• Home
  • Wolfgang Eckstein
Wolfgang Eckstein

Wolfgang Eckstein
MVTec Software GmbH · Strategic Marketing

Dr

About

64
Publications
9,215
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
1,001
Citations
Introduction
Skills and Expertise

Publications

Publications (64)
Chapter
Full-text available
Zwei Ereignisse im 20. Jahrhundert veränderten das Bild der Mathematik nachhaltig. Zuerst bewies Kurt Gödel, dass es kein endlich axiomatisierbares Kalkül geben kann, in dem alle wahren Aussagen über natürliche Zahlen beschreibbar sind. Einige Jahre später wies Alan Turing die Existenz von nicht berechenbaren Funktionen nach. Die ursprünglichen Aus...
Article
Full-text available
A method for accurate depth measurement in depth from focus is presented. With depth from focus the 3D shape of objects can be measured. To do so, an image series with different distances between camera and object is evaluated to extract focused parts. Combining the results of an edge detection filter with the distance information at which the imag...
Article
Full-text available
An approach for automatic road extraction from digital aerial imagery is presented. The extraction is based on a semantic model for roads. The images are divided into different so-called "global contexts": rural, forest, and urban. Different parts of the road model and different strategies are used in the different global contexts. In rural areas,...
Article
Full-text available
An approach for the automatic extraction of roads from digital aerial imagery is proposed. It is optimized for rural areas and makes use of versions of an aerial image with different resolutions. Roads are modeled as a network of intersections and links between these intersections, and are found by a grouping process. The context of roads is hierar...
Article
Full-text available
This paper presents a software architecture that is capable of easy extension and maintenance and enables the user to develop applications rapidly and in a flexible manner. This is achieved by using an object-oriented design, both for the data structures and the operators which process this data. The granularity of the operators is chosen such that...
Article
Full-text available
In this paper we report on research on road extraction at TUM (Technische Universitat Munchen). We propose a scheme for road extraction in rural areas that integrates three different modules with specific strengths. The first module employs local grouping and uses multiple scales and context to reliably extract most parts of the road network. In or...
Article
Full-text available
In this paper we present concepts for modelling image processing operators within an object-oriented system like the Smalltalk environment. The operators, which are described in a knowledge-base and receive their functionality from a large, proven, and well-established C library, are integrated into the Smalltalk environment in order to provide thi...
Article
Full-text available
This paper describes the state of the art in segmentation algorithms of aerial images. Different approaches and object classes are described and their advantages and limitations are shown. First the advantage of multiple input data (e.g., color, infrared, DEM) and the information that can be derived from these sources is discussed. Besides sensor d...
Article
Full-text available
The paper introduces the extraction of objects that are higher than their surroundings, like houses, trees, or bridges, by the segmentation of a DTM (digital terrain model). We propose a new algorithm (called dual rank) to extract the ground in the DTM, which can be seen as an extension of the gray opening. The extraction of high objects can simply...
Article
Full-text available
The paper introduces generalized operators for gray morphology. These operators are defined in a way that the dilation and erosion are special cases. Based on these, the dual rank operator is defined. This operator unifies the important operators for gray morphology (opening and closing) with the rank operator, especially the median. Different char...
Article
Full-text available
The main focus of this contribution is the description of mechanisms for an intuitive input of user-data with a direct response to gain fast parameter selection. To build polymorph input methods for different parameter classes, a database containing knowledge about each operator is needed. This information about the syntactical and semantical struc...
Article
Full-text available
The paper introduces the extraction of objects that are higher than their surroundings, like houses, trees, or bridges, by combining the results of a segmentation of a DTM and a texture analysis of the gray image. We propose a new algorithm (called dual rank) to extract the ground in the DTM, which can be seen as an extension of the gray opening. T...
Article
Full-text available
: In this paper a new multi resolution approach to automatic road extraction from aerial images is described. It is made use of the fact that different characteristics of objects such as roads can be best detected in different scales. Two different resolutions of the same image are used, a coarse one with 2 m per pixel, and a fine one with 0.25 m p...
Article
Full-text available
Parallelization of image analysis tasks forms a basic key for processing huge image data in realtime. At this, suitable subtasks for parallel processing have to be extracted and mapped to components of a distributed system. Basically, this task should be done by the processing system and not by the user, as automatical parallelization allows a flex...
Article
. An evaluation function is presented, which tests the results of an object boundary segmentation algorithm to choose an optimal approximation using lines and circular arcs. An angle detection procedure is used to find sharp corners of the object boundary. Because this procedure does not detect smooth changes, the boundary segments defined by the s...
Article
Full-text available
Parallel processing of image analysis tasks is an essential method to speed up image processing and helps to exploit the full capacity of distributed systems. However, writing parallel code is a difficult and time-consuming process and often leads to an architecture-dependent program that has to be re-implemented when changing the hardware. Therefo...
Article
Abstract. We propose a new approach for automatic road extraction from aerial imagery with a model and a strategy mainly based on the multi-scale detection of roads in combination with geometry-constrained edge extraction using snakes. A main advantage of our approach is, that it allows for the first time a bridging of shadows and partially occlude...
Article
Full-text available
An approach for the automatic extraction of roads from digital aerial imagery is proposed. It makes use of several versions of the same aerial image with different resolutions. Roads are modeled as a network of intersections and links between these intersections, and are found by a grouping process. The context of roads is hierarchically structured...
Article
Full-text available
A methodology for proving the accuracy of a system for high precision measurement of cutting tools is described. The tools are backlit by infrared illumination. Images are acquired by telecentric lenses and standard cameras and framegrabbers. The outer boundary of the tools is extracted by subpixel edge detection and is then approximated by lines a...
Article
Full-text available
Extracting roads from aerial imagery is important for the update of Geographic Information Systems (GIS), e.g., to capture new roads or to improve the geometric accuracy of existing roads. It can be broken down into two phases: First, old data is verified to determine which parts of the data have changed. Then, newly built roads are added. In this...
Article
Parallel processing of image analysis tasks is an essential method to speed up image processing and helps to exploit the full capacity of distributed systems. However, writing parallel code is a difficult and time-consuming process and often leads to an architecture-dependent program that has to be re-implemented when changing the hardware. Therefo...
Article
Full-text available
A methodology for proving the accuracy of a system for high precision measurement of cutting tools is described. The tools are backlit by infrared illumination. Images are acquired by telecentric lenses and standard cameras and framegrabbers. The outer boundary of the tools is extracted by subpixel edge detection and is then approximated by lines a...
Article
Full-text available
Parallelization of image analysis tasks forms a basic key for processing huge image data in real time. At this, suitable subtasks for parallel processing have to be extracted and mapped to components of a distributed system. Basically, this task should be done by the processing system and not by the user, as automatical parallelization allows a fle...
Article
Full-text available
Contextual information can facilitate automatic extraction of objects from digital imagery. This paper addresses the use of context for the automatic extraction of roads from aerial imagery. Context is restricted to knowledge about relations between roads and other objects and is hierarchically structured. More specific, context is used to guide ro...
Article
Full-text available
This paper presents a multi-resolution approach for automatic extraction of roads from digital aerial imagery. Roads are modeled as a network of intersections and links between the intersections. For different context regions, i.e., rural, forest, and urban areas, the model describes relations between background objects, e.g., buildings or trees, a...
Article
Full-text available
This paper presents a multi-resolution approach for automatic extraction of roads from digital aerial imagery. Roads are modeled as a network of intersections and links between the intersections. For different context regions, i.e., rural, forest, and urban areas, the model describes different relations between background objects, e.g., buildings o...
Article
Full-text available
An integrated program development environment for computer vision tasks is presented. The first component of the system is concerned with the visualization of 2D image data. This is done in an object-oriented manner. Programming of the visualization process is achieved by arranging the representations of iconic data in an interactively customizable...
Conference Paper
This article propose a way to automatically retrieve images from the world-wide-web using a semantic description for images and an agent concept for the retrieval of images. The system represents image in a textual way, e.g. look for a portrait of the a specific person, or fetch an image showing a countryside in Southern California. This textual de...
Conference Paper
Full-text available
In the field of object recognition, it is customary to evaluate the generated matching hypotheses with different methods. Several of these methods use a weight (constants, feature statistics, etc.) to produce an improved evaluation. These weights are calculated in the training phase of the model generation and applied later to recognize an object....
Article
Full-text available
Aerial imagery is an important source for the acquisition and update of GIS data. By using digital imagery it is possible to automate some parts of these tasks. In this context this paper proposes a new approach for the automatic update, i.e., verification and extraction, of roads from aerial imagery. The verification process evaluates road axes fr...
Article
Aerial imagery is an important source for the acquisition and update of GIS data. By using digital imagery it is possible to automate some parts of these tasks. In this context this paper proposes a new approach for the automatic update, i.e., verification and extraction, of roads from aerial imagery. The verification process evaluates road axes fr...
Article
Full-text available
The paper introduces the extraction of objects that are higher than their surroundings, like houses, trees, or bridges, by combining the results of a segmentation of a DTM and a texture analysis of the gray image. We propose a new algorithm (called dual rank) to extract the ground in the DTM, which can be seen as an extension of the gray opening. T...
Article
Full-text available
In this paper, an integrated program development environment for computer vision tasks is presented. The first component of the system is concerned with the visualization of 2D image data. This is done in an object oriented manner. Programming of the visualization process is achieved by arranging the representations of iconic data in an interactive...
Article
Full-text available
In this paper we present an approach to the automatic extraction of roads from aerial images. We argue that a model for road extraction is needed in every step of the image interpretation process. The model needs to include knowledge about different aspects of roads, like geometry, radiometry, topology, and context. The main part of this paper disc...
Conference Paper
Full-text available
An evaluation function is presented, which tests the results of an object boundary segmentation algorithm to choose an optimal approximation using lines and circular arcs. An angle detection procedure is used to find sharp corners of the object boundary. Because this procedure does not detect smooth changes, the boundary segments defined by the sha...
Conference Paper
Full-text available
The paper introduces the extraction of objects that are higher than their surroundings, like houses, trees, or bridges, by combining the results of a segmentation of a DTM and a texture analysis of the gray image. We propose a new algorithm (called dual rank) to extract the ground in the DTM, which can be seen as an extension of the gray opening. T...
Conference Paper
Die Erforschung der Morphologie von Eiskristallen in Cirrus-Bewölkung ist ein aktueller Schwerpunkt in der Erstellung globaler Khmamodelle. Zur Ermittlung der Häufigkeiten bestimmter Kristalle soll die Erkennung und Klassifikation der Formen automatisiert werden. In dieser Arbeit werden neben einem Überblick über die in Frage kommenden Bildaufnahme...
Conference Paper
Full-text available
erscheinen.F ur die L osung dieser Aufgabe haben sich der Kopf und die Schulternals sichere Erkennungsmerkmale herausgestellt. Auch der Oberk orper selbst und,eingeschr ankt, die Arme k onnen f ur die Erkennung herangezogen werden.Die Suche nach Personen erfolgt sowohl datenals auch modellgetrieben. Imersten Schritt wird aus der bisherigen Bildfolg...
Article
Full-text available
Werkzeug mit dem alle Stufen des Analyse-prozesses von der Vorverarbeitung bis zur Interpretation problemorientiert durchgeführt werden können. Basierend auf hierarchischen Datentypen (von der Bildmatrix zu relatio-nalen Strukturen) und funktionalen Operatoren ermöglichen verschiedene Module (Benut-zerschnittstelle, Operatordatenbasis, automatische...
Conference Paper
Full-text available
Zusammenfassung. Vorgestellt werden Architektur, Datenstrukturen und Hauptmodule des Bildanalysesystems HORUS. HORUS ist ein Werkzeug mit dem alle Stufen des Analyseprozesses von der Vorver-arbeitung bis zur Interpretation problemorientiert durchgef uhrt werden k onnen. Basierend auf hierarchischen Datentypen (von der Bildmatrix zu relationalen Str...
Conference Paper
Full-text available
Canny (1983) presented criteria for measuring the quality of edge detectors and derived an optimal FIR-filter for step edges by optimizing them. Four years later Deriche proposed an approach to edge detection based on Canny's design utilizing IIR-filters which can be implemented very efficiently and recursively. Using the Deriche-filter leads to a...
Conference Paper
Full-text available
To understand the meaning of an image or image sequence, to reduce theeffort in the design process and increase the reliability and the reusability of image understanding systems, a wide spectrum of AI techniques is applied. Solving an image understanding problem corresponds to specifying an image understanding system which implements the solution...
Conference Paper
Full-text available
Für die Bestimmung des optischen Flusses in Bildfolgen kennt man verschiedene Verfahren. Die robustesten davon sind die Korrelationsmethoden. In der Praxis zeigt sich allerdings, daß die Verschiebungsvektoren nicht an allen Stellen berechenbar sind. Mit einem mehrkanaligen Korrelationsansatz kann aus mehreren Merkmalen der optische Fluß nun genauer...
Conference Paper
Full-text available
Canny stellte 1983 Gütekriterien für Kantendetektoren vor, durch deren Optimierung er einen optimalen eindimensionalen FIR-Filter für Stufenkanten gewann. Vier Jahre später präsentierte Deriche ein Verfahren zur Kantendetektion, das unmittelbar auf Cannys Ansatz aufbaut, jedoch IIR-Filter verwendet, die sich sehr effizient rekursiv implementieren l...
Conference Paper
Full-text available
Die automatisierte Segmentation dreidimensionalerBilddatensätze, die durch Kernspin- oder Computer-Tomographen und durch Ultraschall-Geräte erzeugt werden, ist im Hinblick auf eine weitere klinische Auswertung ein zentrales Problem der medizinischen Bildverarbeitung. Diesel- Artikel stellt ein neues Verfahren vor, das nach Filterung der Bilder mit...
Conference Paper
Full-text available
Das netzwerkfähige HORUS ist ein System, daß eine gleichmäßige Auslastung von Rechnerresourcen in einem Netzwerk (LAN) realisiert. Dazu werden über einen Daemon beliebig viele Rechner unterschiedlichster Bauart zu einem Cluster zusammengefaßt, auf die die verteilte HORUS-Applikation zugreifen kann. Bei einer Berechnungsabfolge zur Bildverarbeitung...
Conference Paper
Full-text available
Das Ziel der Arbeit war die Entwicklung eines Prototyp-Systems zur automatischen, wissensbasierten Erkennung von Bondierungsdefekten bei der Mikrochipherstellung. Geeignete Bildverarbeitungsverfahren wurden entwickelt und ausgewählt, um aus Aufnahmen von Bondierungsstellen die bewertungsrelevanten Bildausschnitte zu segmentieren. Diese Bildinformat...
Conference Paper
Full-text available
Vorgestellt wird eine neue Sprache zur Bildanalyse und Bildinterpretation. TRIAS-1 vereint Prinzipien von logik- und regelbasierten Systemen mit Petrinetzen. Syntaktisch wird ein Programm durch Regeln und darauf aufbauenden Prozeduren beschrieben, mit denen prozedurales und deklaratives Wissen dargestellt werden kann Die Sprache stellt darüberhinau...
Conference Paper
Full-text available
HORUS ist die Verbindung von Bildverarbeitungoperationen mit einer relationalen Datenbank und einer standardisierten Ein/Ausgabe- Schnittstelle zu einem alle Ebenen der Bildverarbeitung umfassenden Programmiersystem. Dieses System kann als Erweiterung in übliche KI-Sprachen eingebunden werden. Damit wird eine einheitliche wissensbasierte Behandlung...
Conference Paper
Full-text available
Conference Paper
Full-text available
Die Arbeit betrifft eine programm- und gerätetechnische Lösung zur rechnerunterstützten Auswertung forstwirtschaftlicher Luftbilder, wobei eine praktische Nutzung angestrebt wird.
Conference Paper
Full-text available
Beschrieben wird die Programmiersprache PSIWAG (Picture Segmentation and Interpretation With Attributed Graphs), die als Werkzeug für verschiedene Ansätze der Bildanalyse entwickelt wurde. Gekennzeichnet ist sie durch einen regelorientierten Programmaufbau, wobei einzelne Regeln Prozeduren repräsentieren, über die die Datenbasis manipuliert wird. K...
Article
Thesis (doctoral)--Technische Universität München, 1992. Includes bibliographical references (p. 181-191) and index.