Automated analysis of 3-D shape and surface strain distributions of a moving object using stereo vision
Automated analysis of 3-D shape and surface strain distributions of a moving object using stereo vision
复制标题
使用立体视觉自动分析移动物体的 3D 形状和表面应变分布
DOI:
10.1016/0143-8166(93)90027-i
复制
发表时间:
1993
影响因子:
4.6
通讯作者:
M. Fujigaki
中科院分区:
文献类型:
--
作者:
Y. Morimoto;M. Fujigaki
An automated method for analyzing the three-dimensional (3-D) shape and the surface strain distributions of a moving object, using stereo vision, is proposed. A two-dimensional (2-D) cross grating is drawn on the surface of the object and it is recorded as images with two charge coupled device cameras. For correspondence between the surface of the object and the images, the Fourier transform grid method (FTGM) is used. By separating each directional first harmonic of the Fourier spectrum of the images using the FTGM, the 2-D cross grating can be easily separated into each directional grating, and each directional phase information of the grating brightness distribution is obtained. The correspondence between arbitrary points on the surface of the object and the images is determined, using the phase values instead of the grating line numbers. The correspondence algorithm is much more easily applied than the conventional image processing method, using the pixel positions of the center lines of the grating lines. The analysis of the 3-D shape is faster. Accordingly, analysis of large quantity images becomes possible and consequently the effective measurement of a dynamically deforming object can be performed. As an application, in the case of elastic-plastic buckling of a curved steel plate, the 3-D shape and the surface strain distributions changing with time are measured.