数字图像处理实验（17）：PROJECT 06-04，Color Image Segmentation 标签： 图像处理MATLAB 2017-05-27 21:13

实验报告：

Objective:

Color image segmentation is a big issue in image processing. This students need to know the basics of this topic.

Main requirements:

Ability of programming with C, C++, or Matlab.

Instruction manual:

Download Fig. 6.28(b) and duplicate Example 6.15, but segment instead the darkest regions in the image.

本实验是彩色图像分割，从彩色图像中分割出特定颜色。这里我选了红色，从下面图片分割红色出来，主要是草莓的部分有较多的红色分量。

原图像：

实验代码：

%
close all;
clc;
clear all;

%
img = imread('Fig6.30(01).jpg');
figure;
subplot(2, 2, 1);
imshow(img);
title('original image');

%
img1 = im2double(img);
R = img1(:, :, 1);
G = img1(:, :, 2);
B = img1(:, :, 3);

subplot(2, 3, 4);
imshow(R);
title('Red');
subplot(2, 3, 5);
imshow(G);
title('Green');
subplot(2, 3, 6);
imshow(B);
title('Blue');

%
R1 = R(129:256, 86:170);
R1_ave = mean(mean(R1(:)));
[M, N] = size(R1);
sd = 0.0;
for i = 1:M
    for j = 1:N
        sd = sd + (R1(i, j) - R1_ave) * (R1(i, j) - R1_ave);
    end
end
R1_d = sqrt(sd/(M*N));

R2 = zeros(size(img, 1), size(img, 2));
index = find((R > R1_ave - 1.25*R1_d) & (R < R1_ave + 1.25*R1_d));
R2(index) = 1;

subplot(2, 2, 2);
imshow(R2);
title('segment red');

实验结果：

从颜色分割的结果来看，白色部分主要为草莓的那部分区域，也正好是我们所要区分的红色的区域。