高斯滤波对规则斑点图像的亚像素匹配精度影响研究

Research on the influence of Gaussian filtering on the sub-pixel matching accuracy of regular spot image

针对数字图像相关法中亚像素匹配误差,提出一种适用于规则斑点图像的高斯滤波预处理方法来降低亚像素匹配误差,模拟了恒定、线性、二次和三次位移场,并采用不同标准差的高斯滤波器分别对图像进行处理。试验结果表明:在不同的位移场下,当采用标准差σ=0.75 px的高斯滤波器处理后,图像的亚像素匹配误差降低到最低,并且通过与随机散斑图像对比分析,证明该方法只适用于规则斑点图像,最后对焊接图像进行分析,验证了该方法的稳定性。

Aiming at the sub-pixel matching error in the digital image correlation method,a Gaussian filter preprocessing method suitable for regular blob images is proposed to reduce the sub-pixel matching error.The constant displacement field,linear displacement field,secondary displacement field and tertiary displacement field are simulated,and Gaussian filters with different standard deviations are used to process the images respectively.The test results show that under different displacement fields,after the image is processed with a Gaussian filter with standard deviation σ=0.75 px,the sub-pixel matching error of the image is reduced to the minimum.And through comparative analysis with random speckle images,it is proved that this method is only suitable for regular speckle images.Finally,the welding image is analyzed to verify the stability of the method.