Binary Segmentation is a computer vision technique that separates an image into two classes, usually distinguishing a target object or region from its background. Instead of assigning every pixel to one of many categories, the model produces a binary mask that identifies whether each pixel belongs to the area of interest. This makes it useful when the primary goal is to isolate a specific structure rather than classify multiple objects.
A common application is medical imaging, where binary segmentation can outline a tumor, organ, or lesion from surrounding tissue. It is also used in industrial inspection to isolate surface defects and in robotics to distinguish objects from their surroundings. Deep learning models such as U-Net and encoder-decoder architectures are frequently used for this task.
