基于结构光的低反射工件三维重建方法研究  

Research of low-reflection workpieces 3D reconstruction method based on structured light

作　　者：贾军营 杨爽 杨海波 李子琦 JIA Junying;YANG Shuang;YANG Haibo;LI Ziqi(School of Information Science and Engineering,Shenyang University of Technology,Shenyang 110870,China)

机构地区：[1]沈阳工业大学信息科学与工程学院,沈阳110870

出　　处：《激光杂志》2025年第4期88-96,共9页Laser Journal

基　　金：辽宁省应用基础研究项目(No.2022JH2/101300243);辽宁省自然科学基金(面上)项目(No.2022-MS-438);沈阳市科学技术计划“揭榜挂帅”产业共性技术项目(No.22-316-1-07)。

摘　　要：结构光技术在工业三维视觉测量领域已得到广泛应用。由于表面反射率和环境光影响,低反射工件三维重建常常面临点云稀疏和平整度低等问题。因此,提出一种基于结构光的低反射工件三维重建改进方法。首先,采用双目结构光三维成像技术构建了一个双导向相移融合模型。该模型对低反射目标区域进行曝光导向的图像融合,并生成掩膜对其他目标区域进行模板导向的相位融合。此外,针对低反射目标区域的格雷码图像编码错误,提出了一种改进区域加权滤波算法,以消除编码错误并抑制级次跳变误差。经实验分析和验证,该方法相较于其他方法在点云密度上平均提高了31.07%,点云平整度达到了0.9997,显著提升了低反射工件三维重建质量和精度。Structured light technology has been widely used in industry for three-dimensional(3D)vision measurement.Due to the influence of surface reflectivity and ambient light,3D reconstruction of low-reflective workpieces often faces problems such as sparse point clouds and low flatness Therefore,an improved method for 3D reconstruction of low-reflective workpieces based on structured light is proposed.Firstly,a dual-guided phase shift fusion model was constructed through binocular structured light 3D imaging.Exposure-guided image fusion was applied to low-reflection target regions,and masks were generated to facilitate template-guided phase fusion in other target areas.Additionally,an improved region-weighted filtering algorithm was proposed to address the gray code image coding errors in low-reflection target regions,effectively eliminating coding errors and suppressing jump errors.After experimental analysis and verification,the method improves the point cloud density by 31.07%on average compared with other methods,with point cloud flatness reaching 0.9997.This study significantly improves the quality and accuracy of 3D reconstruction of low-reflective workpieces.

关 键 词：三维重建 结构光 低反射工件 格雷码 相位展开 

分 类 号：TN201[电子电信—物理电子学] TP391[自动化与计算机技术—计算机应用技术]