近点光源照射下球面坐标PSFS三维形貌重建方法  

PSFS-Based 3D Shape Recovery Method with Near Point Light Source Based on Spherical Coordinates

作　　者：王国珲[1] 谢倩 WANG Guohui;XIE Qian(School of Optoelectronic Engineering,Xi’an Technological University,Xi’an 710021,China)

机构地区：[1]西安工业大学光电工程学院,西安710021

出　　处：《西安工业大学学报》2018年第1期14-19,34,共7页Journal of Xi’an Technological University

基　　金：陕西省自然科学基础研究计划项目(2016JM6041);西安工业大学陕西省光电测试与仪器技术重点实验室开放基金(2016SZSJ-60-1);国家自然科学基金(61102144)

摘　　要：为了解决基于三维笛卡尔坐标系建立的透视投影SFS重建方法中图像辐照度方程复杂等问题,文中提出了基于球面坐标的近点光源PSFS三维形貌重建方法.在透视投影和球面坐标系下,构建了近点光源照射下物体三维形貌与图像灰度相对应的偏微分辐照度方程.将图像辐照度方程转换为Eikonal类偏微分方程,利用改进的FMM方法对方程进行了求解.研究结果表明:与基于三维笛卡尔坐标的PSFS方法相比,基于球面坐标的PSFS方法建立了求解相对简便的图像辐照度方程,可获得较高的重建精度,其高度平均相对误差降低了0.02%.In the perspective SFS recovery methods based on the 3D Cartesian coordinate system,the image irradiance equation is very complicated.To solve this problem,the paper presents a PSFS based 3D shape recovery method with a near point light source based on spherical coordinates.The partial differential irradiance equation between the 3D shape illuminated by a near light source and image gray is constructed under the perspective projection and spherical coordinate system.The image irradiance equation is transformed into an Eikonal partial differential equation,and then solved by an improved FMM.The research results demonstrate that compared with the PSFS method based on 3D Cartesian coordinates,the proposed method can establish a relatively simple image irradiance equation and can obtain a higher recovery accuracy whose mean relative error of the height is reduced by 0.02%.

关 键 词：PSFS 球面坐标 近点光源 三维形貌重建 偏微分辐照度方程 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]