How to Separate Connected Objects in an Image

I am trying to count the number of objects in a picture using bwconncomp, but the results are incorrect because some of the regions blend together. I used the image from https://www.publicdomainpictures.net/pictures/290000/velka/field-of-flowers-1553807209QZG.jpg to illustrate my problem.

This is the original image, and I'd like to write code that counts the total number of flowers in the picture.

I used the following code:

image = imread(fileName);

%Convert to grayscale
imageGray = rgb2gray(image);

%Remove noise
SE = strel('disk', 5);
afterOpening = imopen(image, SE); 

%Convert to binary image and filter out small objects
imageBin = imbinarize(afterOpening);
imageBin = bwareaOpen(imageBin, 300);

%Count the number of objects
CC = bwconncomp(imageBin);
numObjects = CC.NumObjects;

I am pretty sure that bwconncomp will count the three flowers highlighted above as one region, thereby skewing the count in numObjects. Is there a way that I can get Matlab to recognize the highlighted box as having 3 objects instead of 1?

Answer 1

[Matt J]

This might help:

https://blogs.mathworks.com/steve/2006/06/02/cell-segmentation/

https://www.mathworks.com/company/newsletters/articles/the-watershed-transform-strategies-for-image-segmentation.html