Skip to main content
main-content
Top

About this book

This textbook provides an accessible general introduction to the essential topics in computer vision. Classroom-tested programming exercises and review questions are also supplied at the end of each chapter. Features: provides an introduction to the basic notation and mathematical concepts for describing an image and the key concepts for mapping an image into an image; explains the topologic and geometric basics for analysing image regions and distributions of image values and discusses identifying patterns in an image; introduces optic flow for representing dense motion and various topics in sparse motion analysis; describes special approaches for image binarization and segmentation of still images or video frames; examines the basic components of a computer vision system; reviews different techniques for vision-based 3D shape reconstruction; includes a discussion of stereo matchers and the phase-congruency model for image features; presents an introduction into classification and learning.

Table of Contents

1. Image Data

Abstract
This chapter introduces basic notation and mathematical concepts for describing an image in a regular grid in the spatial domain or in the frequency domain. It also details ways for specifying colour and introduces colour images.
Reinhard Klette

2. Image Processing

Abstract
This chapter introduces basic concepts for mapping an image into an image, typically used for improving image quality or for purposes defined by a more complex context of a computer vision process.
Reinhard Klette

3. Image Analysis

Abstract
This chapter provides topologic and geometric basics for analysing image regions, as well as two common ways for analysing distributions of image values. It also discusses line and circle detection as examples for identifying particular patterns in an image.
Reinhard Klette

4. Dense Motion Analysis

Abstract
This chapter discusses optic flow, the standard representation in computer vision for dense motion. Every pixel is labelled by a motion vector, indicating the change in image data from time t to time t+1. Sparse motion analysis (also known as tracking) will be a subject in Chap. 9.
Reinhard Klette

5. Image Segmentation

Abstract
In this chapter we explain special approaches for image binarization and segmentation of still images or video frames, in the latter case with attention to ensuring temporal consistency. We discuss mean-shift segmentation in detail. We also provide a general view on image segmentation as (another) labelling example in computer vision, introduce segmentation this way from an abstract point of view, and discuss belief-propagation solutions for this labelling framework.
Reinhard Klette

6. Cameras, Coordinates, and Calibration

Abstract
This chapter describes three basic components of a computer vision system. The geometry and photometry of the used cameras needs to be understood (to some degree). For modelling the projective mapping of the 3D world into images, and for the steps involved in camera calibration, we have to deal with several coordinate systems. By calibration we map recorded images into normalized (e.g. geometrically rectified) representations, thus simplifying subsequent vision procedures.
Reinhard Klette

7. 3D Shape Reconstruction

Abstract
This chapter describes three different techniques for vision-based reconstruction of 3D shapes. The use of structured lighting is a relatively simple but accurate method. Stereo vision might be called the 3D shape-reconstruction method in computer vision; its actual stereo-matching challenges are a subject in the following chapter; here we only discuss how results of stereo matching are used to derive 3D shape. Finally, as an alternative technique, we briefly describe shading-based 3D-shape understanding.
Reinhard Klette

8. Stereo Matching

Abstract
This chapter discusses the search for corresponding pixels in a pair of stereo images. We consider at first correspondence search as a labelling problem, defined by data and smoothness error functions, but also by the applied control structure. We describe belief-propagation stereo and semi-global matching. Finally we also discuss how to evaluate the accuracy of stereo-matching results on real-world input data, particularly on stereo video data.
Reinhard Klette

9. Feature Detection and Tracking

Abstract
This chapter describes the detection of keypoints and the definition of descriptors for those; a keypoint and a descriptor define a feature. The given examples are SIFT, SURF, and ORB, where we introduce BRIEF and FAST for providing ORB. We discuss the invariance of features in general, and of the provided examples in particular. The chapter also discusses three ways for tracking features: KLT, particle filter, and Kalman filter.
Reinhard Klette

10. Object Detection

Abstract
This final chapter provides an introduction into classification and learning, with a detailed description of basic AdaBoost and the use of random forests. These concepts are illustrated by applications for face detection, and for pedestrian detection, respectively.
Reinhard Klette
Additional information