用 Matlab GUI搭建一个简单的CBIR图像检索系统（二）

上一篇文章介绍了CBIR的流程和图像特征的提取以及图像特征库的建立。在这一篇文章中，将创建一个基于直方图的图像检索GUI。

如上图所示，就是利用matlab gui建立的一个图像检索的框架，其中用到的控件有按钮、单选按钮、静态文本框、坐标轴框以及面板。

ImageRead按钮的Callbacall函数为

% --- Executes on button press in ChooseImage.function ChooseImage_Callback(hObject, eventdata, handles)% hObject    handle to ChooseImage (see GCBO)% eventdata  reserved - to be defined in a future version of MATLAB% handles    structure with handles and user data (see GUIDATA)[filename,PathName] = uigetfile({'*.BMP';'*.bmp';'*.tif';'*.jpg';'*.png'},'选择图像');str = [PathName filename];if PathName ~=0    OrginImage_O = imread(str);    OrginImage_O = OrginImage_O(:,:,1:3);    if get(handles.btn_total,'value')        OrginImage=OrginImage_O;        OrginImage = imresize(OrginImage,[256,256]);    elseif get(handles.btn_part,'value')        figure(2),imshow(OrginImage_O);        OrginImage=imcrop();      %这里把要裁剪的图像框出来        OrginImage = imresize(OrginImage,[256,256]);        close(2);    end    axes(handles.axes1);    imshow(OrginImage);    setappdata(handles.ChooseImage,'OrginImage',OrginImage);end

Image Capture为调用相机程序获取图片的方法，

function pushbutton4_Callback(hObject, eventdata, handles)% hObject    handle to pushbutton4 (see GCBO)% eventdata  reserved - to be defined in a future version of MATLAB% handles    structure with handles and user data (see GUIDATA)open ('C:\Program Files (x86)\Canon\EOS Utility\EOS Utility.exe');

Retrieval按钮对应的事件：

（１）提取读取图像的对应特征

（２）计算检索图像的特征和特征库中的特征的相似度

（３）按照相似度的大小将检索结果显示在右边的坐标轴框中

% --- Executes on button press in pushbutton3.function pushbutton3_Callback(hObject, eventdata, handles)% hObject    handle to pushbutton3 (see GCBO)% eventdata  reserved - to be defined in a future version of MATLAB% handles    structure with handles and user data (see GUIDATA)IndexOfPoints=[];OrginImage=getappdata(handles.ChooseImage,'OrginImage');if get(handles.btn_ch,'value')    load('HISTFeatures.mat');    [m,n]=size(HISTFeatures);    OrginImage_Gray = rgb2gray(OrginImage);    [Count1,x] = imhist(OrginImage_Gray) ;        for index = 1:m        Count2 = HISTFeatures{index,2};        PathData = HISTFeatures{index,1};        Sum1=sum(Count1);Sum2=sum(Count2);        Sumup = sqrt(Count1.*Count2);        SumDown = sqrt(Sum1*Sum2);        Sumup = sum(Sumup);        IndexOfBashi=1-sqrt(1-Sumup/SumDown);        IndexOfPoint{index,1}=IndexOfBashi;        IndexOfPoint{index,2}=PathData;    end endIndexOfPoint=sortrows(IndexOfPoint);axes(handles.axes2); Serch = imread(IndexOfPoint{m,2});Serch = Serch(:,:,1:3);imshow(Serch);axes(handles.axes3); Serch = imread(IndexOfPoint{(m-1),2});Serch = Serch(:,:,1:3);imshow(Serch);axes(handles.axes4); Serch = imread(IndexOfPoint{(m-2),2});Serch = Serch(:,:,1:3);imshow(Serch);axes(handles.axes5); Serch = imread(IndexOfPoint{(m-3),2});Serch = Serch(:,:,1:3);imshow(Serch);axes(handles.axes6); Serch = imread(IndexOfPoint{(m-4),2});Serch = Serch(:,:,1:3);imshow(Serch);axes(handles.axes7); Serch = imread(IndexOfPoint{(m-5),2});Serch = Serch(:,:,1:3);imshow(Serch);axes(handles.axes8); Serch = imread(IndexOfPoint{(m-6),2});Serch = Serch(:,:,1:3);imshow(Serch);axes(handles.axes9); Serch = imread(IndexOfPoint{(m-7),2});Serch = Serch(:,:,1:3);imshow(Serch);axes(handles.axes10); Serch = imread(IndexOfPoint{(m-8),2});Serch = Serch(:,:,1:3);imshow(Serch);axes(handles.axes11); Serch = imread(IndexOfPoint{(m-9),2});Serch = Serch(:,:,1:3);imshow(Serch);axes(handles.axes12); Serch = imread(IndexOfPoint{(m-10),2});Serch = Serch(:,:,1:3);imshow(Serch);axes(handles.axes13); Serch = imread(IndexOfPoint{(m-11),2});Serch = Serch(:,:,1:3);imshow(Serch);

最终效果图