激光点阵图像特征点提取算法研究

  • 打印
  • 收藏
收藏成功


打开文本图片集

关键词:激光点阵图像;阈值分割;特征点提取

中图分类号:TP317.4 文献标志码:A 文章编号:1003-5168(2025)08-0048-05

DOI:10.19968/j.cnki.hnkj.1003-5168.2025.08.008

Abstract: [Purposes] In order to solve the problem that SIFT algorithm extracts multiple feature points from aspot area in the process of feature point extraction of laser latice image,which leads to the inability to accurately extract the center coordinates of the spot,a feature point extraction algorithm of laser lattice image is proposed.[Methods] The algorithm uses Gaussian filtering for noise reduction.Firstly,an adaptive threshold segmentation algorithm based on area and quantity constraints is used to binarize the image;secondly,the contour search algorithm is used to process the binary image ;then,thirdly ellipse fitting of the spot contour is carried out,and the geometric shape ofthe fitting result is screened to obtain the spot center coordinates; finally,the simulation experiment is carried out to select the appropriate Gaussian filter parameters.[Findings] The proposed algorithm and SIFT algorithm were used to extract feature points from laser dot matrix images.The experimental results show that the proposed algorithm can accurately extract 2 6O1 feature points from 5 1 × 5 1 laser dot matrix images.[Conclusions] In this paper,only one feature point is extracted from a laser region,and the interference of environmental fea ture points is eliminated. Keywords: laser dot matrix image; threshold segmentation; feature point extraction

0 引言

激光点阵图像处理是研究激光散斑测量技术的关键,精确定位激光特征点坐标是高精度三维重建的基础。(剩余5329字)

目录
monitor