Can computers see?Can computers understand visual data?

This article presents a brief introduction to the problem of computer vision. Computers, using cameras and other sensors as input devices record digital images in its memory. The computer programs must interpret these images to come up with a description of the scene. The computer vision problem can be stated simply as follows: Given a two-dimensional image, infer the objects that produced it, including their shapes, locations, colors and sizes. The concepts of low level, mid level image processing and high-level image understanding are presented. Various application areas including satellite imaging, assembly line manufacturing, handwriting and face recognition are discussed.