hiii....I want to separate the portion from the image(I attached image below). Plz help me in separating that portion

In one image diseased portion is there and other image the boundary is detected. My problem is how to extract only the diseased portion from the boundary image.

Answer 1

[Image Analyst]

There is nothing in MATLAB to do that (dental caries segmentation) at a high level. You'll have to build it up from lower level functions. Look to the published image analysis or dental literature for algorithms that you can code up.

[Bhagyashree Shivpuje]

Okay. Now tell me one thing. I have calculated the area by using regionprops instruction. in that I want the portion to be displayed which has maximum area. How can I do that? my code is written here.

clc;

clear all;

close all;

imtool close all;

workspace;

%% Read input image

[I,pathname] = uigetfile('*.jpg','Select an input');

str = strcat(pathname, I);

A = imread(str);

A = rgb2gray(A);

figure, imshow(A); title('Original grayscale Image');

%% Thresholding operation based on given range

A1 = Thresholding(A);

figure, imshow(A1); title('thresholded image');

%% Segmentation using connected component labeling

A1 = uint8(A1);

A2 = immultiply(A,A1);

figure, imshow(A2,[]); title('Multiplied Image');

A3 = A2;

A2 = logical(A2);

cc = bwconncomp(A2, 8);

n = cc.NumObjects;

Area = zeros(n,1);

Perimeter = zeros(n,1);

%% Feature Extraction

k = regionprops(A2,'Area', 'Perimeter', 'MajorAxis', 'MinorAxis');

for i = 1:n

 Area(i) = k(i).Area;

 Perimeter(i) = k(i).Perimeter;

end

%% Feature Storage

graindata(1,1) = mean(Area);

graindata(2,1) = mean(Perimeter);

%% Boundary Detection using Canny Operator

BW = edge(A3, 'canny',0.3);

figure, imshow(BW); title('Boundary of the multiplied Image');

-----------------------------------------------------------------

%function thresholding used in main program

function im = Thresholding(A)

[r,c] = size(A);

im = zeros(r,c);

for i = 1:r

    for j = 1:c;

        if A(i,j)>45 & A(i,j)<100

            im(i,j) = 1;

        end

    end

end

im = bwareaopen(im,5);

im = imfill(im, 'holes');

[Image Analyst]

labeledImage = bwlabel(A2);
props = regionsprop(labeledImage, 'Area', 'Perimeter', 'MajorAxis', 'MinorAxis');
allAreas = [k.Area];
[maxArea, indexOfMax] = max(allAreas);
binaryImageMaxOnly = ismember(labeledImage, indexOfMax) > 0;

Or, a lot more simply if you have a recent version of MATLAB:

binaryImageMaxOnly = bwareafilt(A2);

[Bhagyashree Shivpuje]

Thank you..it works..

[Image Analyst]

Then can you please "Accept this answer"?