基于libsvm的图像分割.doc

测试数据来源：《MATLAB高效编程技巧与应用：25个案例分析》吴鹏(著)|北京航空航天大学出版社《MATLAB统计分析与应用：40个案例分析》谢中华(著)|北京航空航天大学出版社该帖子中测试程序需要使用如下工具箱libsvm-mat-2.89-3[FarutoUltimate3.0]HYPERLINK"http://www.matlabsky.com/thread-9327-1-1.html"\t"_blank"http://www.matlabsky.com/thread-9327-1-1.html更多关于SVM的事情请看关于SVM的那点破事[长期更新整理byfaruto]HYPERLINK"http://www.matlabsky.com/thread-10966-1-1.html"\t"_blank"http://www.matlabsky.com/thread-10966-1-1.html
测试图片：现用libsvm来实现图像分割，测试图片用的亦是25cases和40cases中的那个littleduck测试图片。主体程序思想为25cases中的代码过程，改进之处为可以让用户利用ginput来提取背景的样本点和前景（待分割出来的目标）的样本点作为训练样本，而不需实现指定背景和前景的样本点，也不用额外的小软件来查看某点的RGB值，ginput即可。O(∩_∩)O~测试代码：
本帖隐藏的内容
%%ImSegmentLibsvm
%alittertestofimagesegmentbasedonlibsvm
%byfaruto
%Email:[e-mail]
%QQ:516667408
%http://blog.sina.com.cn/faruto
%lastmodified2010.11.05
%www.matlabsky.com
%www.mfun.la
%%
tic;
closeall;
clear;
clc;
formatcompact;
%%
pic=imread('littleduck.jpg');
figure;
imshow(pic);
%%确定训练集

TrainData_background=zeros(20,3,'double');
TrainData_foreground=zeros(20,3,'double');

%背景采样
msgbox('Pleaseget20backgroundsamples','BackgroundSamples','help');
pause;
forrun=1:20
[x,y]=ginput(1);
holdon;
plot(x,y,'r*');
x=uint8(x);
y=uint8(y);
TrainData_background(run,1)=pic(x,y,1);
TrainData_background(run,2)=pic(x,y,2);
TrainData_background(run,3)=pic(x,y,3);
end
%待分割出来的前景采样
msgbox('Pleaseget20foregroundsampleswhichistheparttobesegmented','ForegroundSamples','help');
pause;
forrun=1:20
[x,y]=ginput(1);
holdon;
plot(x,y,'ro');
x=uint8(x);
y=uint8(y);
TrainData_foreground(run,1)=pic(x,y,1);
TrainData_foreground(run,2)=pic(x,y,2);
TrainData_foreground(run,3)=pic(x,y,3);
end
%%背景训练样本10*3
%TrainData_background=...
%[527487;
%76117150;
%194862;
%356482;
%465836;
%505723;
%110127135;
%156173189;
%246242232;
%166174151];
%%前景训练样本8*3
%TrainData_foreground=...
%[211192107;
%202193164;
%32250;
%213201151;
%1157516;
%101700;
%16913122;
%15013387];

%letbackgroundbe0&foreground1
TrainLabel=[zeros(length(TrainData_background),1);...
ones(length(TrainData_foreground),1)];