大视场双目投影条纹方法中的相机参数标定与优化研究

投影条纹法具有高精度、高分辨的特点,且实验设备简单,对实验环境要求低,适宜于不同尺度的三维形貌测量.双目投影条纹法通常采用有标准参照物的相机标定方法获取相机参数,以实现物体表面三维形貌重建.然而,在大型结构的三维形貌测量中参照物的相对尺度较小,传统的基于重投影的相机标定方法在特征检测中引入的误差会被放大,从而影响三维表...     

大型结构变形及形貌摄像测量技术研究进展

船体、机翼、工程建筑等大型结构的变形与形貌的精确测量是实验力学的基本任务之一。摄像测量具有测量精度高、范围大、非接触、可动态测量等优点,是结构形貌与运动测量的重要手段。本文介绍了作者及其所在科研团队近年来利用摄像测量技术在大型结构变形、形貌测量等方面的研究成果和典型应用范例,主要包括：提出折线光路像机链摄像测量方法与技术,用于船体等大型结构,不稳定平台的静态基准转换,以及地下工程、边坡等的多点变形长期动态监测;采用多摄像机联合组网,用于机翼、风电叶片等大型结构形貌与变形的高精度测量;在投影轮廓线辅助下进行多部像机组网,用于对大型堆场三维形貌、料堆体积和高程等的高精度测量等。     

微尺度位移、形貌测量系统及应用

研制了一套可应用于MEMS器件的微尺度测量系统,可以在受载状态下实时检测MEMS器件的面内位移、离面位移和三维形貌.该系统中,面内位移测量是一个基于白光数字散斑相关方法的显微光学测量系统,与相应的力学加载系统结合,可以得到MEMS器件在受载状态下的实时面内位移;离面位移和三维形貌测量则是一个基于相移显微投影光栅方法的光学测量系统,与相应的力学加载系统结合,可以得到MEMS器件在受载状态下的实时三维形貌和离面位移.最后给出了几个典型的MEMS器件面内位移、离面位移和三维形貌的实测结果.     

用于物体表面形貌和变形测量的三维数字图像相关方法

使用单个摄像机的二维数字图像相关方法通常仅局限于平面物体的面内变形测量,而使用两个摄像机基于双目立体视觉原理的三维数字图像相关方法克服了这一局限,可对平面和曲面物体表面的三维形貌和载荷作用下的三维变形进行测量。本文介绍了三维数字图像相关方法的基本原理及其关键技术,并用两个典型的实验验证了该方法的有效性。     

微尺度位移、形貌测量系统及应用

研制了一套可应用于MEMS器件的微尺度测量系统，可以在受载状态下实时检测MEMS器件的面内位移、离面位移和三维形貌。该系统中，面内位移测量是一个基于白光数字散斑相关方法的显微光学测量系统，与相应的力学加载系统结合，可以得到MEMS器件在受载状态下的实时面内位移；离面位移和三维形貌测量则是一个基于相移显微投影光栅方法的光学测量系统，与相应的力学加载系统结合，可以得到MEMS器件在受载状态下的实时三维形貌和离面位移。最后给出了几个典型的MEMS器件面内位移、离面位移和三维形貌的实测结果。     

一种基于投影云纹法的三维形貌测量技术及其标定方法

投影云纹法测量物体的三维形貌具有非接触式、快速、全场测量的优势,且计算结果的分辨率和精度都很高。本文提出了一种新的投影云纹系统测量三维形貌的计算和标定方法,这种方法适用于由投影仪直接投影栅线的投影云纹系统,对光路没有特殊相对位置要求,标定过程对设备的要求低,简单易行,仅需一大一小两块平整的标定板即可实现对整个系统的标定。其可行性、正确性以及高精度的特点都被真实物体测量证实。     

基于视觉测量的应变片空间定位及应用

在大型航天结构舱段试验中，通常需要粘贴大量的应变片来获取结构的应变测试数据。为了能够准确定位结构舱段上的应变片位置，本文研究了基于单目视觉测量技术的应变片空间坐标重建方法。该方法采用一台单反相机，从不同视角连续对构件进行拍摄，获得一组包含圆环编码点和应变片的图像序列；通过检测编码点图像坐标，构建相邻帧之间的匹配关系，然后基于相对定位原理估计相机姿态信息；再利用深度学习技术提取图像中的应变片坐标，结合估计的姿态信息对提取的应变片进行立体匹配，并且采用三角测量方法重构应变片空间坐标，从而实现应变片的三维定位。利用所研究的方法，分别对铝质平板和圆筒上粘贴应变片的三维坐标、碳纤维标尺的长度进行了定位。定位结果表明，该方法的重复定位精度优于0.05mm,三维点距离测量精度优于0.1mm。     

基于DIC的印制线路板三维形貌高精度测量

生产过程中对产品三维形貌的高精度实时测量是有效的质量监控手段,具有迫切的需求。本文基于三维数字图像相关(DIC-3D)方法,全场测量了印制线路板上的台阶状焊点和印制线路,得到焊点大小、高度和印制线路线宽、线高等三维形貌数字图像数据;使用台阶仪采集面内和离面形貌数据,并进行比较。实验结果表明,三维数字图像相关方法在测量印制线路板形貌时,可测量出十数微米高的印刷线路台阶,精度达3μm;对500×500pixels图像,在i7 4790k CPU硬件条件,步长3pixels、子区19×19pixels的参数下,立体图像的重建速度小于1s,可满足线路板的在线实时检测。     

一种单相机测量三维运动轨迹的方法

论文提出了一种基于交汇式双目视觉模型的单个CCD相机三维立体测量方法.在CCD相机前放置两面成一定夹角的平面镜,通过一次拍摄,得到两幅存在视差的图像,该图像相当于相机在平面镜中两个虚拟相机对目标从不同角度拍摄所得到图像,根据双目视觉原理可实现三维立体测量.详细介绍了实验系统的布置方法,建立了三维坐标计算的数学模型,对系统的内外参数进行了标定.测量了静态物体的形态和尺寸,验证了系统的可行性与可靠性.并对运动物体进行了动态测量,通过视频记录物体两个虚像的平面运动图像序列,计算还原了物体空间的三维运动轨迹,并应用误差理论对实验误差进行了分析.     

用DSPI方法和栅线投影法测量任意曲表面的三维变形

本文采用数字散斑干涉法和栅线投影法同时测量了任意曲表面的三维位移和三维形貌，并运用数字图像处理技术分析了曲表面的三维应变场，用实验方法获得了曲表面的主应变和主方向。     

Measuring the kinematics of a free-flying hawk-moth （Macroglossum stellatarum） by a comb-fringe projection method

We describe a 2D fringe projection method that involves projecting two groups of sharp comb fringes onto a free-flying hawk-moth from different directions and recording the images of distorted fringes by two high speed cameras from two orthogonal views. By calculating the 3D coordinates of the points on the hawk-moth and three-dimensional reconstruction of the wing, the flight trajectory, body attitude and wing kinematics including flapping angle, elevation angle, torsion angle, and camber deformation are obtained.     

三维射流PIV图像处理方法的研究及其精度分析

本文利用三台照相机系统记录三维射流的流动，通过照相机系统的标定、三维空间粒子的重构、空间粒子的对应及误对应向量的判断和消除等，建立了一种三个图像记录设备组成的三维PIV图像处理算法，并对其精度进行了分析，通过对三维射流图像的处理和分析表明，该方法是切实可行的。     

Visualization of three-dimensional incompressible flows by quasi-two-dimensional divergence-free projections in arbitrary flow regions

A visualization of three-dimensional incompressible flows by divergence-free quasi-two-dimensional projections of the velocity field onto three coordinate planes is revisited. An alternative and more general way to compute the projections is proposed. The approach is based on the Chorin projection combined with a SIMPLE-like iteration. Compared to the previous methodology based on divergence-free Galerkin–Chebyshev bases, this technique, formulated in general curvilinear coordinates, is applicable to any flow region and allows for faster computations. To illustrate this visualization method, examples in Cartesian and spherical coordinates, as well as post-processing of experimental 3D-PTV data, are presented.     

Accurate measurement of three-dimensional deformations in deformable and rigid bodies using computer vision

Recently, digital-image-correlation techniques have been used to accurately determine two-dimensional in-plane displacements and strains. An extension of the two-dimensional method to the acquisition of accurate, three-dimensional surfacedisplacement data from a stereo pair of CCD cameras is presented in this paper. A pin-hole camera model is used to express the transformation relating three-dimensional world coordinates to two-dimensional computer-image coordinates by the use of camera extrinsic and intrinsic parameters. Accurate camera model parameters are obtained for each camera independently by (a) using several points which have three-dimensional world coordinates that are accurate within 0.001 mm and (b) using two-dimensional image-correlation methods that are accurate to within 0.05 pixels to obtain the computer-image coordinates of various object positions. A nonlinear, least-squares method is used to select the optimal camera parameters such that the deviations between the measured and estimated image positions are minimized. Using multiple orientations of the cameras, the accuracy of the methodology is tested by performing translation tests. Using theoretical error estimates, error analyses are presented. To verify the methodology for actual tests both the displacement field for a cantilever beam and also the surface, three-dimensional displacement and strain fields for a 304L stainless-steel compact-tension specimen were experimentally obtained using stereo vision. Results indicate that the three-dimensional measurement methodology, when combined with two-dimensional digital correlation for subpixel accuracy, is a viable tool for the accurate measurement of surface displacements and strains. Paper was presented at the 1989 SEM Spring Conference on Experimental Mechanics held in Cambridge, MA on May 29–June 1.     

Shape Monitoring of a Beam Structure from Measured Strain or Curvature

This paper presents results of numerical simulations and an experimental investigation of a method to determine shape of a beam from curvature and/or strain measurements. The purpose of this work was to develop an efficient and accurate method that can be used in real time shape monitoring of beam type structures with possible extension to more complex systems. A method based upon solving a set of continuity equations is presented. Numerical simulations were implemented to minimize the number of sensors and to determine the most beneficial sensor locations and sensor/model configuration to capture the shape in a timely and effective manner. Simulations showed that dividing the beam into segments (elements) and placing sensors at the Gauss point locations of each segment gave only 0.14% systematic error while using three elements and two Gauss points per element. An experiment was designed using an aluminum beam combined with a data acquisition system and a shape reconstruction algorithm. The real-time reconstruction of shape from curvature data was accomplished using strain gages for the curvature estimates. The results were compared to a technique based on position only data and point cloud image data. Overall, consistent results were obtained. The percent difference between the experimental and photogrammetry results fluctuated from 1.4 to 3.5% when various magnitudes of concentrated loads were applied to the beam. This methodology may be useful in real-time shape control and shape modification systems with potential applications in structural health monitoring and damage detection.     

A Topology-Based Approach to Exploiting Sparsity in Multibody Dynamics: Joint Formulation*

ABSTRACT

In this paper, advantage is taken of the problem structure in multibody dynamics simulation when the mechanical system is modeled using a minimal set of generalized coordinates. It is shown that the inertia matrix associated with any open- or closed-loop mechanism is positive definite by finding a simple mathematical expression for the quadratic form expressing the kinetic energy in an associated state space. Based on this result, an algorithm that efficiently solves for second time derivatives of the generalized coordinates is presented. Significant speed-ups accrue due to both the no fill-in factorization of the composite inertia matrix technique and the degree of parallelism attainable with the new algorithm.     

Optical tomography applied to speckle photographic measurement of asymmetric flows with variable density

Optical tomography is applied to the speckle photographic measurement of an asymmetric flow field with variable fluid density. The convolution back projection algorithm is used for obtaining the 3-D density distribution. Noise in the experimental data is reduced by spline smoothing. The method is verified with a steady, laminar, axisymmetric helium jet exhausting vertically into the ambient air, and then applied to a non-axisymmetric helium jet for determining the helium concentration. It is found that speckle photographic recordings are very adequate for tomographic reconstruction, because they provide a high number of data points from each projection. The influence of the limited number of projections on the reconstruction quality is particularly investigated. Dedicated to Professor Dr.-Ing. J. Zierep on the occasion of his 60th birthday     

An experimental study of a turbulent vortex ring: a three-dimensional representation

This paper presents a reconstruction of the three-dimensional velocity field of a turbulent vortex ring by means of Taylor’s hypothesis. Stereoscopic PIV is used to acquire three velocity component information on a plane. The accuracy of the Taylor’s hypothesis for this particular flow pattern is first discussed, and the three-dimensional velocity and vorticity information are then presented. This study also introduces an azimuthally averaging method in order to give a mean structure in cylindrical coordinates from a single realization and from which turbulent stresses and production can be estimated. The azimuthally averaged quantities are then compared with the ensemble-averaged results from the previous planar (two-dimensional and stereoscopic) PIV experiments.     

An efficient simultaneous reconstruction technique for tomographic particle image velocimetry

To date, Tomo-PIV has involved the use of the multiplicative algebraic reconstruction technique (MART), where the intensity of each 3D voxel is iteratively corrected to satisfy one recorded projection, or pixel intensity, at a time. This results in reconstruction times of multiple hours for each velocity field and requires considerable computer memory in order to store the associated weighting coefficients and intensity values for each point in the volume. In this paper, a rapid and less memory intensive reconstruction algorithm is presented based on a multiplicative line-of-sight (MLOS) estimation that determines possible particle locations in the volume, followed by simultaneous iterative correction. Reconstructions of simulated images are presented for two simultaneous algorithms (SART and SMART) as well as the now standard MART algorithm, which indicate that the same accuracy as MART can be achieved 5.5 times faster or 77 times faster with 15 times less memory if the processing and storage of the weighting matrix is considered. Application of MLOS-SMART and MART to a turbulent boundary layer at Re _θ = 2200 using a 4 camera Tomo-PIV system with a volume of 1,000 × 1,000 × 160 voxels is discussed. Results indicate improvements in reconstruction speed of 15 times that of MART with precalculated weighting matrix, or 65 times if calculation of the weighting matrix is considered. Furthermore the memory needed to store a large weighting matrix and volume intensity is reduced by almost 40 times in this case.     

具有非线性homologous变形约束的桁架结构形态分析

研究了具有非线性homologous变形约束条件的桁架结构形态分析问题。在已有的线性homologous变形约束桁架形态分析的基础上,将结构的节点分成三类:homologous变形约束节点,形状可变节点和边界点。运用Moore-Penrose广义逆矩阵性质,将基础方程组解的存在条件表示为包含形状可变节点未知坐标的非线性方程组,为采用Newton-Raphson方法求解非线性方程组,对AA (A为任意矩阵,A 为A的Moore-Penrose广义逆矩阵)求偏导数,找到了满足保型要求的形态,给出的桁架算例说明了本文方法的有效性。