Pengertian Ektraksi Citra dan Contohnya



Ektraksi citra -Pemrograman Matlab



Pengetrian - Ekstraksi ciri citra merupakan tahapan mengekstrak ciri/informasi dari objek di dalam citra yang ingin dikenali/dibedakan dengan objek lainnya. Ciri yang telah diekstrak kemudian digunakan sebagai parameter/nilai masukan untuk membedakan antara objek satu dengan lainnya pada tahapan identifikasi/ klasifikasi. Ciri yang telah diekstrak kemudian digunakan sebagai parameter/nilai masukan untuk membedakan antara objek satu dengan lain.


Untuk membedakan bentuk objek satu dengan objek lainnya, dapat menggunakan parameter yang disebut dengan ‘eccentricity’. Eccentricity merupakan nilai perbandingan antara jarak foci ellips minor dengan foci ellips mayor suatu objek. Eccentricity memiliki rentang nilai antara 0 hingga 1. Objek yang berbentuk memanjang/mendekati bentuk garis lurus, nilai eccentricitynya mendekati angka 1, sedangkan objek yang berbentuk bulat/lingkaran, nilai eccentricitynya mendekati angka 0. 



Untuk fungsi nya sebagai berikut :

clc;clear;close all;

 

RGB = imread('figure2.jpeg');

imshow(RGB);

 

I = rgb2gray(RGB);

threshold = graythresh(I);

bw = im2bw(I,threshold);

bw = bwareaopen(bw,30);

 

se = strel('disk',2);

bw = imclose(bw,se);

bw = imfill(bw,'holes');

[B,L] = bwboundaries(bw,'noholes');

 

hold on

for k = 1:length(B)

    boundary = B{k};

    plot(boundary(:,2), boundary(:,1), 'w', 'LineWidth', 2)

end

 

stats = regionprops(L,'Area','Centroid','Eccentricity');

 

for k = 1:length(B)

    boundary = B{k};

    delta_sq = diff(boundary).^2;

    perimeter = sum(sqrt(sum(delta_sq,2)));

    area = stats(k).Area;

    eccentricity = stats(k).Eccentricity;

    metric = 4*pi*area/perimeter^2;

     

    centroid = stats(k).Centroid;

    text(centroid(1),centroid(2)-16,num2str(k),'Color','r',...

        'FontSize',20,'FontWeight','bold');

    disp('===================================')

    disp(strcat(['Object number = ', num2str(k)]))

    disp(strcat(['Area = ',num2str(area)]))

    disp(strcat(['Perimeter = ',num2str(perimeter)]))

    disp(strcat(['Metric = ',num2str(metric)]))

    disp(strcat(['Eccentricity = ',num2str(eccentricity)]))

     

    if metric>0.8

        text(centroid(1)-16,centroid(2)+16,'Bulat','Color','r',...

            'FontSize',20,'FontWeight','bold');

    else

        text(centroid(1)-16,centroid(2)+16,'Tidak Bulat','Color','b',...

            'FontSize',20,'FontWeight','bold');

    end

end

 

RGB = imread('figure2.jpeg');

subplot ()

imshow(RGB);

 

I = rgb2gray(RGB);

threshold = graythresh(I);

bw = im2bw(I,threshold);

bw = bwareaopen(bw,30);

 

se = strel('disk',2);

bw = imclose(bw,se);

bw = imfill(bw,'holes');

[B,L] = bwboundaries(bw,'noholes');

 

hold on

for k = 1:length(B)

    boundary = B{k};

    plot(boundary(:,2), boundary(:,1), 'w', 'LineWidth', 2)

end

 

stats = regionprops(L,'Area','Centroid','Eccentricity');

 

for k = 1:length(B)

    boundary = B{k};

    delta_sq = diff(boundary).^2;

    perimeter = sum(sqrt(sum(delta_sq,2)));

    area = stats(k).Area;

    eccentricity = stats(k).Eccentricity;

    metric = 4*pi*area/perimeter^2;

     

    centroid = stats(k).Centroid;

    text(centroid(1),centroid(2)-16,num2str(k),'Color','r',...

        'FontSize',20,'FontWeight','bold');

    disp('===================================')

    disp(strcat(['Object number = ', num2str(k)]))

    disp(strcat(['Area = ',num2str(area)]))

    disp(strcat(['Perimeter = ',num2str(perimeter)]))

    disp(strcat(['Metric = ',num2str(metric)]))

    disp(strcat(['Eccentricity = ',num2str(eccentricity)]))

     

    if metric>0.8

        text(centroid(1)-16,centroid(2)+16,'Bulat','Color','r',...

            'FontSize',20,'FontWeight','bold');

    else

        text(centroid(1)-16,centroid(2)+16,'Tidak Bulat','Color','b',...

            'FontSize',20,'FontWeight','bold');

    end

end


untuk hasil yang di dapat adalah sebagai berikut :



2. Ekstraksi Ciri Tekstur
Untuk membedakan tekstur objek satu dengan objek lainnya dapat menggunakan ciri statistik orde pertama atau ciri statistik orde dua. Ciri orde pertama didasarkan pada karakteristik histogram citra. Ciri orde pertama umumnya digunakan untuk membedakan tekstur makrostruktur (perulangan pola lokal secara periodik). Ciri orde pertama antara lain: mean, variance, skewness, kurtosis, dan entropy. Sedangkan ciri orde dua didasarkan pada probabilitas hubungan ketetanggaan antara dua piksel pada jarak dan orientasi sudut tertentu. Ciri orde dua umumnya digunakan untuk membedakan tekstur mikrostruktur (pola lokal dan perulangan tidak begitu jelas). Ciri orde dua antara lain: Angular Second MomentContrast, Correlation, Variance, Inverse Different Moment, dan Entropy.
Analisis tekstur juga dapat dilakukan dalam domain frekuensi antara lain menggunakan filter bank gabor. fanction nya yaitu :


function varargout = Texture_Analysis(varargin)

% TEXTURE_ANALYSIS MATLAB code for Texture_Analysis.fig

%      TEXTURE_ANALYSIS, by itself, creates a new TEXTURE_ANALYSIS or raises the existing

%      singleton*.

%

%      H = TEXTURE_ANALYSIS returns the handle to a new TEXTURE_ANALYSIS or the handle to

%      the existing singleton*.

%

%      TEXTURE_ANALYSIS('CALLBACK',hObject,eventData,handles,...) calls the local

%      function named CALLBACK in TEXTURE_ANALYSIS.M with the given input arguments.

%

%      TEXTURE_ANALYSIS('Property','Value',...) creates a new TEXTURE_ANALYSIS or raises the

%      existing singleton*.  Starting from the left, property value pairs are

%      applied to the GUI before Texture_Analysis_OpeningFcn gets called.  An

%      unrecognized property name or invalid value makes property application

%      stop.  All inputs are passed to Texture_Analysis_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 Texture_Analysis

 

% Last Modified by GUIDE v2.5 12-Oct-2022 12:59:32

 

% Begin initialization code - DO NOT EDIT

gui_Singleton = 1;

gui_State = struct('gui_Name',       mfilename, ...

                   'gui_Singleton',  gui_Singleton, ...

                   'gui_OpeningFcn', @Texture_Analysis_OpeningFcn, ...

                   'gui_OutputFcn',  @Texture_Analysis_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 Texture_Analysis is made visible.

function Texture_Analysis_OpeningFcn(hObject, eventdata, handles, varargin)

% This function has no output args, see OutputFcn.

% hObject    handle to figure

% eventdata  reserved - to be defined in a future version of MATLAB

% handles    structure with handles and user data (see GUIDATA)

% varargin   command line arguments to Texture_Analysis (see VARARGIN)

 

% Choose default command line output for Texture_Analysis

handles.output = hObject;

 

% Update handles structure

guidata(hObject, handles);

 

% UIWAIT makes Texture_Analysis wait for user response (see UIRESUME)

% uiwait(handles.figure1);

 

 

% --- Outputs from this function are returned to the command line.

function varargout = Texture_Analysis_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;

 

 

% --- Executes on button press in pushbutton1.

function pushbutton1_Callback(hObject, eventdata, handles)

% hObject    handle to pushbutton1 (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({'*.*'});

 

if ~isequal (filename,0)

    info = imfinfo(fullfile(pathname,filename));

    if info.BitDepth == 1

        msgbox('Citra masukkan harus citra RGB atau Grayscale');

        return

    elseif info.BitDepth == 8

        Img = imread (fullfile(pathname,filename));

        axes(handles.axes1)

        cla('reset')

        imshow(Img)

        title('Grayscale Image')

    else

        Img = rgb2gray(imread(fullfile(pathname,filename)));

        axes(handles.axes1)

        cla('reset')

        imshow(Img)

        title('Grayscale Image')

    end

else

    return

end

 

handles.Img = Img;

guidata(hObject,handles)

 

% --- Executes on button press in pushbutton2.

function pushbutton2_Callback(hObject, eventdata, handles)

% hObject    handle to pushbutton2 (see GCBO)

% eventdata  reserved - to be defined in a future version of MATLAB

% handles    structure with handles and user data (see GUIDATA)

Img = handles.Img;

pixel_dist = str2double(get(handles.edit1,'string'));

GLCM = graycomatrix(Img,'offset',[0 pixel_dist; -pixel_dist pixel_dist; pixel_dist 0 ; -pixel_dist -pixel_dist ]);

stats = graycoprops(GLCM,{'Contrast','Correlation','energy','homogeneity'});

 

Contrast = stats.Contrast;

Correlation = stats.Correlation;

Energy = stats.Energy;

Homogeneity = stats.Homogeneity;

 

data = get(handles.uitable1,'Data');

data{1,1} = num2str(Contrast(1));

data{1,2} = num2str(Contrast(2));

data{1,3} = num2str(Contrast(3));

data{1,4} = num2str(Contrast(4));

data{1,5} = num2str(mean(Contrast));

 

data{2,1} = num2str(Correlation(1));

data{2,2} = num2str(Correlation(2));

data{2,3} = num2str(Correlation(2));

data{2,4} = num2str(Correlation(4));

data{2,5} = num2str(mean(Correlation));

 

data{3,1} = num2str(Energy(1));

data{3,2} = num2str(Energy(2));

data{3,3} = num2str(Energy(3));

data{3,4} = num2str(Energy(4));

data{3,5} = num2str(mean(Energy));

 

data{4,1} = num2str(Homogeneity(1));

data{4,2} = num2str(Homogeneity(2));

data{4,3} = num2str(Homogeneity(3));

data{4,4} = num2str(Homogeneity(4));

data{4,5} = num2str(mean(Homogeneity));

 

set(handles.uitable1,'Data',data,'ForegroundColor',[0 0 0])

 

function edit1_Callback(hObject, eventdata, handles)

% hObject    handle to edit1 (see GCBO)

% eventdata  reserved - to be defined in a future version of MATLAB

% handles    structure with handles and user data (see GUIDATA)

 

% Hints: get(hObject,'String') returns contents of edit1 as text

%        str2double(get(hObject,'String')) returns contents of edit1 as a double

 

 

% --- Executes during object creation, after setting all properties.

function edit1_CreateFcn(hObject, eventdata, handles)

% hObject    handle to edit1 (see GCBO)

% eventdata  reserved - to be defined in a future version of MATLAB

% handles    empty - handles not created until after all CreateFcns called

 

% Hint: edit controls usually have a white background on Windows.

%       See ISPC and COMPUTER.

if ispc && isequal(get(hObject,'BackgroundColor'), get(0,'defaultUicontrolBackgroundColor'))

    set(hObject,'BackgroundColor','white');

end

 

 

% --- Executes during object creation, after setting all properties.

function pushbutton1_CreateFcn(hObject, eventdata, handles)

% hObject    handle to pushbutton1 (see GCBO)

% eventdata  reserved - to be defined in a future version of MATLAB

% handles    empty - handles not created until after all CreateFcns called

 

 

% --- Executes on key press with focus on pushbutton1 and none of its controls.

function pushbutton1_KeyPressFcn(hObject, eventdata, handles)

% hObject    handle to pushbutton1 (see GCBO)

% eventdata  structure with the following fields (see MATLAB.UI.CONTROL.UICONTROL)

%   Key: name of the key that was pressed, in lower case

%   Character: character interpretation of the key(s) that was pressed

%   Modifier: name(s) of the modifier key(s) (i.e., control, shift) pressed

% handles    structure with handles and user data (see GUIDATA)

 

 Berikut ini merupakan contoh aplikasi pemrograman gui matlab untuk analisis tekstur menggunakan metode Gray-Level Co-Occurrence Matrix (GLCM)

Gray-Level Co-Occurrence Matrix (GLCM) dengan pixel distance = 1





Gray-Level Co-Occurrence Matrix (GLCM) dengan pixel distance = 2










 

Komentar

Postingan populer dari blog ini

Menggubah Citra RGB ( Red Green Blue )

Mengubah citra RGB ke grayscale (abu-abu) dan ke BW (black and white)