computederivatives2.m

来自「图像配准的源程序」· M 代码 · 共 23 行

function [fx,fy,ft] = computeDerivatives2(im1,im2)
% 
% function [fx,fy,fz] = computeDerivatives2(im1,im2)
%
% im1 and im2 are images
%
% [fx,fy,ft] are images, derivatives of the input images

filter = [0.03504 0.24878 0.43234 0.24878 0.03504];
dfilter = [0.10689 0.28461 0.0  -0.28461  -0.10689];

dx1 = conv2sep(im1,dfilter,filter,'valid');
dy1 = conv2sep(im1,filter,dfilter,'valid');
blur1 = conv2sep(im1,filter,filter,'valid');
dx2 = conv2sep(im2,dfilter,filter,'valid');
dy2 = conv2sep(im2,filter,dfilter,'valid');
blur2 = conv2sep(im2,filter,filter,'valid');

fx=(dx1+dx2)/2;
fy=(dy1+dy2)/2;
ft=(blur2-blur1);