Research Group
Digital Signal and Image Processing
Institute of Chemical Technology, Prague, Department of Computing and Control Engineering
Technická 5, 166 28 Praha 6
Digital signal and image processing
Linear and Non-Linear Methods of Time Series Prediction
Neural Networks in Digital Signal Processing
Biomedical Image Processing and Classification
Environmental Image Analysis and Processing
Keywords
 
Digital signal and image processing  

(DSP) represents a very general discipline that joins completely different topics of scientific research using very similar mathematical tools and algorithms. In this way it corresponds to the vision of a very famous philosopher G. W. Leibnitz (1646-1716) who tried to find the joint basis of different scientific disciplines to minimize communication problems of specialists of different areas of science. The feature of DSP as a link between different scientific problems is emphasized also now in papers of Prof. B. Widraw devoted to artificial neural networks.

In the modern world digital signal processing forms a basis for different research areas including information engineering, measuring engineering, control engineering, communication technologies, technical cybernetics and biomedical engineering. The major research topics of the DSP group cover both theoretical methods including signal decomposition and reconstruction based on the Wavelet transform and applications in different areas.

10: Signal Modelling
11: Detection, Estimation and Identification
12: Signal Segmentation and Classification
13: Modelling, Approximation and Optimisation
14: Non-linear Signal Modelling
20: Signal and Image Analysis
21: Time-Frequency and Time-Scale Signal Analysis
22: Discrete Wavelet Transform
23: Statistical Methods
24: Adaptive Methods in Signal Processing
30: Signal Prediction
31: Statistical Methods for Signal Prediction
32: Neural Networks and Learning Theory
33: Evolutionary Computation
34: Digital Filters
40: Applications
41: Biomedical Signal Processing
42: Image Processing
43: Enviromental Signal and Image Processing
44: Engineering Signal Processing
 
Linear and Non-Linear Methods of Time Series Prediction

Project goal is in the analysis of autoregressive methods and neural networks for signal prediction, reliability limits evaluation and model simplification by principal components analysis and orthogonal decomposition. Mathematical methods are applied to gas consumption data processing.

 
Prediction
 
Neural Networks in Digital Signal Processing

Project includes methods of change-point detection, signal features extraction and signal classification with application to biomedical (EEG) signals. The mathematical background includes application of the Short time Fourier transform and Wavelet transform.

 
Neural networks
 
Biomedical Image Processing and Classification

Methods of image analysis and their linear and non-linear processing are used to perform signal de-noising, enhancement, detection of its components and their identification. The goal is in biomedical (NMR) image processing using Wavelet image decomposition and reconstruction, spline interpolation and gradient methods.

 
NMR
 
Environmental Image Analysis and Processing

Project goal is in image analysis, correlation and de-noising by two-dimensional Wavelet transform. Application is in the detection of dust particles by satellite observations and verification of suggested methods and algorithms by ground measuring stations.

 
Ovzdusi
 
Keywords

Digital Signal Processing - Image Processing - Time Series Analysis – Discrete Fourier Transform - Spectral Analysis - Wavelet Transform –Signal Decomposition and Reconstruction - Signal De-Noising - Signal Segmentation - Change-Point Detection - Feature Extraction - Signal Classification – Principal Component Analysis – Orthogonal Decomposition

Artificial Neural Networks - Wavelet Networks - Optimisation - Simulated Annealing - Genetic Algorithms – Linear and Non-Linear Modelling – Numerical Methods - Signal Prediction – Reliability Limits Estimation - Adaptive Systems - Artificial Intelligence

Information Engineering – Data Acquisition – Sensors – Computer Networks - Data Processing – Database Systems

Biomedical Signal Processing – Biomedical Image Analysis and Enhancement - Environmental Protection - Satellite Data Processing – Energy Consumption Prediction