Student Report Software Engineering FACH-0097

BibliographyWiselka, Matthias; Schieberle, Christian; Gerhardt, Andreas: Verfahren zur Geräuschpegelanalyse zur Situationserkennnung.
University of Stuttgart, Faculty of Computer Science, Electrical Engineering, and Information Technology, Student Report Software Engineering No. 97 (2008).
61 pages, german.
CR-SchemaI.2.10 (Vision and Scene Understanding)
I.4.8 (Image Processing and Computer Vision Scene Analysis)
I.5.4 (Pattern Recognition Applications)
I.6.4 (Model Validation and Analysis)
KeywordsNEXUS, Geräuschpegelanalyse, Künstliche Neuronale Netze, Logistische Regression, Multinomiale Regression, Situationserkennung
Abstract

Within the scope of the NEXUS project, a possibility was needed to determine if a room was occupied based on measured audio signals. The goal of this case study was to analyze and compare the quality of different methods that allow recognition of the ongoing situation. The analyzed methods include a noise level based approach, artificial neural networks and logistic regression. Also, a comparison was made for different numbers of recognizable situations and how well they could be recognized. This case study included the development of a program for measuring the audio signals of a room and another program for the evaluation of the measured audio data with the analyzed methods.

Full text and
other links
PDF (6310325 Bytes)
Access to students' publications restricted to the faculty due to current privacy regulations
ContactE-Mail: m.wiselka@gmail.com
Department(s)University of Stuttgart, Institute of Parallel and Distributed Systems, Image Understanding
Superviser(s)Käppeler, Uwe-Philipp
Project(s)Abteilung Bildverstehen
Entry dateSeptember 9, 2009
   Publ. Computer Science