find multiple objects and their number in an image

2 views (last 30 days)
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.
  3 Comments
KALYAN ACHARJYA
KALYAN ACHARJYA on 12 Mar 2018
Edited: KALYAN ACHARJYA on 12 Mar 2018
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
Liliana Malik on 12 Mar 2018
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

Sign in to comment.

Accepted Answer

Image Analyst
Image Analyst on 12 Mar 2018
See my image segmentation tutorial in my File Exchange.
  1 Comment
Image Analyst
Image Analyst on 13 Mar 2018
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.

Sign in to comment.

More Answers (0)

Categories

Find more on Recognition, Object Detection, and Semantic Segmentation in Help Center and File Exchange

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!