Plenary Lecture

Controlled Automatic Detection of Defects in Dependent Image Data

Professor Ansgar Steland
RWTH Aachen University
Germany
E-mail: steland@stochastik.rwth-aachen.de

Abstract: Various areas in engineering sciences produce huge high-dimensional data streams which need to be monitored to detect parts of the data or locations in the data stream which indicate faults or critical events. We consider the specific problem to analyze image data, in order to detect automatically suspicious areas. This problem arises, for instance, in photovoltaics when evaluating the quality of solar cells and photovoltaic modules by electroluminescence imaging. We consider an automatic detector which proceeds in two stages: At the first stage the image is locally smoothed to reduce noise. The second stage scans the images by local statistics to detect the most suspicious area. The local statistics are standardized by an estimate based on residuals calculated from a smooth estimate of the in-control background which is allowed to vary but on a larger scale. The procedure is designed to ensure that a positive detection occurs with a controlled error probability. This is achieved by applying new results on Gumbel-type extreme value asymptotics for random fields. Simulations for models of correlated image data show that, for properly chosen parameters of the local statistics, the approximation of the significance level is accurate for realistic image resolutions and moderate correlations. The method is illustrated by an application to electroluminescence images of photovoltaic modules.

Brief Biography of the Speaker: Ansgar Steland received the M.Sc. and Ph.D. degrees in mathematics from the University of Göttingen, Germany. He held positions as an assistant at the Technical University of Berlin, Berlin, Germany, as a consultant in industry, as a postdoc at the European University Viadrina of Frankfurt/Oder, Germany, and as a lecturer at the Faculty of Mathematics at the Ruhr-University Bochum. Since 2006 he has been a Professor at RWTH Aachen University, Germany, where he holds the Chair of Stochastics at the Institute of Statistics. He is an Elected Member of the International Statistical Institute (ISI) and acts as the chair of the Society for Reliability, Quality and Safety and as the chair of the Section on Statistics in Natural Sciences and Technology of the German Statistical Society. He organized and co-organized several international workshops and invited sessions. His current research interests are in nonparametric statistics, high-dimensional inference, signal and change-point detection, sequential analysis and quality control, empirical processes, time series analysis, econometrics and applications to photovoltaics.

Bulletin Board

Currently:

The Conference Guide is online.

The paper submission deadline has expired. Please choose a future conference to submit your paper.


Plenary Speakers

WSEAS Main Site

Publication Ethics and Malpractice Statement