n-dim difference of offset Gaussian DooG filter (Gaussian derivative). Creates a nd derivative of Gaussian kernel. For all but d==1 use primarily for visualization purposes -- for filtering better to use the indvidiual seperable kernels for efficiency purposes. USAGE dG = filterDoog( dims, sigmas, nderivs, [show] ) INPUTS dims - nd element vector of dimensions of final Gaussian sigmas - sigmas for nd Gaussian nderivs - order of derivative along each of the nd dimensions show - [0] figure to use for optional display OUTPUTS dG - The derivative of Gaussian filter EXAMPLE dG1 = filterDoog( 43, 2, 3, 1 ); %1D dG2 = filterDoog( [41 41], [3 3], [1,1], 2 ); %2D dG3 = filterDoog( [101 101 101], [4,4,10], [1,1,0], 3 ); %3D See also FILTERGAUSS, NORMPDF2, FILTERDOG2D, FILTERGABOR2D Piotr's Computer Vision Matlab Toolbox Version 2.0 Copyright 2014 Piotr Dollar. [pdollar-at-gmail.com] Licensed under the Simplified BSD License [see external/bsd.txt]