Computer Vision

This book is an introduction to the broad field of computer vision. Without a doubt, machines can be built to see; for example, machines inspect millions of light bulb filaments and miles of fabric each day. Automatic teller machines (ATMs) have been built to scan the human eye for user identification and cars have been driven by a computer using camera input. This chapter introduces several important problem areas where computer vision provides solutions. After reading this chapter, you should have a broad view of some problems and methods of computer vision.

1 Definition

The goal of computer vision is to make useful decisions about real physical objects and scenes based on sensed images.

In order to make decisions about real objects, it is almost always necessary to construct some description or model of them from the image. Because of this, many experts will say that the goal of computer vision is the construction of scene descriptions from images. Although our study of computer vision is problem-oriented, fundamental issues will be addressed. Critical issues raised in this chapter and studied in the remainder of the text include the following.

Sensing: How do sensors obtain images of the world? How do the images encode properties of the world, such as material, shape, illumination and spatial relationships?
Encoded Information: How do images yield information for understanding the 3D world, including the geometry, texture, motion, and identity of objects in it?
Representations: What representations should be used for stored descriptions of objects, their parts, properties and relationships?
Algorithms: What methods are there to process image information and construct descriptions of the world and its objects?