Extended dataflow model for automated parallel execution of algorithms. - In: Proceedings, ISBN 978-1-936338-63-4, (2012), S. 148-152
The standard model of dataflow can express algorithms consisting of command sequences where each and every command is executed exactly once. This paper presents extensions to the model that allows control flow to be expressed within that model as well. Using these extensions algorithms containing branches, loops and even recursion can be composed as dataflow.
Modeling automated inspection plans for straight line detection in image scenes. - In: Proceedings, ISBN 978-1-936338-63-4, (2012), S. 142-147
This article explains an approach to further enhance the detection of geometric primitives namely straight lines, based on the spline filter contour decomposition method as shown in [1]. The aim is to split complex features into simple primitives for which further processing steps will lead to automated measurement results. Thus the inspection plan generated is reusable without employing additional CAD software tools. The feature recognition approach proposed in this paper incorporates an adaptive spline filter scale space method for smoothing contour point clouds at subpixel-scale precision with the aim to reduce the user influence on the setup of ROIs (regions of interest) in image scenes and therefore increasing usability of software interfaces.
State based application modeling for industrial image processing. - In: Proceedings, ISBN 978-1-936338-63-4, (2012), S. 138-141
This paper presents in detail the relevance and the potential of a novel approach for process modeling in creation of various measurement applications using industrial image processing and quality measurement.
Indentification of construction and demolition waste by using image processing in the visual and near-infrared spectrum and machine learning methods. - In: Tagungsbericht, (2012), S. 1026-1033
A new optical solution for the identification and separation of CDW materials was discussed in front of the given recycling situation and first results were demonstrated. The given problem, especially the chosen material classes were analysed by determination of their specific phenotypical characteristics in form of the feature parameters in VIS and NIR spectrum. In this approach a part of relevant building material classes were used. Only five classes were used out of the plurality of building classes.
Automatic petrographic inspection by using image processing and machine learning. - In: Proceedings of XX IMEKO World Congress, ISBN 978-89-950000-5-2, (2012), insges. 7 S.
In this paper we present a method for an automatic inspection of mineral aggregates. Certain components of aggregates can negatively affect the mechanical strength of produced concrete or asphalt as well as cause the destructive alkali silica reaction. The automatic recognition of such aggregates could be successfully implemented by means of image processing and machine learning algorithms. We achieved a total recognition rate of 88% for this very complex recognition problem. Therefore every step in the pattern recognition process has been successfully optimized. In conjunction with the object singularization we achieved a fully automatic testing instrument for petrographic inspection.
Two-viewing-planes-projection: the new way of measuring spectacle frames. - In: Proceedings of XX IMEKO World Congress, ISBN 978-89-950000-5-2, (2012), insges. 3 S.
Measuring edges on pivot-mounted objects during rotation. - In: Advanced mathematical and computational tools in metrology and testing IX, (2012), S. 358-365
Application of spline surface profile filters to subpixel contour decomposition problems. - In: Advanced mathematical and computational tools in metrology and testing IX, (2012), S. 334-341
Evaluation of a frame bending measuring system for large diameters in wind power industry. - In: 15th International Symposium Mechatronika, 2012, (2012), insges. 5 S.
http://ieeexplore.ieee.org/xpl/articleDetails.jsp?tp=&arnumber=6415031
Concepts of a scanning hardware platform for high-resolution image processing with lab-on-a chip analysis. - In: 15th International Symposium Mechatronika, 2012, (2012), insges. 4 S.
This paper describes the disadvantages of the quality assurance of meat according germ contaminations. Because of the latest regulations and the consumption behavior of the costumers there is a need for faster testing methods. One possible solution could be fluorescent markers which react with specific germs. The fluorescence effect could be acquired with an image processing system. The requirements and possible solutions of such image processing system are presented in this paper.
http://ieeexplore.ieee.org/xpl/articleDetails.jsp?tp=&arnumber=6415029