testgaussgradient.m

本算法是基于高斯分布的梯度边缘检测法

clear;close all;
im=imread('rice.png');
fim=mat2gray(im);
%draw magnitude of gradient
figure('name','Magnitude of Gradient');
[imx,imy]=gaussgradient(fim,0.5);
subplot(2,2,1); imshow(abs(imx)+abs(imy)); title('sigma=0.5');
[imx,imy]=gaussgradient(fim,1.0);
subplot(2,2,2); imshow(abs(imx)+abs(imy)); title('sigma=1.0');
[imx,imy]=gaussgradient(fim,1.5);
subplot(2,2,3); imshow(abs(imx)+abs(imy)); title('sigma=1.5');
[imx,imy]=gaussgradient(fim,2.0);
subplot(2,2,4); imshow(abs(imx)+abs(imy)); title('sigma=2.0');

%draw gradeint in a small region with sigma=1.0
[imx,imy]=gaussgradient(fim,1.0);
figure('name','Gradient');
imshow(fim(1:50,1:50),'InitialMagnification','fit');
hold on;
quiver(imx(1:50,1:50),imy(1:50,1:50));
title('sigma=1.0');