Tadej Vodopivec: Filtering methods for signal processing
Signal processing is a key component of pattern recognition; in order to obtain good result from it, filtering is needed in the preprocessing phase when dealing with any non-ideal measurements (i.e. noisy data). In this seminar Tadej Vodopivec will talk about effects of various filters (average, median, low pass, FFT, etc.) on the noisy signals. Main purpose of this presentation is to get a better understanding of how certain filters perform and which filters are appropriate in which situation.