The reconstruction of object surfaces is a special discipline in Computer
Vision. This discipline is directed towards the recovery of object
shapes or recovery of distances between the camera and objects in a scene.
This textbook is recommended for a one-semester university course at third
or fourth year level in this field of surface reconstruction, for example
in bachelor or master programs in Computer Science, in Applied Mathematics,
or in Engineering.
The book provides a selection of fundamentals, often illustrated and explained by examples, and of comprehensible algorithmic solutions. Some of the recent results of the authors' research are included in this text. The exercises which follow each chapter are not only theoretic questions but are also directed towards practical applications. For these exercises it is recommended that the reader has a software system available which allows at least pixelwise write/read access to picture data. This will allow the reader to experience the discussed algorithms. Some image processing systems can be downloaded as public domain software from the Internet. Generally these systems are characterized by a certain number of basic procedures, e.g. picture enhancement, edge detection or picture representation. Information about such systems is available on the Internet.
The material in this book has been used during the last eight years for various courses in Computer Science at the University of Auckland, at the University of Otago, Dunedin, and at the Berlin Technical University. Section 4.3.2 is based on a text provided by Georgy Gimel'farb (University of Auckland). Alan McIvor (Industrial Research Ltd., Auckland) provided two figures for publication. We want to thank those colleagues who contributed to the manuscript with their comments: Ryszard Kozera (UWA Perth), Richard Lobb (University of Auckland), Volker Rodehorst (TU Berlin), Horst Völz (FU Berlin), and Piero Zamperoni (TU Braunschweig). Many of our students have contributed to the creation of this book. In particular, we would like to thank Petra Bonfert, Peter Handschack, Tapani Hegewald, Wolfgang Huber, Richard Lewis-Shell, Greg Maddigan, Dirk Mehren, Arno Mitritz, Detlev Rumpel, Kathrin Spiller, and Matthias Teschner.
Reinhard Klette, Karsten Schlüns, and Andreas Koschan