It is possible to do image segmentation of images in MATLAB.
Refer to this Link to see the various options available: Image Segmentation
As you want to segment these image pixels into 5 categories , you could use either of the two types of image segmentation techniques depending on your need.
1 ) Semantic segmentation: In this Objects shown in an image are grouped based on defined categories. It treats multiple objects within a single category as one entity. For Example: A street would be segmented by “pedestrians,” “bikes,” “vehicles,” “sidewalks,” and so on.
You can also refer to this example which shows how to create and train a simple semantic segmentation
2) Instance segmentation: This is a refined version of semantic segmentation. It detects the instances of each category. For Example : Categories like “vehicles” are split into “cars,” “motorcycles,” “buses,” and so on.
You can refer to this Link to know more about how to get started: Instance Segmentation
You should use the grayscale images for the image segmentation algorithm and Ground truths to access the performance of image segmentation where each pixel of the image is compared with corresponding ground truth data.The objective is to minimize the error between the segmented image and ground truth information.
