In machine learning, image segmentation means dividing a digital image into different segments or regions, each made of a group of pixels. The purpose is to make the image simpler to analyze and more meaningful. Image segmentation is often used to find objects and boundaries (like lines and curves) in images. To do this, annotators usually use instruments such as bounding boxes or polygons. During annotation, each pixel is given a label, so that pixels with the same label share certain characteristics.
Back to all