Publications
2007

 
Haptic Rendering for Blind and Severely Visually Impaired Children
Frank Weichert, Mathias Wagner, Andreas Streng, Andreas Groh, Josef Ingenerf, Werner Liese, Tereza Richards, Ali Shamaa, Roland Linder

The prevalence of severely visually impairement to the point of blindness is in industrialised countries approx. 0,2 percent, i.e. in Germany approx. 160.000 people suffering under heavy visual deficits. 6 percent of the rele-vant people (approx. 8.700) are younger than 18 years. The last group is mostly in schooling to get a good education and prospects of success at the first employment market. So this leads to developing special methods and devices to make the hole media-catalog available and more than accessable for their students. In particular this affects the natural science subjects. In chemistry, physics and biology it is one point to teach the formulas and statistics, but it is another point to give the students the possibility to watch a live-experiment to feel and realize what the hole theory is about. The concept is to provide visual information by tactile analogies. A haptic device is used, i.e., a hardware device that helps present tactile information. The haptic device can generate a tactile representation of geometric objects and physical bodies.

MEDINFO 2007, Proceedings of the 12th World Congress on Health (Medical) Informatics, 20. – 24. August, Brisbane, Australia, pp. 342-343, 2007

 
A Topology-Independent Similarity Measure for High-Dimensional Feature Spaces
Jochen Kerdels and Gabriele Peters

In the field of computer vision feature matching in high dimensional feature spaces is a commonly used technique for object recognition. One major problem is to find an adequate similarity measure for the particular feature space, as there is usually only little knowledge about the structure of that space. As a possible solution to this problem this paper presents a method to obtain a similarity measure suitable for the task of feature matching without the need for structural information of the particular feature space. As the described similarity measure is based on the topology of the feature space and the topology is generated by a growing neural gas, no knowledge about the particular structure of the feature space is needed. In addition, the used neural gas quantizes the feature vectors and thus reduces the amount of data which has to be stored and retrieved for the purpose of object recognition.

International Conference on Artificial Neural Networks (ICANN 2007), Porto, Portugal, September 9-13, 2007

    HTTP-Download HTTP-Download
ICANN2007.pdf 		    Whole text, PDF, 350 K

 
Multivariate data analysis and visualization of position stationary and position-invariant calculated data of Hollow Cone Nozzles
Daniel Feggeler, Peter Walzel, Frank Weichert, Marcel Gaspar, Heinrich Müller

This paper deals with the analysis and visualization of multiphase flows during the process of sheet disintegration at Hollow Cone Nozzles (HCN). The characterization of surface waves and their propagation is based on the analysis of the sheet contour, considering the wavelength, the amplitude, the frequency and the growth rates of the waves. Our data from the CCD Camera are evaluated by an extended signal-theoretical approach. The investigations have shown wavelengths in the range between 0.00092 < x < 0.1 m. The highest growth rates are obtained for the 8mm orifice diameter at low Ohnesorge numbers Oh. The evaluations reveal that a dominating oscillation frequency occurs at each flow condition. The corresponding Strouhal number Stp exhibits in the investigated ranges a nearly constant wavelength of oscillations.

6th International Conference on Multiphase Flow, ICMF 2007, Leipzig, Germany, July 9 – 13, 2007

 
Image Segmentation Based on Height Maps
Gabriele Peters and Jochen Kerdels

In this paper we introduce a new method for image segmentation. It is based on a height map generated from the input image. The height map characterizes the image content in such a way that the application of the watershed concept provides a proper segmentation of the image. The height map enables the watershed method to provide better segmentation results on difficult images, e.g., images of natural objects, than without the intermediate height map generation. Markers used for the watershed concept are generated automatically from the input data holding the advantage of a more autonomous segmentation. In addition, we introduce a new edge detector which has some advantages over the Canny edge detector. We demonstrate our methods by means of a number of segmentation examples.

12th International Conference on Computer Analysis of Images and Patterns (CAIP 2007), Vienna, Austria, August 27-29, 2007

    HTTP-Download HTTP-Download
peters_kerdels_caip07.pdf 		    Whole text, PDF, 402 K

 
Aesthetic Primitives of Images for Visualization
Gabriele Peters

Images play an important role in visualization. As users are more willing to adopt a product if it evokes pleasurable feelings the aesthetic appeal of interfaces becomes more important. Thus, there is a growing need to generate also images which appear aesthetically to the user. Starting with the modularities of the human visual system, we derive six dimensions of visual aesthetics. For each dimension we explore, inspired by principles of the visual arts and insights of cognitive neuroscience, which pecularities of the dimensions are particularly adequate for an aesthetic impression. Accompanied by a fair number of image examples, these considerations result in an easy to understand guideline for computer scientists and interface designers how to deal with images in terms of aesthetics.

11th International Conference on Information Visualisation (IV 2007), Symposium on Design and Aesthetics in Visualisation (DAViz 2007), Zurich, Switzerland, July 2-6, 2007

    HTTP-Download HTTP-Download
peters-AestheticPrimitives.pdf 		    Whole text, PDF, 6.7 M

 
Multivariate data analysis and visualization of position stationary and position-invariant calculated data of Hollow Cone Nozzles
Frank Weichert, Marcel Gaspar, Heinrich Müller, Daniel Feggeler, Frank Landwehr, Peter Walzel

This paper introduces the analysis of multiphase flows during the process of sheet disintegration at Hollow Cone Nozzles (HCN). Despite the impact on different fields of application e.g. fuel injectors of motor vehicles or paint-spray lines, there is no deep understanding of the process of sheet disintegration. Taking into account multivariate measuring data, the focus of our work is the achievement of accurate geometrical parameters and signal-theoretic characterization of the liquid/gas interface for better understanding of the behavior of the sheet. By the given results it will be attempted to develop an improved physical-mathematical model of the sheet disintegration of Hollow Cone Nozzles. By incorporating the geometrical data it is possible to apply an extended signal-theoretical approach. Taking into account the assumed agglomerative wave spectrum, for differentiated examination a wavelet analysis is applied in addition to the global frequency analysis. Frequencies which vary by time and place, identification of relevant frequencies and calculating possible periodicity are relevant aspects in this context. If the last aspect takes place, a time series analysis could accomplish a delimitation to a chaotic system additionally. By projection of the calculated data onto a virtual 3D-spray cone a unified view is possible.

Statistics 2007 - Statistics under one Umbrella, Bielefeld, March 27-30, 2007

 
Dark Days - New York
Gabriele Peters

These photographs are a small excerpt of about 70 works which emerged from journeys to serveral cities in the winter 05/06, among them New York, Prague, and Venice. The whole cycle is called Dark Days - Venice, Prague, New York. Its topic is the isolation of the modern human beeing in todays big cities. The reality of the city inhabitants undergoes an estrangement by stylistic devices such as coarse grain, blur, camera shake, and multiple exposures. The selected images show scenes of crossroads in New York. They have been taken from a large distance at night. The humans shown in the photographs are not visible in the negatives with the naked eye. Only an extreme enlargement makes them visible. The technical process to generate these images is a hybrid one. The photographs have been taken on panchromatic film. The relevant sections of the negatives have been scanned with high resolution (4000 ppi). After applying the usual image enhancement techniques, artifical grain is added to the images to further exaggerate the effect of the film grain. Only after this the images are scaled up to their final size. It is important that the grain is added before the enlargement, because otherwise the pointilist effect, especially interesting from different viewing distances, is missed. Finally, a triplex filter with some yellowish and purple tones is applied to the black and white image for atmospheric reasons.

Online Exhibition and Physical Exhibition at Symposium on Digital Art (D-ART), 11th International Conference on Information Visualisation (IV 2007), Zurich, Switzerland, July 2-6, 2007
and
Online Exhibition at Symposium on Digital Art (D-ART), 4th International Conference Computer Graphics, Imaging and Visualization (CGIV 2007), Bangkok, Thailand, August 13-16, 2007

    HTTP-Download HTTP-Download
Dark_Days_expose.pdf 		    Whole text, PDF, 9.6 M
    HTTP-Download HTTP-Download
Dark_Days_artwork.pdf 		    Whole text, PDF, 6.6 M

 
Extension of the Generalized Image Rectification - Catching the Infinity Cases
Klaus Häming and Gabriele Peters

This paper addresses the topic of image rectification, a widely used technique in 3D-reconstruction and stereo vision. The most popular algorithm uses a projective transformation to map the epipoles of the images to infinity. This algorithm fails whenever an epipole lies inside an image. To overcome this drawback, a rectification scheme known as polar rectification can be used. This, however, fails whenever an epipole lies at infinity. For autonomous systems exploring their environment, it can happen that successive camera positions constitute cases where we have an image pair with one epipole at infinity and the other inside an image. So neither of the previous algorithms can be applied directly. We present an extension to the polar rectification scheme. This extension allows the rectification of image pairs whose epipoles lie even at such difficult positions. Additionally, we discuss the necessary computation of the orientation of the epipolar geometry in terms of the fundamental matrix directly, avoiding the computation of a line homography as in the original polar rectification process.

4th International Conference on Informatics in Control, Automation, and Robotics (ICINCO 2007), Angers, France, May 9-12, 2007

    HTTP-Download HTTP-Download
ICINCO07.pdf 		    Whole text, PDF, 129 k

 
Dynamic Learning Of Action Patterns For Object Acquisition
Gabriele Peters and Thomas Leopold

We propose an active vision system for the acquisition of internal object representations. The core of the approach is an agent which learns goal-directed action patterns depending on the perceived environment via reinforcement learning. The user supervision is restricted to the definition of this goal in form of a reward function. We demonstrate this approach by means of learning a strategy to scan an object. The agent moves a virtual camera around an object and is able to adapt her scan path dynamically to different conditions of the environment such as different objects and different goals of the data acquisition. The purpose of the acquisition we consider here is the view-based reconstruction of non-acquired views. The scan pattern obtained after the learned path has stabilized allows a better reconstruction of unfamiliar views than random scan paths.

International Journal of Intelligent Systems Technologies and Applications 2(2/3), pp. 113-124, 2007

UniDoLogo FBI-Logo LS7-Logo Back Top This page was last updated Sunday, 02-Sep-2007 12:45:46 CEST by wwwadmin@ls7.cs.tu-dortmund.de.
| Legal |