基于约束流体化的激光图像特征三维重建算法（英文）  

Three dimensional reconstruction algorithm of laser image feature based on constrained fluid

作　　者：赵宇枫 徐科[3] Yu-feng ZHAO Ke XU(School of Computer, Ctmngqing University, Chongqing 401331, China Informatton Engineering School, Chongqing Industrial Vocational and Technical College, Chongqing 401120 lnstitute of Computer Science, South-Central University for Nationalities, Wuhan 430074, China)

机构地区：[1]重庆大学计算机学院,重庆401331 [2]重庆工业职业技术学院信息工程学院,重庆401120 [3]中南民族大学计算机科学学院,武汉430074

出　　处：《机床与液压》2017年第18期34-40,共7页Machine Tool & Hydraulics

基　　金：supported by doctoral program of the Ministry of education of China(No:20121101110037)

摘　　要：针对当前的三维重建方法一直存在无法真实体现激光图像的三维特征、且匹配精度较差等缺点,提出基于约束流体化的激光图像特征三维重建方法。结合激光图像的特有特征设计特征点检测算法,提取激光图像特征点,采用最优二叉树法对激光图像特征进行匹配,运用非线性流体分析方法获取激光图像特征三维重建所需参数矩阵;引入不同约束权重比例法对约束权重系数进行选择;实现对激光图像特征的三维重建。实验结果证明:采用该算法充分利用图像特征点进行激光图像的三维重建,其重建效率高,时间短,与原图的匹配度较高,具有一定的实用性。According to the detects that current 3D reconstruction method has been unable to truly reflect the three-dimensional characteristics of the laser image, and poor matching accuracy and so on, 3D reconstruction method based on constraint fluidize laser image characteristics was proposed in this paper. By adopting Harris cor- ner detection algorithm, we extracted the laser image feature points, and by the method of the optimal binary tree, we had the laser image feature matching and used the nonlinear fluid analysis method to get parameter matrix of 3D reconstruction for the laser image characteristics; Using different constraint weight proportion method, we se- lected the constraint weight coefficient and realized 3D reconstruction for the laser image characteristics. The ex- perimental results show that the 3D reconstruction of image feature points makes full use of laser image by using this algorithm, and the reconstruction time is short, and it has high efficiency and high matching degree of original image for the industrial application.

关 键 词：激光图像 特征算法 三维重建 约束流体化 图像特征 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]