Masking out black area from the created binary mask

Hi all,

I am new to Image processing and I am doing the image processing to mask out the ROI.

From the result image, As I already defined the ROI (from this ref: https://in.mathworks.com/help/images/detecting-a-cell-using-image-segmentation.html?s_tid=srchtitle_cell%20detection_2), I want to mask out again the middel part of the image that is black and gray area. Is it possible to segment that black area from masked out image? I want to convert that maskedout image into again binary image and then want to count the black pixels and white pixels inside the mask.

I = imread('2022-05-10_11h25_21.png');
imshow(I)

I masked out the original image with the binary mask and then converted it into binary image. But here what I want to do is to detect only black area which is inside the white shape. Later I want to fill it up that black part to get rid of those small white dots. see the sample below.

I2 = imread('2022-05-10_11h36_40.png');
imshow(I2)

Please help me which method should I follow to get the results? Or is it possible to solve such problems? Thank you in advance.

[Dhrupad Prajapati]

@Image Analyst Please help me out with this. Thank you in advance.

[Image Analyst]

Not sure how I missed this a year and a half ago. Not sure what you want but it sounds like you wanted the dark gray rectangle with the white circle to be your ROI. Then you wanted to binarize that to turn it into black and white and count the number of black and white pixels. Then you wanted to make the small white blobs inside the gray rectangle (not including the large white circle) into black, and again count the number of clack and white pixels. Is that what you wanted to do? If so, could you attach the original (not screenshot) gray scale image if it is still a need for you?

Answer 1

[Githin George]

Hello,

I understand you are trying to binarize the left image from 2nd attachment and fill the small white dots inside the right image, with black color.

The following code can be used to achieve the above requirement:

[X,map] = imread('yourImage.png'); 
% Make your 2D grayscale image into a binary image 
X = X(:,:,1); 
X = imbinarize(X); 
imshow(X) 

% Fill the small 'white' holes with less than 400 pixel connections 
Ifill = bwareaopen(X, 400); 
figure 
montage({X,Ifill}) 

The “imbinarize” function converts your Grayscale image into a Binary Image and “bwareaopen” fills the white dots inside the the image. Please, adjust parameters for both these functions to obtain the desired result.

Kindly refer to the following documentation links for the above-mentioned functions below,

https://www.mathworks.com/help/images/ref/imbinarize.html#bu1w0rc-1-T

https://www.mathworks.com/help/images/ref/bwareaopen.html

Below is the output obtained using the above code for reference,

I hope this helps.