Low-level image processing

Abstract

An image that is originally acquired by a camera is called a raw image, and raw images must be processed in some way to extract the useful information that they contain. The first step in this procedure is called pre-processing, and it is done to prepare the image for further processing at higher levels. The two primary goals of image preprocessing are to eliminate unwanted attributes in the image that would aggravate subsequent processing steps, and to extract primitive features that represent the pertinent information present in the image. Unwanted or undesirable image attributes include noise, areas of featureless space, or areas where the features can be ignored. Primitive features include edges, corners or vertices, lines and line junctions. These features typically represent the boundaries or surface details of objects that are valuable when those objects are to be located and identified in an image.