How can I train a network to use clustering on a set of images, to extract an object?

I have used color based segmentation to roughly isolate said an object (for the purpose of this question lets say the object is an irregular shape, that is perdominantly green but has a couple of red patches near the border).

https://au.mathworks.com/help/images/color-based-segmentation-using-k-means-clustering.html

I would like to be able to further this by 1) Isolating the entire object not just the green part, 2) Do this for a larger set of images with the same type of shape.

My understanding is that the best way to approach this method would be to use Machine Learning such as the clustering I have used, to train a network of some sort to detect the object in any set of given images. I'm just unsure of how to approach this problem.

Answer 1

[Shashank Gupta]

Hi Tom,

Object segmentation using K-means clustering algorithm works only when the object in the image is differentiable quite well on the basis of either intensity or color, this are very high-level feature and we cannot generalize it to work on general images. Instead if you have access to segmentation map of some images (i.e. ground truth), try using semantic segmentation algorithm, I am attaching some example which may give you a good head start.

https://www.mathworks.com/help/vision/semantic-segmentation.html

In case, if you don’t have access to ground truths, then there are recent work done Autoencoder, you can read upon them and try training Autoencoder and use the feature vector to segment it in to different classes. It is one of classical way of dealing with unsupervised Image segmentation.

I hope it helps you or at least give you an idea where to look for.