find multiple objects and their number in an image

I am working on a project that classifies beans in an image how can i do that and how can i tell the number of beans of each type.

[Akira Agata]

If possible, could you share a sample image?

[KALYAN ACHARJYA]

When you classify it based on image segmentation (Apply various variable thresholding techniques) and count the number of objects within an image.

 [label_Image, numberofObject]=bwlabel(segmented_binaryImage);

[Liliana Malik]

this is a sample image it's not exactly what i'm going to do cause the exact number of beans for some types is uncountable in this image, the image i'm going to work with will have less beans

Answer 1

[Image Analyst]

See my image segmentation tutorial in my File Exchange.

[Image Analyst]

Create a feature vector for each pixel that includes the mean hue, saturation, and value in a window around the pixel plus the standard deviation of the values in the window. Use rgb2hsv() and conv2() and stdfilt().

hsvImage = rgb2hsv(rgbImage);
hImage = hsvImage(:, :, 1);
sImage = hsvImage(:, :, 2);
vImage = hsvImage(:, :, 3);
windowsSize = 21; % Some odd number.
kernel = ones(windowSize, windowSize)/windowSize^2;
localMeanH = conv2(hImage, kernel, 'same');
localMeanS = conv2(sImage, kernel, 'same');
localMeanV = conv2(vImage, kernel, 'same');
% Now use stdfilt
localSDH = stdfilt(hImage, ..... etc.

I'm sure you can figure out what to do next.