Handbook of Machine and Computer Vision, 2nd Edition

Chapter 9Machine Vision Algorithms

Carsten Steger

MVTec Software GmbH, Arnulfstraße 205, 80634 München, Germany

In the previous chapters, we have examined the different hardware components that are involved in delivering an image to the computer. Each of the components plays an essential role in the machine vision process. For example, illumination is often crucial to bring out the objects we are interested in. Triggered frame grabbers and cameras are essential if the image is to be captured at the right time with the right exposure. Lenses are important for acquiring a sharp and aberration-free image. Nevertheless, none of these components can “see,” that is, extract the information we are interested in from the image. This is analogous to human vision. Without our eyes we cannot see. Yet, even with eyes we cannot see anything without our brain. The eye is merely a sensor that delivers data to the brain for interpretation. To this analogy a little further, even if we are myopic, we can still see – only worse. Hence, we can see that the processing of the images delivered to the computer by the sensors is truly the core of machine vision. Consequently, in this chapter, we will discuss the most important machine vision algorithms.

9.1 Fundamental Data Structures

Before we can delve into the study of the machine vision algorithms, we need to examine the fundamental data structures that are involved in machine vision applications. Therefore, in this section we will take a look ...