基于体素选择的自适应数字体积相关方法  

Self-adaptive digital volume correlation based on voxel selection

作　　者：袁山明 苏勇[1] 肖宇[1] 胡小方[1] 李永存 许峰[1] YUAN Shanming;SU Yong;XIAO Yu;HU Xiaofang;LI Yongcun;XU Feng(CAS Key Laboratory of Mechanical Behavior and Design of Materials,Department of Modern Mechanics,University of Science and Technology of China,Hefei 230o26,Anhui,China;National Demonstration Center for Experimental Mechanics Education,College of Mechanical and Vehicle Engineering,Taiyuan University of Technology,Taiyuan 030024,Shanxi,China)

机构地区：[1]中国科学技术大学中国科学院材料力学行为和设计重点实验室,安徽合肥230026 [2]太原理工大学机械与运载工程学院国家级力学实验教学示范中心,山西太原030024

出　　处：《实验力学》2023年第2期165-175,共11页Journal of Experimental Mechanics

基　　金：国家自然科学基金(12027901,12041202,11890683,U1832216)资助。

摘　　要：数字体积相关(Digital Volume Correlation,简称DVC)方法是一种非接触、非干涉的内部三维变形场测量方法,其基本原理是通过匹配物体变形前、后的体图像来测量变形,目前已成为实验力学领域最重要的内部变形测量手段之一。DVC方法计算量大,计算时间长,同时,DVC子体块易受成像噪声影响,会降低测量精度。针对上述问题,提出一种基于体素选择的自适应DVC方法(VSA-DVC),该方法将子体块内体素灰度梯度作为判据进行有效体素筛选,构建只包含有效体素的不定形子体块参与配准过程,可以同时提升计算效率和测量精度。Digital volume correlation(DVC)method is a non-contact and non-interference method for internal 3D deformation field measurement.Its basic principle is to measure the deformation by matching the volume images of the object before and after the deformation.It has become one of the most important internal deformation measurement methods in the field of experimental mechanics.However,DVC has a large amount of calculation and a long calculation time.At the same time,subvolumes are vulnerable to imaging noise,which will reduce the accuracy of measurement.To solve the above problems,this paper proposed a self-adaptive DVC method based on voxel selection(VSADVC).This method takes the voxel gray gradient in the sub-volume as the criterion to select the effective voxels,then forms sub-volumes composed of effective voxels with indefinite shape to participate in the registration process,which can improve the calculation efficiency and measurement accuracy at the same time.

关 键 词：自适应数字体积相关 灰度梯度 体素选择 不定形子体块 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]