Differentiate between oval and circle shaped images using MATLAB

I am using the following code to try to differentiate two types of cells. One is circular, and the other is oval to oblong. I have about 100 images of each cell type, and have attached one each .jpg image to this question.

When I run the following code for Ck1.jpg, the rgb2gray image shows that my cell no longer has a continuous boundary. As a result, the code cannot tell what shape this cell has. Unfortunately, all the images I have are not high enough resolution. I was wondering if there was someway to interpolate the cell boundary based on however much is preserved after converting rgb image to gray.

I would greatly appreciate any help or feedback. Thank you!

clear; close all; clc;

ipath = 'D:\3) Candida yeast measurement\4) Multi Frequency Measurement - 100 Data Points\5) Ca versus Ck\';
fid = 'Ck1.jpg';

image_read = imread(strcat(ipath,fid));
figure (1)
image_RGB = imshow(image_read);

image_gray = rgb2gray(image_read);
B_W = imbinarize(image_gray);
figure (2)
image_B_W = imshow(B_W);

BW = bwareaopen(B_W,10);
figure (3)
image_BW = imshow(BW);

[B,L] = bwboundaries(BW,'holes');

figure (6)
imshow(label2rgb(L,@jet,[.5 0 .5]))

hold on
for k = 1:length(B)
    boundary = B{k};
    plot(boundary(:,2),boundary(:,1),'w','LineWidth',2)
end

stats = regionprops(L,'Area','Centroid');
threshold = 0.95;
% loop over the boundaries
for k = 1:length(B)
    
    % obtain (X,Y) boundary coordinates corresponding to label 'k'
    boundary = B{k};
    
    % compute a simple estimate of the object's perimeter
    delta_sq = diff(boundary).^2;
    perimeter = sum(sqrt(sum(delta_sq,2)));
    
    % obtain the area calculation corresponding to label 'k'
    area = stats(k).Area;
    
    % compute the roundness metric
    metric = 4*pi*area/perimeter^2;
    
    % display the results
    metric_string = sprintf('%2.2f',metric);
    
    % mark objects above the threshold with a black circle
    if metric > threshold
        metric = 0;
        %         centroid = stats(k).Centroid;
        %         plot(centroid(1),centroid(2),'ko');
    end
    
    text(boundary(1,2)-35,boundary(1,1)+13,metric_string,'Color','r',...
        'FontSize',14,'FontWeight','bold')
    
end

Answer 1

[Simon Chan]

You may try function bwconvhull by adding the following 2 lines.

However, I am not sure how robust it is for other pictures.

B_W = imbinarize(image_gray);
CH = bwconvhull(~B_W);                          % use function bwconvhull
B_W = ~CH;                                      % Convert back to B_W

[Neelima Dahal]

Thank you for responding to my question. While your solution works perfectly for Ck1.jpg file, it does not for the other attached image Ca1.jpg. I also tried it on other oval cell images I have, but it doesn't work for them either.

Answer 2

[Image Analyst]

Assuming you can get a mask for the blobs, use the 'Eccentricity' option in regionprops

props = regionprops(mask, 'Eccentricity');

Alternatively use bwferet().