IMPRESS - Intelligent Methods for Processing and Exploration of Signals and Systems

In the research project Intelligent Methods for Processing and Exploration of Signals and Systems (IMPRESS), neural computing and other intelligent and learning methods are applied in the preprocessing, coding, variable selection, detection, classification, analysis and modelling of signals, images, measurements and textual data originating from different processes. All concerned tasks are related to central data-analysis applications of the participating companies.

IMPRESS is a subproject of AISA.

The IMPRESS project is lead by
Prof. Erkki Oja
Helsinki University of Technology
P.O. Box 5400
02015 HUT
Finland
Tel. +358-9-451 3265
Fax +358-9-451 3277
Email: erkki.oja@hut.fi

• Adaptive real-time image analysis
• Data mining and visualization by self-organizing map
• Data-fusion and neural networks in complex models
• Adaptive signal processing
• Non-linear disturbance analysis
• Analysis of economical information by WEBSOM
• DAEMON - Data analysis and building representations via neural networks
• Estimation of time dependent faults by spatio-temporal neural network
• Optical characterization of microstructures