LAMPIRAN SOURCE CODE MATLAB 7.1

Download PDF
12 downloads 313 Views 481KB Size Report
Comment
Jun 7, 2007 ... LAMPIRAN SOURCE CODE MATLAB 7.1. APLIKASI. SEGMENTASI CITRA SPOT DENGAN MENGGUNAKAN. PENDEKATAN FUZZY C- ...
L1

LAMPIRAN SOURCE CODE MATLAB 7.1 APLIKASI SEGMENTASI CITRA SPOT DENGAN MENGGUNAKAN PENDEKATAN FUZZY C-MEANS •

Main Program Source Code

function varargout = Image_Segmentation_Using_Fuzzy_C_Means(varargin) % IMAGE_SEGMENTATION_USING_FUZZY_C_MEANS M-file for Image_Segmentation_Using_Fuzzy_C_Means.fig % IMAGE_SEGMENTATION_USING_FUZZY_C_MEANS, by itself, creates a new IMAGE_SEGMENTATION_USING_FUZZY_C_MEANS or raises the existing % singleton*. % % H = IMAGE_SEGMENTATION_USING_FUZZY_C_MEANS returns the handle to a new IMAGE_SEGMENTATION_USING_FUZZY_C_MEANS or the handle to % the existing singleton*. % % IMAGE_SEGMENTATION_USING_FUZZY_C_MEANS('CALLBACK',hObject,eventData,han dles,...) calls the local % function named CALLBACK in IMAGE_SEGMENTATION_USING_FUZZY_C_MEANS.M with the given input arguments. % % IMAGE_SEGMENTATION_USING_FUZZY_C_MEANS('Property','Value',...) creates a new IMAGE_SEGMENTATION_USING_FUZZY_C_MEANS or raises the % existing singleton*. Starting from the left, property value pairs are % applied to the GUI before Image_Segmentation_Using_Fuzzy_C_Means_OpeningFunction gets called. An % unrecognized property name or invalid value makes property application % stop. All inputs are passed to Image_Segmentation_Using_Fuzzy_C_Means_OpeningFcn via varargin. % % *See GUI Options on GUIDE's Tools menu. Choose "GUI allows only one % instance to run (singleton)". % % See also: GUIDE, GUIDATA, GUIHANDLES % Edit the above text to modify the response to help Image_Segmentation_Using_Fuzzy_C_Means % Last Modified by GUIDE v2.5 07-Jun-2007 03:37:13 % Begin initialization code - DO NOT EDIT gui_Singleton = 1; gui_State = struct('gui_Name', mfilename, ... 'gui_Singleton', gui_Singleton, ... 'gui_OpeningFcn', @Image_Segmentation_Using_Fuzzy_C_Means_OpeningFcn, ...

L2 'gui_OutputFcn', @Image_Segmentation_Using_Fuzzy_C_Means_OutputFcn, ... 'gui_LayoutFcn', [] , ... 'gui_Callback', []); if nargin && ischar(varargin{1}) gui_State.gui_Callback = str2func(varargin{1}); end if nargout [varargout{1:nargout}] = gui_mainfcn(gui_State, varargin{:}); else gui_mainfcn(gui_State, varargin{:}); end % End initialization code - DO NOT EDIT

% --- Executes just before Image_Segmentation_Using_Fuzzy_C_Means is made visible. function Image_Segmentation_Using_Fuzzy_C_Means_OpeningFcn(hObject, eventdata, handles, varargin) global img; set(handles.Image3,'visible','off'); set(handles.cluster1,'visible','off'); set(handles.cluster2,'visible','off'); set(handles.cluster3,'visible','off'); set(handles.cluster4,'visible','off'); set(handles.cluster5,'visible','off'); set(handles.cluster6,'visible','off'); set(handles.btncluster1,'visible','off'); set(handles.btncluster2,'visible','off'); set(handles.btncluster3,'visible','off'); set(handles.btncluster4,'visible','off'); set(handles.btncluster5,'visible','off'); set(handles.btncluster6,'visible','off'); set(handles.valuecluster1,'visible','off'); set(handles.valuecluster2,'visible','off'); set(handles.valuecluster3,'visible','off'); set(handles.valuecluster4,'visible','off'); set(handles.valuecluster5,'visible','off'); set(handles.valuecluster6,'visible','off'); set(handles.clusterbox1,'visible','off'); set(handles.clusterbox2,'visible','off'); set(handles.clusterbox3,'visible','off'); set(handles.clusterbox4,'visible','off'); set(handles.clusterbox5,'visible','off'); set(handles.clusterbox6,'visible','off'); handles.output = hObject; img.status = 0; % Update handles structure guidata(hObject, handles); % UIWAIT makes Image_Segmentation_Using_Fuzzy_C_Means wait for user response (see UIRESUME) % uiwait(handles.info);

L3

% --- Outputs from this function are returned to the command line. function varargout = Image_Segmentation_Using_Fuzzy_C_Means_OutputFcn(hObject, eventdata, handles) % varargout cell array for returning output args (see VARARGOUT); % hObject handle to figure % eventdata reserved - to be defined in a future version of MATLAB % handles structure with handles and user data (see GUIDATA) % Get default command line output from handles structure varargout{1} = handles.output;

% -------------------------------------------------------------------function open_Callback(hObject, eventdata, handles) global img;

img.open = uigetfile({'*.jpg';'*.gif'},'Open Image'); img.read = imread (img.open); axes(handles.Image3); img.load = imshow (img.read); img.status = 1; % -------------------------------------------------------------------function save_Callback(hObject, eventdata, handles) global img; img.save = uiputfile ({'*.jpg';'*.gif';'*.tif';'*.raw';'*.png'},'Save Image'); img.get = getimage(handles.Image3); imwrite(img.get,img.save); % -------------------------------------------------------------------function file_Callback(hObject, eventdata, handles) % hObject handle to file (see GCBO) % eventdata reserved - to be defined in a future version of MATLAB % handles structure with handles and user data (see GUIDATA)

% -------------------------------------------------------------------function about_Callback(hObject, eventdata, handles) % hObject handle to about (see GCBO) % eventdata reserved - to be defined in a future version of MATLAB % handles structure with handles and user data (see GUIDATA) helpdlg(sprintf(' Jimmy Johan / 0700708953 \n Sherwin Sasmita Rachmat / 0700710314'),'About Us'); % -------------------------------------------------------------------function exit_Callback(hObject, eventdata, handles) button = questdlg('Do you want to close application?','Exit','Yes','No','Yes'); if button=='Yes' close(figure(Image_Segmentation_Using_Fuzzy_C_Means));

L4 elseif button=='No' clc; Image_Segmentation_Using_Fuzzy_C_Means_OpeningFcn; end; % --- Executes on button press in start. function start_Callback(hObject, eventdata, handles) global img; if(img.status == 0) msgbox(' Please Load Image First','Image Load'); elseif (img.status == 2) msgbox(' The Image Has been Classified','Image Classification'); else

%Parse and initialize input and output arguments if ndims(img.read)~=3 msgbox('Input image must be m x n x 3 (RGB)'); end %Operate on image as double, regardless of input type if ~isa(img.read,'double') img.read = im2double(img.read); end %Get individual RGB colors included in segmentation [a,b,c]=size(img.read); rc = img.read(:,:,1); gc = img.read(:,:,2); bc = img.read(:,:,3); img.temp = img.read; segcolors = double([rc(:) gc(:) bc(:)]); axes(handles.Image3); img.load = imshow (img.temp); nClusters = 6; title('Segmenting. Please wait.......'); drawnow; [Idx,U,fcnobj] = fcm(segcolors,nClusters); [g,j]=size(U); img.status = zeros(a,b); img.clus1 = 0; img.clus2 = 0; img.clus3 = 0; img.clus4 = 0; img.clus5 = 0; img.clus6 = 0;

degree = max(U); for i=1 : j if ( U(1,i)== degree(1,i) )

L5 Id(i,:) = Idx(1,:); img.clus1 = img.clus1+1; elseif ( U(2,i)== degree(1,i) Id(i,:) = Idx(2,:); img.clus2 = img.clus2+1; elseif ( U(3,i)== degree(1,i) Id(i,:) = Idx(3,:); img.clus3 = img.clus3+1; elseif ( U(4,i)== degree(1,i) Id(i,:) = Idx(4,:); img.clus4 = img.clus4+1; elseif ( U(5,i)== degree(1,i) Id(i,:) = Idx(5,:); img.clus5 = img.clus5+1; elseif ( U(6,i)== degree(1,i) Id(i,:) = Idx(6,:); img.clus6 = img.clus6+1; end; end;

)

)

)

)

)

img.size = j; %------------------------------------------------------------------------% Cluster #1 clus1(1,:) = U(1,:); cluster1 = reshape(clus1,size(img.read(:,:,1))); markcluster1 = zeros(a,b); img.imagecluster1 = zeros(a,b,c); for x=1 : a for y=1 :b if cluster1(x,y)> 0.5 markcluster1(x,y) = 1; end; end; end; [pixel,I] = contour_trace(markcluster1,1,1); [batasx,batasy]=size(I); I(batasx,:) = 0; I(batasx-1,:) = 0; I(:,batasy) = 0; I(:,batasy-1) = 0; for x=1 : a for y=1 :b if I(x,y) == 255 img.imagecluster1(x,y,1)=Idx(1,1); img.imagecluster1(x,y,2)=Idx(1,2); img.imagecluster1(x,y,3)=Idx(1,3); end; end; end; %------------------------------------------------------------------------% Cluster #2 clus2(1,:) = U(2,:); cluster2 = reshape(clus2,size(img.read(:,:,1))); markcluster2 = zeros(a,b);

L6 img.imagecluster2 = zeros(a,b,c); for x=1 : a for y=1 :b if cluster2(x,y)> 0.5 markcluster2(x,y) = 1; end; end; end; [pixel,I] = contour_trace(markcluster2,1,1); [batasx,batasy]=size(I); I(batasx,:) = 0; I(batasx-1,:) = 0; I(:,batasy) = 0; I(:,batasy-1) = 0; for x=1 : a for y=1 :b if I(x,y) == 255 img.imagecluster2(x,y,1)=Idx(2,1); img.imagecluster2(x,y,2)=Idx(2,2); img.imagecluster2(x,y,3)=Idx(2,3); end; end; end; %------------------------------------------------------------------------% Cluster #3 clus3(1,:) = U(3,:); cluster3 = reshape(clus3,size(img.read(:,:,1))); markcluster3 = zeros(a,b); img.imagecluster3 = zeros(a,b,c); for x=1 : a for y=1 :b if cluster3(x,y)> 0.5 markcluster3(x,y) = 1; end; end; end; [pixel,I] = contour_trace(markcluster3,1,1); [batasx,batasy]=size(I); I(batasx,:) = 0; I(batasx-1,:) = 0; I(:,batasy) = 0; I(:,batasy-1) = 0; for x=1 : a for y=1 :b if I(x,y) == 255 img.imagecluster3(x,y,1)=Idx(3,1); img.imagecluster3(x,y,2)=Idx(3,2); img.imagecluster3(x,y,3)=Idx(3,3); end; end; end; %------------------------------------------------------------------------% Cluster #4 clus4(1,:) = U(4,:); cluster4 = reshape(clus4,size(img.read(:,:,1)));

L7 markcluster4 = zeros(a,b); img.imagecluster4 = zeros(a,b,c); for x=1 : a for y=1 :b if cluster4(x,y)> 0.5 markcluster4(x,y) = 1; end; end; end; [pixel,I] = contour_trace(markcluster4,1,1); [batasx,batasy]=size(I); I(batasx,:) = 0; I(batasx-1,:) = 0; I(:,batasy) = 0; I(:,batasy-1) = 0; for x=1 : a for y=1 :b if I(x,y) == 255 img.imagecluster4(x,y,1)=Idx(4,1); img.imagecluster4(x,y,2)=Idx(4,2); img.imagecluster4(x,y,3)=Idx(4,3); end; end; end; %------------------------------------------------------------------------% Cluster #5 clus5(1,:) = U(5,:); cluster5 = reshape(clus5,size(img.read(:,:,1))); markcluster5 = zeros(a,b); img.imagecluster5 = zeros(a,b,c); for x=1 : a for y=1 :b if cluster5(x,y)> 0.5 markcluster5(x,y) = 1; end; end; end; [pixel,I] = contour_trace(markcluster5,1,1); [batasx,batasy]=size(I); I(batasx,:) = 0; I(batasx-1,:) = 0; I(:,batasy) = 0; I(:,batasy-1) = 0; for x=1 : a for y=1 :b if I(x,y) == 255 img.imagecluster5(x,y,1)=Idx(5,1); img.imagecluster5(x,y,2)=Idx(5,2); img.imagecluster5(x,y,3)=Idx(5,3); img.status(x,y) =5; end; end; end; %-------------------------------------------------------------------------

L8 % Cluster #6 clus6(1,:) = U(6,:); cluster6 = reshape(clus6,size(img.read(:,:,1))); markcluster6 = zeros(a,b); img.imagecluster6 = zeros(a,b,c); for x=1 : a for y=1 :b if cluster6(x,y)> 0.5 markcluster6(x,y) = 1; end; end; end; [pixel,I] = contour_trace(markcluster6,1,1); [batasx,batasy]=size(I); I(batasx,:) = 0; I(batasx-1,:) = 0; I(:,batasy) = 0; I(:,batasy-1) = 0; for x=1 : a for y=1 :b if I(x,y) == 255 img.imagecluster6(x,y,1)=Idx(6,1); img.imagecluster6(x,y,2)=Idx(6,2); img.imagecluster6(x,y,3)=Idx(6,3); end; end; end; %-------------------------------------------------------------------------

pixel_labels = reshape(Id,size(img.read)); axes(handles.Image3); img.load = imshow(pixel_labels,[]); title('Image labeled by (FCM) Fuzzy C-Means Clustering'); img.Idx = Idx; colormap(Idx); tmp = colorbar; set(tmp,'ytick',1:nClusters); img.status = 2; img.classimage = pixel_labels; end % --- Executes on button press in information. function information_Callback(hObject, eventdata, handles) % hObject handle to information (see GCBO) % eventdata reserved - to be defined in a future version of MATLAB % handles structure with handles and user data (see GUIDATA) global img; if(img.status == 0) msgbox(' Please Load Image First','Image Load'); elseif (img.status == 1) msgbox(' Please Classification The Image First','Image Classification'); else

L9

set(handles.btncluster1,'visible','on'); set(handles.btncluster2,'visible','on'); set(handles.btncluster3,'visible','on'); set(handles.btncluster4,'visible','on'); set(handles.btncluster5,'visible','on'); set(handles.btncluster6,'visible','on'); set(handles.valuecluster1,'visible','on'); set(handles.valuecluster2,'visible','on'); set(handles.valuecluster3,'visible','on'); set(handles.valuecluster4,'visible','on'); set(handles.valuecluster5,'visible','on'); set(handles.valuecluster6,'visible','on'); set(handles.clusterbox1,'visible','on'); set(handles.clusterbox2,'visible','on'); set(handles.clusterbox3,'visible','on'); set(handles.clusterbox4,'visible','on'); set(handles.clusterbox5,'visible','on'); set(handles.clusterbox6,'visible','on'); %Pembuatan Image Cluster cluster1 = zeros(10,10,3); cluster1(:,:,1)=img.Idx(1,1); cluster1(:,:,2)=img.Idx(1,2); cluster1(:,:,3)=img.Idx(1,3); cluster2 = zeros(10,10,3); cluster2(:,:,1)=img.Idx(2,1); cluster2(:,:,2)=img.Idx(2,2); cluster2(:,:,3)=img.Idx(2,3); cluster3 = zeros(10,10,3); cluster3(:,:,1)=img.Idx(3,1); cluster3(:,:,2)=img.Idx(3,2); cluster3(:,:,3)=img.Idx(3,3); cluster4 = zeros(10,10,3); cluster4(:,:,1)=img.Idx(4,1); cluster4(:,:,2)=img.Idx(4,2); cluster4(:,:,3)=img.Idx(4,3); cluster5 = zeros(10,10,3); cluster5(:,:,1)=img.Idx(5,1); cluster5(:,:,2)=img.Idx(5,2); cluster5(:,:,3)=img.Idx(5,3); cluster6 = zeros(10,10,3); cluster6(:,:,1)=img.Idx(6,1); cluster6(:,:,2)=img.Idx(6,2); cluster6(:,:,3)=img.Idx(6,3); %Menampilkan Image Cluster axes(handles.cluster1) imshow(cluster1); axes(handles.cluster2) imshow(cluster2); axes(handles.cluster3)

L10 imshow(cluster3); axes(handles.cluster4) imshow(cluster4); axes(handles.cluster5) imshow(cluster5); axes(handles.cluster6) imshow(cluster6); nilaicluster1 nilaicluster2 nilaicluster3 nilaicluster4 nilaicluster5 nilaicluster6

= = = = = =

0; 0; 0; 0; 0; 0;

nilaicluster1 nilaicluster2 nilaicluster3 nilaicluster4 nilaicluster5 nilaicluster6

= = = = = =

(img.clus1/img.size)*100; (img.clus2/img.size)*100; (img.clus3/img.size)*100; (img.clus4/img.size)*100; (img.clus5/img.size)*100; (img.clus6/img.size)*100;

set(handles.valuecluster1,'String',[nilaicluster1]); set(handles.valuecluster2,'String',[nilaicluster2]); set(handles.valuecluster3,'String',[nilaicluster3]); set(handles.valuecluster4,'String',[nilaicluster4]); set(handles.valuecluster5,'String',[nilaicluster5]); set(handles.valuecluster6,'String',[nilaicluster6]);

for i =1 :6 if ((img.Idx(i,1)>=110/255) && (img.Idx(i,1)=120/255) && (img.Idx(i,2)=100/255) && (img.Idx(i,3)=130/255)&& (img.Idx(i,1)=140/255) && (img.Idx(i,2)=130/255) && (img.Idx(i,3)=45/255) && (img.Idx(i,1)=55/255) && (img.Idx(i,2)=50/255) && (img.Idx(i,3)=85/255)&& (img.Idx(i,1)=100/255) && (img.Idx(i,2)=85/255) && (img.Idx(i,3)=70/255)&& (img.Idx(i,1)=80/255) && (img.Idx(i,2)=70/255) && (img.Idx(i,3)=170/255) && (img.Idx(i,1)=180/255) && (img.Idx(i,2)=170/255) && (img.Idx(i,3)=1 && (sp_x(loop,2)-1)>=1) if (sp_x(loop,1)+1 >= batasx) sp_x(loop,1)= sp_x(loop,1)-1; end; if (sp_x(loop,2)+1 >= batasy) sp_x(loop,2)= sp_x(loop,2)-1; end; if (BW((sp_x(loop,1)-1),(sp_x(loop,2)-1))==0) k=k+1; sp_x(k,2)=sp_x(loop,2)-1; sp_x(k,1)=sp_x(loop,1)-1; BW((sp_x(loop,1)-1),(sp_x(loop,2)-1))=1; end end if ((sp_x(loop,1))>=1 && (sp_x(loop,2)-1)>=1) if (sp_x(loop,1)+1 >= batasx) sp_x(loop,1)= sp_x(loop,1)-1; end; if (sp_x(loop,2)+1 >= batasy) sp_x(loop,2)= sp_x(loop,2)-1; end; if (BW((sp_x(loop,1)),(sp_x(loop,2)-1))==0) k=k+1; sp_x(k,2)=sp_x(loop,2)-1; sp_x(k,1)=sp_x(loop,1); BW((sp_x(loop,1)),(sp_x(loop,2)-1))=1;

L15 end end

if ((sp_x(loop,1)+1)>=1 && ((sp_x(loop,2)-1)>=1)) if (sp_x(loop,1)+1 >= batasx) sp_x(loop,1)= sp_x(loop,1)-1; end; if (sp_x(loop,2)+1 >= batasy) sp_x(loop,2)= sp_x(loop,2)-1; end; if (BW((sp_x(loop,1)+1),(sp_x(loop,2)-1))==0) k=k+1; sp_x(k,2)=sp_x(loop,2)-1; sp_x(k,1)=sp_x(loop,1)+1; BW((sp_x(loop,1)+1),(sp_x(loop,2)-1))=1; end end

if ((sp_x(loop,1)-1)>=1 && ((sp_x(loop,2)+1)>=1)) if (sp_x(loop,1)+1 >= batasx) sp_x(loop,1)= sp_x(loop,1)-1; end; if (sp_x(loop,2)+1 >= batasy) sp_x(loop,2)= sp_x(loop,2)-1; end; if (BW((sp_x(loop,1)-1),(sp_x(loop,2)+1))==0) k=k+1; sp_x(k,2)=sp_x(loop,2)+1; sp_x(k,1)=sp_x(loop,1)-1; BW((sp_x(loop,1)-1),(sp_x(loop,2)+1))=1; end end if ((sp_x(loop,1))>=1 && ((sp_x(loop,2)+1)>=1)) if (sp_x(loop,1)+1 >= batasx) sp_x(loop,1)= sp_x(loop,1)-1; end; if (sp_x(loop,2)+1 >= batasy) sp_x(loop,2)= sp_x(loop,2)-1; end; if (BW((sp_x(loop,1)),(sp_x(loop,2)+1))==0) k=k+1; sp_x(k,2)=sp_x(loop,2)+1; sp_x(k,1)=sp_x(loop,1); BW((sp_x(loop,1)),(sp_x(loop,2)+1))=1; end end

L16 if ((sp_x(loop,1)+1)>=1 && ((sp_x(loop,2)+1)>=1)) if (sp_x(loop,1)+1 >= batasx) sp_x(loop,1)= sp_x(loop,1)-1; end; if (sp_x(loop,2)+1 >= batasy) sp_x(loop,2)= sp_x(loop,2)-1; end; if (BW((sp_x(loop,1)+1),(sp_x(loop,2)+1))==0) k=k+1; sp_x(k,2)=sp_x(loop,2)+1; sp_x(k,1)=sp_x(loop,1)+1; BW((sp_x(loop,1)+1),(sp_x(loop,2)+1))=1; end end if ((sp_x(loop,1)+1)>=1 && ((sp_x(loop,2))>=1)) if (sp_x(loop,1)+1 >= batasx) sp_x(loop,1)= sp_x(loop,1)-1; end; if (sp_x(loop,2)+1 >= batasy) sp_x(loop,2)= sp_x(loop,2)-1; end; if (BW((sp_x(loop,1)+1),(sp_x(loop,2)))==0) k=k+1; sp_x(k,2)=sp_x(loop,2); sp_x(k,1)=sp_x(loop,1)+1; BW((sp_x(loop,1)+1),(sp_x(loop,2)))=1; end end if ((sp_x(loop,1)-1)>=1 && ((sp_x(loop,2))>=1)) if (sp_x(loop,1)+1 >= batasx) sp_x(loop,1)= sp_x(loop,1)-1; end; if (sp_x(loop,2)+1 >= batasy) sp_x(loop,2)= sp_x(loop,2)-1; end; if (BW((sp_x(loop,1)-1),(sp_x(loop,2)))==0) k=k+1; sp_x(k,2)=sp_x(loop,2); sp_x(k,1)=sp_x(loop,1)-1; BW((sp_x(loop,1)-1),(sp_x(loop,2)))=1; end end if (loop>=prev_size) break; end end % end of while if (prev_size==size(sp_x,1)) break; end end % end of while for loop=1:size(sp_x,1) C(sp_x(loop,1),sp_x(loop,2))=0; end C=uint8(C);

L17

LAMPIRAN FIGURE MATLAB 7.1 APLIKASI SEGMENTASI CITRA SPOT DENGAN MENGGUNAKAN PENDEKATAN FUZZY C-MEANS

L18

LAMPIRAN CITRA

L19

L20

L21

L22

L23