SUB-PIXEL DISPLACEMENT MEASUREMENT METHOD BASED ON TIKHONOV REGULARIZATION

摘要

Disclosed is a sub-pixel displacement measurement method based on Tikhonov regularization. In a digital image correlation method, a grayscale gradient of a speckle pattern needs to be obtained when sub-pixel displacement of the speckle pattern is calculated. In a traditional calculation method, derivation is carried out for a grayscale of the speckle pattern by means of a finite difference method; however, numerical value derivation is highly unstable and is very sensitive to image noise, and a tiny measurement error can cause the grayscale gradient obtained by means of calculation to seriously deviate from a real grayscale gradient. With regard to this problem, a sub-pixel displacement measurement method based on Tikhonov regularization is provided. In the method, a grayscale of a speckle pattern is fitted by using a smooth cubic spline function, a grayscale gradient of the speckle pattern is the derivative of a cubic spline, and thus, sub-pixel displacement of a structure is obtained by using the sub-pixel displacement measurement method, thereby solving the problem of the poor noise resistance capability of traditional measurement methods, and being capable of effectively improving the measurement accuracy.