模拟散斑尺寸对数字图像相关算法影响分析

Influence of stimulated speckle size on digital image correlation algorithm

数字散斑测试精度的影响因素相对较多、精准控制难度较大,定量分析程度不足。为了提高数字散斑测试精度,文章通过模拟散斑的方法分析观测散斑尺寸大小对计算结果的影响。结果表明:随着散斑尺寸的增加,位移计算误差呈现先降后升的趋势,即散斑尺寸过大或者过小都不利于数字图像相关的位移计算,最优散斑尺寸为5~7像素。研究结果为数字图像相关计算的观测对象设计提供了数据基准,为实际现场监测的观测目标变形信息的精度提高提供了数据基础。

There are many factors affecting the accuracy of digital speckle measurement. It is difficult to accurately control these factors, and the degree of quantitative analysis is insufficient. In order to improve the accuracy of digital speckle measurement, the speckle simulation focusing on the influence of the size of observed speckle on the calculation results was conducted. The results show that the displacement calculation error decreases first and then increases with the increase of speckle size, suggesting that too large or too small speckle size is not conducive to the displacement calculation of digital image correlation, and the optimal speckle size is 5-7 pixel. The research results provide a data benchmark for the observation object design of digital image correlation calculation, and provide a data basis for improving the accuracy of observation target deformation information monitored in the actual field.