Read more
The human ability to recognize objects on various backgrounds is amazing. Many times, industrial image processing tried to imitate this ability by its own techniques. This book discusses the recognition of defects on free-form edges and - homogeneous surfaces. My many years of experience has shown that such a task can be solved e?ciently only under particular conditions. Inevitably, the following questions must be answered: How did the defect come about? How and why is a person able to recognize a speci?c defect? In short, one needs an analysis of the process of defect creation as well as an analysis of its detection. As soon as the principle of these processes is understood, the processes can be described mathematically on the basis of an appropriate physical model and can then be captured in an algorithm for defect detection. This approach can be described as "image processing from a physicist's perspective". I have successfully used this approach in the development of several industrial image processingsystemsandimprovedupontheminthecourseoftime.Iwouldlike to present the achieved results in a hands-on book on the basis of edge-based algorithms for defect detection on edges and surfaces. I would like to thank all who have supported me in writing this book.
List of contents
Introduction.- Edge detection.- Defect recognition on an edge.- Defect Recognition on an inhomogenious high-contrast surface.- Defect Recognition on an inhomogenious structured surface.- Grading, optimizing, winning!- A turbo-method of defect recognition.- Before an image processing system comes to use.
About the author
Born on February 26, 1954 in Kiev, Ukraine.
1971 – 1976 Study at the State University of Woronezh, Russia. Graduation: Qualified physicist with award.
1977 – 1992 Scientific research at the Institute of Material Problems of the Science Academy, Kiev, Ukraine. Area of studies: plasma physics, composed materials.
1987 Promotion with the grade: Doctor of Engineering Sciences
1992 – 2006 hema electronic GmbH, Aalen, Department: machine vision, Development engineer, Project director. Main topic: Algorithmics for defect recognition on edges and surfaces
Since April 2006 Thermosensorik GmbH, Erlangen, Manager Software Engineering, IP Expert
Summary
The human ability to recognize objects on various backgrounds is amazing. Many times, industrial image processing tried to imitate this ability by its own techniques. This book discusses the recognition of defects on free-form edges and - homogeneous surfaces. My many years of experience has shown that such a task can be solved e?ciently only under particular conditions. Inevitably, the following questions must be answered: How did the defect come about? How and why is a person able to recognize a speci?c defect? In short, one needs an analysis of the process of defect creation as well as an analysis of its detection. As soon as the principle of these processes is understood, the processes can be described mathematically on the basis of an appropriate physical model and can then be captured in an algorithm for defect detection. This approach can be described as “image processing from a physicist’s perspective”. I have successfully used this approach in the development of several industrial image processingsystemsandimprovedupontheminthecourseoftime.Iwouldlike to present the achieved results in a hands-on book on the basis of edge-based algorithms for defect detection on edges and surfaces. I would like to thank all who have supported me in writing this book.
Additional text
Aus den Rezensionen:
“... Die HeIligkeitsverhältnisse an der Kante einer Materialbeschädigung werden als Gauß'sche Verteilung einer Strahlung interpretiert und in einem physikalischen Modell erfasst. Basierend auf diesem Modell wurden neue Methoden entwickelt, mit denen unterschiedliche Fehlertypen unabhängig von den Helligkeitsbedingungen eines aufgenommenen Bildes ermittelt werden können. ... Zahlreiche Anwendungsbeispiele veranschaulichen die theoretischen Ausführungen.“ (in: QZ Qualität und Zuverlässigkeit, March/2010, Issue 3, S. 13)
Report
Aus den Rezensionen: "... Die HeIligkeitsverhältnisse an der Kante einer Materialbeschädigung werden als Gauß'sche Verteilung einer Strahlung interpretiert und in einem physikalischen Modell erfasst. Basierend auf diesem Modell wurden neue Methoden entwickelt, mit denen unterschiedliche Fehlertypen unabhängig von den Helligkeitsbedingungen eines aufgenommenen Bildes ermittelt werden können. ... Zahlreiche Anwendungsbeispiele veranschaulichen die theoretischen Ausführungen." (in: QZ Qualität und Zuverlässigkeit, March/2010, Issue 3, S. 13)
