 37(9): 2179-2189.
Gao Shuguo, Ji Shengchang, Meng Lingming, et al.Operation state evaluation method of high-voltage shunt reactorbased on on-line monitoring system and vibro-acoustic characteristic prediction model[J]. Transactions of China Electrotechnical Society, 2022, 37(9):2179-2189.
[4] 田勇, 杨昊, 胡超, 等. 基于毫米波雷达的电动汽车无线充电运动异物检测与跟踪[J]. 电工技术学报, 2023, 38(2): 297-308.
Tian Yong, Yang Hao, Hu Chao, et al.Moving foreign object detection and track for electric vehicle wireless charging based on millimeter wave radar[J]. Transactions of China Electrotechnical Society, 2023, 38(2): 297-308.
[5] 谢庆, 蔡扬, 谢军, 等. 基于ALBERT的电力变压器运维知识图谱构建方法与应用研究[J].电工技术学报, 2023, 38(1): 95-106.
Xie Qing, Cai Yang, Xie Jun, et al.Research on construction method and application of knowledge graph for power transformer operation and maintenance based on ALBERT[J]. Transactions of China Electro-technical Society, 2023, 38(1): 95-106.
[6] 王小虎, 郭广鑫, 董佳涵, 等. 变电站应用实景复制技术建模和网络安全监控[J]. 中国电力, 2021, 54(11): 221-228.
Wang Xiaohu, Guo Guangxin, Dong Jiahan, et al.Substation modelling and network security monitoring based on scene replication technology[J]. Electric Power, 2021, 54(11): 221-228
[7] 刘云鹏, 刘一瑾, 刘刚, 等. 电力变压器智能运维的数字孪生体构想[J]. 中国电机工程学报, 2023, 43(22): 8636-8652.
Liu Yunpeng, Liu Yijin, Liu Gang, et al.Digital twin conception of intelligent operation and maintenance of power transformer[J]. Proceedings of the CSEE, 2023, 43(22): 8636-8652.
[8] 宋立业, 刘帅, 王凯, 等. 基于改进EfficientDet的电网元件及缺陷识别方法[J]. 电工技术学报, 2022, 37(9): 2241-2251.
Song Liye, Liu Shuai, Wang Kai, et al.Identification method of power grid components and defects based on improved EfficientDet[J]. Transactions of China Electrotechnical Society, 2022, 37(9):2241-2251.
[9] 王伟杰, 雍明超, 黄金魁, 等. 高压设备数字孪生体构建及状态分析技术研究[J]. 高压电器, 2023, 59(11): 119-128.
Wang Weijie, Yong Mingchao, Huang Jinkui, et al.Research on construction of condition analysis technology of digital twin for highvoltage equipment[J]. High Voltage Apparatus, 2023,59(11): 119-128.
[10] 刘衍, 邹阳, 谭舒宁, 等. 基于地面激光雷达的隔离开关合闸状态自动检测方法[J]. 高压电器, 2022, 58(6): 193-198.
Liu Yan, Zou Yang, TanShuning, et al. Automatic detection method of closing state for disconnector based on ground lidar[J]. High Voltage Apparatus, 2022, 58(6): 193-198.
[11] Lipman Y, Cohen-Or D, Levin D, et al.Parameterization-free projection for geometry reconstruction[J]. ACM Transactions on Graphics, 2007, 26(99): 22.
[12] Gezawa A S, Bello Z A, Wang Qicong, et al.A voxelized point clouds representation for object classification and segmentation on 3D data[J]. The Journal of Supercomputing, 2022, 78(1): 1479-1500.
[13] Wang Shuaiqing, Hu Qijun, Xiao Dongsheng, et al.A new point cloud simplification method with feature and integrity preservation by partition strategy[J]. Measurement, 2022, 197: 111173.
[14] Chen Hui, Cui Wen, Bo Caihui, et al.Point cloud simplification for the boundary preservation based on extracted four features[J]. Displays, 2023, 78: 102414.
[15] Lü Chenlei, Lin Weisi, Zhao Baoquan.Voxel structure-based mesh reconstruction from a 3D point cloud[J]. IEEE Transactions on Multimedia, 2022, 24: 1815-1829.
[16] Yang Lina, Li Yuchen, Li Xichun, et al.Efficient plane extraction using normal estimation and RANSAC from 3D point cloud[J]. Computer Standards & Interfaces, 2022, 82: 103608.
[17] Cheng Yuqi, Li Wenlong, Jiang Cheng, et al.A novel point cloud simplification method using local conditional information[J]. Measurement Science and Technology, 2022, 33(12): 125203.
[18] Lü Chenlei, Lin Weisi, Zhao Baoquan.Intrinsic and isotropic resampling for 3D point clouds[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2023, 45(3): 3274-3291.
[19] Xuan Wei, Hua Xianghong, Chen Xijiang, et al.A new progressive simplification method for point cloud using local entropy of normal angle[J]. Journal of the Indian Society of Remote Sensing, 2018, 46(4): 581-589.
[20] Wei Pengcheng, Yan Li, Xie Hong, et al.Automatic coarse registration of point clouds using plane contour shape descriptor and topological graph voting[J]. Automation in Construction, 2022, 134: 104055.
[21] Song Hao, Feng H Y.A progressive point cloud simplification algorithm with preserved sharp edge data[J]. The International Journal of Advanced Manufacturing Technology, 2009, 45(5): 583-592.
[22] Hu Qingyong, Yang Bo, Xie Linhai, et al.Learning semantic segmentation of large-scale point clouds with random sampling[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2022, 44(11): 8338-8354.
[23] Qi Junkun, Hu Wei, Guo Zongming.Feature preserving and uniformity-controllable point cloud simplification on graph[C]//2019 IEEE International Conference on Multimedia and Expo (ICME), Shanghai, China, 2019: 284-289.
[24] Hackel T, Wegner J D, Schindler K.Fast semantic segmentation of 3d point clouds with strongly varying density[J]. ISPRS Annals of Photogrammetry, Remote Sensing and Spatial Information Sciences, 2016, III3: 177-184.
[25] Ortega A, Frossard P, Kovačević J, et al.Graph signal processing: overview, challenges, and applications[J]. Proceedings of the IEEE, 2018, 106(5): 808-828.
[26] Zhang Zicheng, Sun Wei, Min Xiongkuo, et al.No-reference quality assessment for 3D colored point cloud and mesh models[J]. IEEE Transactions on Circuits and Systems for Video Technology, 2022, 32(11): 7618-7631.
[27] Weinmann M, Schmidt A, Mallet C, et al.Contextual classification of point cloud data by exploiting individual 3d neigbourhoods[J]. ISPRS Annals of Photogrammetry, Remote Sensing and Spatial Information Sciences, 2015, II3: 271-278.
[28] Weinmann M, Jutzi B, Mallet C, et al. Geometric features and their relevance for 3d point cloud classification[J]. ISPRS Annals of Photogrammetry, Remote Sensing and Spatial Information Sciences, 2017, Ⅳ-1/W1: 157-164.
[29] Connolly C, Fleiss T.A study of efficiency and accuracy in the transformation from RGB to CIELAB color space[J]. IEEE Transactions on Image Processing: a Publication of the IEEE Signal Processing Society, 1997, 6(7): 1046-1048.
[30] Su H L.Waterloo-Point-Cloud-Database[EB-OL].[2019].https://github.com/qdushl/Waterloo-Point-Cloud-Database.
[31] Liu Qi, Su Honglei, Duanmu Zhengfang, et al.Perceptual quality assessment of colored 3D point clouds[J]. IEEE Transactions on Visualization and Computer Graphics, 2023, 29(8): 3642-3655.