Interpolation bias for the IC-GN and FA-NR algorithm
Su, Yong · Figshare · 2017
This is all original materials of my paper "Interpolation bias for the inverse compositional Gauss-Newton algorithm in digital image correlation", including C++ source code, MATLAB source code, and speckle images. These materials are divided into 2 folders: one for 1-dimensional Gaussian speckle patterns; one for 2-dimensional actual speckle images. In the 1-dimensional case, the calculation and estimation of the interpolation bias and the SSSIG are implemented using a C++ program. The C++ program will generate some "txt" data files. Then these files are read by a MATLAB program to visualize the result. Some other MATLAB programs will illustrate the shapes and power spectrums of speckle patterns, or calculate the transfer function of gradient estimation. In the 2-dimensional case, the calculation of the interpolation bias is implemented using a C++ program, and the results are written into a “txt” file; the estimation of bias is implemented using a MATLAB program. Some other MATLAB programs will calculate the power spectrum or the histograms.I use visual studio community 2013 with update 5 to compile the C++ files; I also use two open-source C++ libraries, OpenCV(2.4.9) and Armadillo(7.400.4), and I think these two libraries are great. I use MATLAB 2016a to run the MATLAB codes. I hope these codes will help readers who are interested in analyzing bias error of pattern match. If you have any questions, you can send me an email. My email address is [email protected].