摘要
基于逆向设计中点云处理的表面识别问题,本文提出了一种基于小波变换的离散点云数据的特征识别算法。首先将离散点云表示成小波变换可以处理计算的形式,然后在此基础上提出了具体的二维和三维离散点云的小波分解算法,最后引入实例,对二维离散点云的小波分解算法进行验证分析。实验结果表明本文提出的算法达到了对点云数据的特征分解的目的。将离散点云数据按特征分解从而提取出不同的特征成分,可以根据后期点云预处理的不同要求,将小波变换后的数据进行进一步的处理。
Because of the surface recognition problem in cloud process in reverse design, this paper presents a feature recognition algorithm for discrete cloud points based on wavelet transformation. Firstly the cloud data is denoted that the wavelet transformation can deal with, and a detailed wavelet decomposition method for 2D and 3D discrete cloud points is represented. Lastly a sample is adopted to analyze and demonstrate the algorithm. The discrete cloud points are decomposed according feature recognition algorithm and different features are extracted and these decomposed cloud data can be further processed to meet different data pretreatment requirement.
出处
《电子测试》
2013年第11期53-56,共4页
Electronic Test
关键词
点云处理
离散点云
小波变换
特征识别
cloud process
discrete cloud
wavelet transform
feature recognition
