Research on Point Cloud Repair Based on Multi-parameter Extraction in Digital Power Plant

2020 5th International Conference on Power and Renewable Energy (ICPRE) Pub Date : 2020-09-12 DOI:10.1109/ICPRE51194.2020.9233241

Chen Hui, Cui Wen, Cui Chenggang, Hu Yunfeng

{"title":"Research on Point Cloud Repair Based on Multi-parameter Extraction in Digital Power Plant","authors":"Chen Hui, Cui Wen, Cui Chenggang, Hu Yunfeng","doi":"10.1109/ICPRE51194.2020.9233241","DOIUrl":null,"url":null,"abstract":"The 3D visualization of the traditional power plant can integrate the information in the scene and manage each area accurately and quickly. The surface missing parts and holes in the reconstruction process of 3D visualization power plant can lead to the loss of information. To solve this problem, an efficient holes repairing algorithm is proposed in this paper. Through the feature discrimination of the original data to simplify processing, and then repair the hole. In the case of retaining the characteristics of the original data, the subsequent repairing efficiency is improved by reducing the number of point clouds. After fast triangulation of the discrete data, a complete network is formed by least square fitting, and then the curvature change of the surface is minimized to make it follow the trend of curvature change around the hole. The experimental results show that this method can greatly improve the repair efficiency of point cloud hole data with a large number of point clouds, and retain its original features.","PeriodicalId":394287,"journal":{"name":"2020 5th International Conference on Power and Renewable Energy (ICPRE)","volume":"13 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2020-09-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2020 5th International Conference on Power and Renewable Energy (ICPRE)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICPRE51194.2020.9233241","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}

引用次数: 1

Abstract

The 3D visualization of the traditional power plant can integrate the information in the scene and manage each area accurately and quickly. The surface missing parts and holes in the reconstruction process of 3D visualization power plant can lead to the loss of information. To solve this problem, an efficient holes repairing algorithm is proposed in this paper. Through the feature discrimination of the original data to simplify processing, and then repair the hole. In the case of retaining the characteristics of the original data, the subsequent repairing efficiency is improved by reducing the number of point clouds. After fast triangulation of the discrete data, a complete network is formed by least square fitting, and then the curvature change of the surface is minimized to make it follow the trend of curvature change around the hole. The experimental results show that this method can greatly improve the repair efficiency of point cloud hole data with a large number of point clouds, and retain its original features.

查看原文本刊更多论文

基于多参数提取的数字电厂点云修复研究

传统电厂的三维可视化可以整合现场信息，对各个区域进行准确、快速的管理。在三维可视化电厂重建过程中，表面缺失的零件和孔洞会导致信息的丢失。为了解决这一问题，本文提出了一种有效的孔洞修复算法。通过特征判别对原始数据进行简化处理，再进行补孔。在保留原始数据特征的情况下，通过减少点云的数量，提高后续修复效率。对离散数据进行快速三角剖分后，通过最小二乘拟合形成一个完整的网络，然后将表面的曲率变化最小化，使其符合孔周围曲率变化的趋势。实验结果表明，该方法可以大大提高具有大量点云的点云孔数据的修复效率，并保持其原有特征。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

2020 5th International Conference on Power and Renewable Energy (ICPRE)

自引率

0.00%

发文量