Classification in remote sensing refers to the process of converting raw pixel data into meaningful information. It involves categorizing all pixels in an image into land cover classes or themes. This process is primarily used to identify and analyze specific features in an image such as vegetation, water bodies, urban areas, etc. It can be done through supervised classification (where you know what to look for) or unsupervised classification (where the software identifies classes or clusters on its own).

Supervised classification in remote sensing involves using known, labeled examples (training data) to classify areas of unknown identity. The analyst specifies the types or classes based on a certain number of pixel locations that are already known, such as urban area, forest, water, etc.