O'Reilly logo

A Practical Introduction to Computer Vision with OpenCV by Kenneth Dawson-Howe

Stay ahead with the world's most comprehensive technology and business learning platform.

With Safari, you learn the way you learn best. Get unlimited access to videos, live online training, learning paths, books, tutorials, and more.

Start Free Trial

No credit card required

8 Recognition

Towards the end of the previous chapter, one of the feature detectors (SIFT) allowed us to recognise objects in images through the comparison of features from a known instance of an object (or objects) and a scene which might (or might not) contain the object(s) in question. This topic of recognition (Cyganek, 2013) is central to most advanced computer vision systems. For example, we may want to

  1. distinguish different types of object (e.g. people vs. cars vs. bicycles);
  2. recognise specific individuals;
  3. automatically read the license plate of a car;
  4. locate specific objects to be manipulated by a robot;
  5. recognise a page of a book, or a specific painting, or building, and so on, so that we can augment reality in some way;
  6. locate eyes so that we can track them to provide advanced user interfaces;
  7. classify objects (such as chocolates on a production line) so that a robot can pick them up and place them in the correct location.

This chapter continues this topic and presents a number of different ways of recognising objects. These are appropriate in different situations and are only a small sample of the wide range of techniques which have been presented in the computer vision literature. This chapter concludes with an introduction to the area of performance assessment.

8.1 Template Matching

Template matching (Brunelli, 2009) is very simply a technique where a sub-image is searched for within an image. The sub-image represents some object of interest – which is effectively ...

With Safari, you learn the way you learn best. Get unlimited access to videos, live online training, learning paths, books, interactive tutorials, and more.

Start Free Trial

No credit card required