EMPA Media Technology

Publications 2011

DIN 6164 for Gamut Mapping?

U. Caluori, D. Küpper, K. Simon

Perceived color is an empirical phenomenon and, to date, is only approximately understood in complex situations. In general, color spaces or color order systems, as a mathematical characterization of such empirical observations, address specific applications such that they may not be adequate in other contexts. In this work, we investigate four device-independent color spaces (color order systems) with regard to their suitability for a specific gamut mapping concept called “unsharp mapping”.

Electronic Imaging 2011: Color Imaging XVI: Displaying, Processing, Hardcopy, and Applications.

bibtex PDF

Coordinates of particular physical samples in a colour space

Perception of Contrast on Different Backgrounds

M. Scheller Lichtenauer, I. Sprow, P. Zolliker

We quantify the effects of background on perception of contrast. For this, we compare pairs of colour differences using the method of constant stimuli. The influence of room illumination is quantitatively measured and discussed.

Midterm Meeting of the International Colour Association (AIC), Zurich

bibtex PDF
Visualization of background effects on perceived contrast

Subsurface imaging with NIR light using polarization gating

M. Sormaz, P. Jenny

In this paper, a numerical study with the near infrared (NIR) laser light and polarization gating is used to show possible contrast improvement of the spherical inclusion hidden in the turbid medium. Inclusion has the same refractive index and scattering properties as surrounding medium but slightly higher absorption.

PIERS Online Journal

Geometry of the simulation

Image fusion for optimizing Gamut Mapping

P. Zolliker, Z. Baranczuk, J. Giesen

In this paper we study a local, image dependent approach to the gamut mapping problem. A structural image quality measure is used to pick an optimal mapping algorithm for image patches from a given class of algorithms. The optimally mapped patches are then fused into a single mapping of the image. We discuss and compare two image fusion methods that are designed to avoid artifacts in the fused image. Psycho-visual experiments confirm that this approach has a good potential to obtain mapped images with higher perceived quality than any of the individual algorithms, on which the method is based.

CIC 2011: 19th Color imaging Conference, San Jose

bibtex PDF
An example image after segmentation, mapping and fusion

SpeechIndexer: A Flexible Software for Audio-Visual Language Learning

U. Glavitsch, K. Simon, J. Szakos

This paper presents SpeechIndexer as a software tool to create teaching and learning material for language courses and as a an e-learning program to train the oral comprehension and speech production. We introduce the player function where students can follow speech and text closely and the role play function that allows the learner to get involved in a dialog. Teachers can create material from the vast amount of speech recordings available (audio books, radio and TV podcasts, language learning CDs, etc.) that specifically match the knowledge level and interest of individual students or the whole class. The software may complement regular language courses or may serve to teach languages where teaching and learning material scarcely exist, e.g. endangered languages that have a pure oral tradition.

International Conference on Education, Informatics and Systemics (icEIC), Orlando, FL, USA, Nov. 29th - Dec. 2nd, 2011

Should you be interested to use the software, please contact U. Glavitsch.
SpeechIndexer main window with speakers marked for role play function

Deriving Fine-Scale Socioeconomic Information of Urban Areas using High-Resolution Satellite Imagery

F. Tapiador, S. Avelar, C. Tavares, R. Zah

This article presents a new approach to derive fine-scale socioeconomic information of urban areas using very high resolution satellite data. The rationale behind the method is to use high resolution satellite data, capable of resolving urban morphology details, to derive a classification of the image. Thus, it is assumed that there is a relationship between the socioeconomic profile and the urban morphology of an area in terms of availability of green areas, sport facilities, private swimming pools or pavement conditions. The method is tested using a case study of Lima, Peru. Using a sample of ground data, a neural network classifier was applied to a pre-classified image in which entropy had been used to mask extensive, non-built up areas that would otherwise have inserted spurious information into the classifier. The result shows a high correlation (0.70 R2) when compared with validation data. The good performances also show that a physiographic satellite view of the city reflects the socioeconomic layout of their inhabitants, thus making remote sensing a complementary tool for social research and urban planning. While the param-eterization of the problem may differ from one area to another, it is shown that an a priori choice of a few parameters may help to automatically characterize large areas in social terms, thus allowing social inequality and its evolution to be mapped in those areas with limited availability of data. In order to make the method widely applicable, the possibilities and limitations of applying the procedure to other large cities are discussed.

International Journal of Remote Sensing 32(21): 6437-6456

bibtex PDF
Panchromatic sector of Lima, Peru and its corresponding spatial entropy.