摘要
点云数据压缩时,外边界点易丢失,影响后续点云压缩品质。为此,提出激光测距前沿回波大数据点云压缩方法。首先通过主回波匹配、回波信号识别、多项式拟合滤波、时间误差修正等操作预处理激光测距前沿回波大数据;然后采用主成分分析法获取数据法矢,再分别提取边界特征点和尖锐特征点,保留点云模型完整边界;最后利用SIFT算法压缩点云数据并保存,完成激光测距前沿回波大数据的点云压缩。实验结果表明,所提方法能够提高压缩率、缩短压缩时间、降低压缩前后表面积比。
When point cloud data is compressed,the outer boundary points are easily lost,which affects the compression quality of subsequent point clouds.Therefore,a cloud compression method for large data points of laser ranging front echo is proposed.Firstly,the front-edge echo data of laser ranging is processed by main echo matching,echo signal recognition,polynomial fitting filtering and time error correction.Then the principal component analysis method is used to obtain the data normal vector,and then the boundary feature points and sharp feature points are extracted respectively,and the complete boundary of the point cloud model is retained.Finally,SIFT algorithm is used to compress the point cloud data and save it,and the point cloud compression of laser ranging front echo big data is completed.The experimental results show that the proposed method can improve the compression rate,shorten the compression time and reduce the surface area ratio before and after compression.
作者
杨道平
王亚
唐晔
YANG Daoping;WANG Ya;TANG Ye(School of information engineering,Zunyi Normal college,Zunyi Guizhou 563006,China)
出处
《激光杂志》
CAS
北大核心
2023年第3期252-256,共5页
Laser Journal
基金
国家自然科学基金(No.61562094)
贵州省科技合作计划项目(No.黔科合J字LKZS[2019]25号)。
关键词
激光测距
前沿回波
大数据
点云压缩
特征点提取
laser ranging
frontier echo
big data
point cloud compression
feature point extraction
