Wyniki wyszukiwania - BazTech

1

Altered postural stability in elderly women following a single session of head-mounted display virtual reality

Cieślik Błażej, Serweta Anna, Federico Sara, Szczepańska-Gieracha Joanna

Acta of Bioengineering and Biomechanics

|

2021

|

Vol. 23, no. 1

107--111

EN

Purpouse: The aim of this study was to evaluate the effect of a single session of head-mounted display virtual reality on postural stability in elderly women. Methods: Forty-seven female subjects underwent a 20-minute virtual reality session. The mean age of the subjects was 70.12 years. As an immersive source, we used a relaxing virtual reality game with a head-mounted display device. The postural stability test was conducted using a Nintendo Wii force plate. Participants completed a set of three 30-s trials in which they took a quiet bipedal eyes-open stance while standing on a hard surface: before the virtual reality session, immediately after the virtual reality session, and 2 minutes after the virtual reality session. Centre of pressure parameters were analysed in the sagittal and frontal planes. Results: Analysing the results obtained immediately after the virtual reality session, significant differences were observed in almost all examined parameters. In the sagittal plane, centre of pressure path velocity increased by 10% (p < 0.01) and path standard deviation by 15% (p < 0.05). In the frontal plane, centre of pressure path velocity increased by 14% (p < 0.01). After 2 minutes, all examined parameters showed no significant difference compared to before the virtual reality session. Conclusions: Immediately after the virtual reality session, there was an increase in almost all examined parameters. However, after 2 minutes, all examined parameters had returned to baseline. Therefore, to reduce fall risk after a virtual reality session, it is recommended that the subject spend at least 2 minutes in a sitting position.

2

Assessment and recognition of pre and co seismic electromagnetic signatures from magnetotelluric data: a case study from Koyna–Warna seismoactive region, India

Vijaya Kumar P. V., Rawat Vishal S., Patro Prasanta K., Gupta A. K., Babu Narendra

Acta Geophysica

|

2021

|

Vol. 69, no. 1

1--15

EN

Continuously monitored magnetotelluric (MT) time series data were used to identify the short-term earthquake co-seismic and pre-seismic electromagnetic phenomenon. The co-seismic behavior of the MT time series data recorded at 15 Hz sampling frequency is analyzed for the earthquake that occurred on November 24, 2007, of Mw =4.6. The wavelet analysis of the MT time series data shows signifcant enhancement at 3–6 Hz frequency band in the scalogram during the earthquake in comparison with pre- and post-time. The signifcant enhancement in the scalogram is related to the onset of the main shock of the earthquake. In this paper, we have also shown the precursory signatures of several earthquake magnitudes (Mw) ranging from 3.9 to 4.9 and the focal depth extending from 5 to 10 km mainly dominated by normal and strike-slip faulting. The spectral polarization ratio technique was implemented on these events to identify the precursory signatures. A few days before the earthquake, a signifcant anomaly was identifed for most of the earthquakes using this technique. This prominent anomaly is correlated with Dst index, which provides information about the ionosphere and magnetosphere responses in the presence of the solar wind and interplanetary magnetic feld. We inferred the unusual behavior prior to the earthquake is related to the precursory signature, but not related to the solar-terrestrial efect. The complex tectonic settings in the study region suggest that both electrokinetic and seismic dynamo mechanisms are the probable mechanisms playing an important role in generation of co- and pre-seismic electromagnetic signals.

3

Faults detection and classification on parallel transmission lines using modified Clarke’s transformation-ANN approach

Saini Makmur, Yunus Shiddiq, Sulthan Ahmed Rizal, Djalal Muh.Ruswandi, Mustafa Mohd. Wasir, Rahimuddin Rahimuddin, Kitta Ikhlas

Przegląd Elektrotechniczny

|

2020

|

R. 96, nr 4

23--27

EN

This paper introduces a comparative study for fault detection and classification on parallel transmission line using cascade forward and feed forward back propagation. Both calculations were based on discrete wavelet transform (DWT) and Clarke’s transformation. Daubechies4 mother wavelet (Db4) was applied to decompose coefficients of wavelet transforms coefficients (WTC) and wavelet energy coefficients (WEC) of high frequency signals. The coefficients were inputs for training of neural network back-propagation (BPNN). The results showed that the feed forward back propagation algorithm of Artificial Neural Network (ANN) models responded better than Cascade forward back propagation algorithm models, particularly in fault detection and classification on parallel transmission. The results showed that the proposed method for fault analysis was able to classify all the faults on the parallel transmission line rapidly and correctly.

PL

W pracy przedstawiono badanie porównawcze wykrywania i klasyfikacji uszkodzeń równoległej linii przesyłowej z wykorzystaniem propagacji kaskadowej do przodu i do tyłu. Oba obliczenia oparto na dyskretnej transformacie falkowej (DWT) i transformacji Clarke'a. Falkę macierzystą Daubechies4 (Db4) zastosowano do dekompozycji współczynników przekształceń falkowych (WTC) i współczynników energii falkowej (WEC) sygnałów wysokiej częstotliwości. Współczynniki stanowiły dane wejściowe do szkolenia propagacji wstecznej sieci neuronowej (BPNN). Wyniki pokazały, że algorytm propagacji wstecznego sprzężenia zwrotnego modeli sztucznej sieci neuronowej (ANN) zareagował lepiej niż modele algorytmu kaskadowego propagacji wstecznej, szczególnie w wykrywaniu błędów i klasyfikacji w transmisji równoległej. Wyniki pokazały, że zaproponowana metoda analizy uszkodzeń była w stanie szybko i poprawnie sklasyfikować wszystkie uszkodzenia na równoległej linii przesyłowej.

4

Complex morlet wavelet design with global parameter optimization for diagnosis of industrial manufacturing faults of tapered roller bearing in noisycondition

Deák Krisztián, Kocsis Imre

Diagnostyka

|

2019

|

Vol. 20, No. 2

77--86

EN

Detecting manufacturing defects of bearings are difficult because of their unique topography. To find adequate methods for diagnosis is important because they could be responsible for serious problems. Wavelet transform is an efficient tool for analyzing the transients in the vibration signal. In this article we are focusing on industrial grinding faults on the outer ring of tapered roller bearings. Nine different real-valued wavelets, Symlet-2, Symlet-5, Symlet-8, Daubechies (2, 6, 10, 14), Morlet and Meyer wavelets are compared to a designed complex Morlet wavelet according to the Energy-to-Shannon-Entropy ratio criteria to determine which is the most efficient for detecting the manufacturing fault. Parameters of the complex Morlet wavelet are adjustable, thus, it has more flexibility for feature extraction. Genetic algorithm is applied to optimize the center frequency and the bandwidth of the designed wavelet. A sophisticated filtering procedure through multi-resolution analysis is applied with autocorrelation enhancement and envelope detection. To determine the efficiency of the designed wavelet and compare to the other wavelets, a test-rig was constructed equipped with high-precision sensors and devices. The designed wavelet is found to be the most effective to detect the manufacturing fault. Therefore, it has the capacity for an industrial testing procedure.

5

Image denoising using new wavelet thresholding function

Dehda B., Melkemi K.

Journal of Applied Mathematics and Computational Mechanics

|

2017

|

Vol. 16, nr 2

55--65

EN

In this paper, we propose a new image denoising method based on wavelet thresholding. In this method, we introduce a new nonlinear thresholding function characterized by a shape parameter and basic properties. These characteristics make the new method able to achieve a compromise between both traditional thresholding techniques such as Hard and Soft thresholding. The experimental results show that our proposed method provides better performance compared to many classical thresholding methods in terms of the visual quality of the denoised image.

6

A mathematical model for file fragment diffusion and a neural predictor to manage priority queues over BitTorrent

Napoli C., Pappalardo G., Tramontana E.

International Journal of Applied Mathematics and Computer Science

|

2016

|

Vol. 26, no. 1

147--160

EN

BitTorrent splits the files that are shared on a P2P network into fragments and then spreads these by giving the highest priority to the rarest fragment. We propose a mathematical model that takes into account several factors such as the peer distance, communication delays, and file fragment availability in a future period also by using a neural network module designed to model the behaviour of the peers. The ensemble comprising the proposed mathematical model and a neural network provides a solution for choosing the file fragments that have to be spread first, in order to ensure their continuous availability, taking into account that some peers will disconnect.

7

Classification of abnormalities in mammograms by new asymmetric fractal features

Beheshti S. M. A., Noubari H. A., Fatemizadeh E., Khalili M.

Biocybernetics and Biomedical Engineering

|

2016

|

Vol. 36, no. 1

56--65

EN

In this paper we use fractal method for detection and diagnosis of abnormalities in mammograms. We have used 168 images that were carefully selected by a radiologist and their abnormalities were also confirmed by biopsy. These images included asymmetric lesions, architectural distortion, normal tissue and mass lesion where in case of mass lesion they included circumscribed benign, ill-defined and spiculated malignant masses. At first, by using wavelet transform and piecewise linear coefficient mapping, image enhancement were done. Secondly detection of lesions was done by fractal method as a ROI. Since in investigation of breast cancer, it is important that fibroglandular tissues in both breasts be symmetric and for each asymmetric density, evaluation for malignancy is necessary, we define new fractal features based on extracting asymmetric information from lesions. The fractal features were evaluated on 5 data sets using SVM classifier which enabled to achieve high accuracy in classification of mammograms and diagnostic results. We have also investigated the performance of image enhancement in classification of each data set which shows different effects of enhancement on different lesion types.

8

Diagnostics of direct current machine based on analysis of acoustic signals with the use of symlet wavelet transform and modified classifier based on words

Glowacz A.

Eksploatacja i Niezawodność

|

2014

|

Vol. 16, no. 4

554--558

EN

In the paper author proposed an original approach for detection and localization of faults occurring in Direct Current machine. A system for diagnosing DC machines was described. The system performed an analysis of the acoustic signals of DC machine. Researches were conducted for two states of Direct Current machines. The studies were conducted for the algorithms of data processing: Symlet wavelet transform and modified classifier based on words. A pattern creation process has been carried out for the 10 sound samples. An identification process has been carried out for the 40 sound samples. The described implementation of the system may be useful for protecting machines. Moreover, this approach will reduce the cost of maintenance and the number of damaged machines.

PL

W pracy autor zaproponował oryginalne podejście do wykrywania, lokalizacji usterek występujących w maszynie prądu stałego. Opisano implementację systemu do diagnostyki maszyn prądu stałego. System przeprowadzał analizę sygnałów akustycznych maszyny prądu stałego. Przeprowadzono badania dla dwóch stanów maszyny prądu stałego. Badania zostały przeprowadzone dla algorytmów przetwarzania danych: Transformacji falkowej Symlet i zmodyfikowanego klasyfikatora opartego na słowach. Proces tworzenia wzorca do rozpoznawania został przeprowadzony dla 10 próbek dźwięku. Proces identyfikacji został przeprowadzony dla 40 próbek dźwięku. Opisana implementacja systemu może być przydatna do ochrony maszyn. Ponadto podejście takie pozwoli zmniejszyć koszty utrzymania i liczbę uszkodzonych maszyn.

9

Assessment of muscle load and fatigue with the usage of frequency and time-frequency analysis of the EMG signal

Bartuzi P., Roman-Liu D.

Acta of Bioengineering and Biomechanics

|

2014

|

Vol. 16, no. 2

31--39

EN

The aim of the study was to determine the effect of the muscle load and fatigue on the values of the parameters calculated on the basis of the time, frequency (Fourier transform) and time-frequency (wavelet transform) analysis of the EMG signal, for low levels of load. Fifteen young men took part in the study. The EMG signal was registered from right side biceps brachii (BB) and trapezius (TR) muscles in static conditions, at load 10%, 20% and 30% MVC (maximal voluntary contraction). On the basis of the analysis there were selected parameters sensitive to force (RMS) and parameters sensitive to fatigue but simultaneously insensitive to force (MPF – mean power frequency determined on the basis of Fourier transform, CMPFdb5 – mean power frequency determined on the basis of the wavelet transform). The results indicate that CMPFdb5 can show similar (muscle BB) or greater (muscle TR) sensitivity to fatigue than MPF. It can suggest that, for low levels of load, the wavelet transform parameters can be more effective in assessing muscle fatigue than the parameters based on the Fourier transform. The obtained results can allow for a more precise analysis of muscle fatigue at low levels of load. Further analysis for a greater number of muscles activated at low levels of load, with the usage of the parameters tested is desirable.

10

Wavelet Decomposition of Signal and Feature Selection by LASSO for Pattern Recognition

Libal U.

Przegląd Elektrotechniczny

|

2013

|

R. 89, nr 4

89-91

EN

There is searched the balance between an increase of pattern recognition risk and a decrease of a model size. The experiments are performed for noisy signals, decomposed in wavelet bases. Wavelet representation of signals, i.e. representation by wavelet coefficients called signal features, constitutes the full model. The presented feature selection method is based on the Lasso algorithm (Least Absolute Shrinkage and Selection Operator). The aim of the experiment is to find an optimal model size and investigate the relations between the risk, the number of signal features and the noise level. A new criterion of feature selection is proposed that minimizes both the risk and the number of signal features. The experimental risk of classification is analysed for all possible reduced by Lasso models and for several values of noise levels.

PL

Poszukiwana jest równowaga pomi˛edzy wzrostem ryzyka rozpoznawania obrazów oraz zmniejszeniem rozmiaru modelu. Badania przeprowadzono dla zaszumionych sygnałów, zdekomponowanych w bazach falkowych. Falkowa reprezentacja sygnałów, czyli reprezentacja za pomoca˛współczynników falkowych zwanych cechami sygnału, stanowi pełny model. Przedstawiona metoda selekcji cech jest oparta o algorytm Lasso (Least Absolute Shrinkage and Selection Operator). Celem eksperymentu jest znalezienie optymalnego rozmiaru modelu oraz zbadanie zale˙znosci pomie˛dzy ryzykiem, liczba˛ cech sygnału oraz poziomem szumu. Zaproponowano nowe kryterium selekcji cech, które minimalizuje ryzyko oraz liczbe˛ cech sygnału. Eksperymentalne ryzyko błe˛dnej klasyfikacji jest badane dla wszystkich moz˙liwych zredukowanych za pomoca˛ Lasso modeli oraz kilku wartosci poziomu szumu.

11

Transformata falkowa wybranych sygnałów symulacyjnych

Józefczyk I., Małecki R.

Problemy Eksploatacji

|

2013

|

nr 1

27--34

PL

W artykule przedstawiono dyskretną transformcję falkową (DWT) dwóch sygnałów symulacyjnych z zastosowaniem falki dmey (dyskretny Meyer), której użyteczność potwierdziły wcześniejsze badania [5]. Przedmiotem badań były sygnały - wzorce oraz sygnały powstałe z wzorca poprzez dodanie tła losowego - zakłóceń. W celu wyeliminowania czynnika losowego sygnały zakłócone poddawano transformacji falkowej. Celem badań było poszukiwanie narzędzi statystycznych do określenia jakości transformacji falką dmey na różnych poziomach dekompozycji.

EN

In the article, there was a Discrete Wavelet Transform (DWT) of two simulation signals shown with the usage of waves which had been proven to be effective in previous research. The subjects of the article are signals-modules and the signals created from the modules by adding a randomised background-interference. In order to eliminate the random factor, DWT was used against disrupted signals. The aim of the research was the search for statistical tools to assess the quality of dmey wave transformation.

12

Ewolucja algorytmu inwersji spektralnej oparta na wydzieleniu dubletu i normalizacji mocy

Żuławiński K.

Nafta-Gaz

|

2013

|

R. 69, nr 12

903--908

PL

Artykuł przedstawia rozwinięcie inwersji spektralnej opartej na dopasowaniu adaptacyjnym. Zawiera skrócony opis metody i analizuje jej słabe strony. Na podstawie analizy sygnału zgodnej z kryteriami rozdzielczości sugeruje możliwości podniesienia jej wydajności oraz dokładności.

EN

The article presents the development of spectral inversion based on adaptive matching. A short description of the method’s disadvantages is presented. There are suggested methods of precision and efficiency improvement by signal resolvability analysis.

13

Continuous wavelet transform as effective tools for detecting motion artefacts in electrogastrographical signals

Mika B., Tkacz E., Kostka P. S.

Journal of Medical Informatics & Technologies

|

2011

|

Vol. 18

59--68

EN

The cutaneous recording of gastric myoelectrical activity of the stomach known as electrogastrography (EGG) seems to be the promising tool for the non-invasive assessment of gastric motility. Unfortunately the EGG recording is usually severely contaminated both by motion artefacts and endogenous biological noise source. In order to use EGG signals as reliable diagnostic tool it is necessity to look for the effective artefacts removal methods. In this paper Continuous Wavelet Transform (CWT) was applied for detection motion artefacts from the EGG data. The set of own mother wavelets extracted directly from EGG signal was created and applied for detecting motion artefacts from one channel EGG recording. The results was compared with the effects obtained by using standard mother wavelets. The proposed method based on CWT with own mother wavelet presents very good performance for detecting motion artefacts from the EGG data.

14

On the convergence of the wavelet-Galerkin method for nonlinear filtering

Nowak Ł. D., Pasławska-Południak M., Twardowska K.

International Journal of Applied Mathematics and Computer Science

|

2010

|

Vol. 20, no 1

93-108

EN

The aim of the paper is to examine the wavelet-Galerkin method for the solution of filtering equations. We use a wavelet biorthogonal basis with compact support for approximations of the solution. Then we compute the Zakai equation for our filtering problem and consider the implicit Euler scheme in time and the Galerkin scheme in space for the solution of the Zakai equation. We give theorems on convergence and its rate. The method is numerically much more efficient than the classical Galerkin method.

15

Multilevel converter in low power drive applications

Iwaszkiewicz J.

Prace Instytutu Elektrotechniki

|

2009

|

Z. 240

195-208

EN

The paper describes a method of generating high quality AC voltage waveforms in low power AC drives. Voltage waveforms are generated by use of multilevel inverters. This technique can be applied particularly in small devices, servomechanisms, robotics. An approach to the synthesis of AC voltage waveforms using analytical methods like Fourier and wavelets transform is presented. The analysis of the obtained waveforms is included to the paper as well as a proposal of adequate converter circuitry.

PL

W artykule opisano metodę generowania przebiegów napięcia przemiennego wysokiej jakości do układów napędowych małej mocy prądu przemiennego (AC). Przebiegi napięcia są generowane z wykorzystaniem wielopoziomowych falowników. Ta technika może być zastosowana szczególnie w małych urządzeniach, serwomechanizmach i robotyce. Przedstawiono metody syntezy przebiegów napięcia przemiennego z wykorzystaniem metod analitycznych, takich jak przekształcenie Fouriera i przekształcenie Falkowe. W artykule przedstawiono również analizę uzyskanych przebiegów oraz propozycje odpowiednich schematów przekształtników.

16

Transformata falkowa wybranych sygnałów symulacyjnych jako narzędzie eliminujące czynnik losowy

Józefczyk I.

Problemy Eksploatacji

|

2008

|

nr 1

101-112

PL

W artykule przedstawiono dyskretną transformcję falkową (DWT) dwóch sygnałów symulacyjnych z zastosowaniem falek, których użyteczność potwierdziły wcześniejsze badania. Przedmiotem badań były sygnały-wzorce oraz sygnały powstałe z wzorca poprzez dodanie tła losowego-zakłóceń. Te drugie poddawano transformacji falkowej, chcąc w ten sposób wyeliminować czynnik losowy. Celem badań było sprawdzenie, jak wielkość zakłóceń wpływa na odzyskanie wzorca.

EN

The paper presents the discrete wavelet transform (DWT) of a two simulated signals using wavelets whose usefulness was confirmed by earlier studies. The subject of the examination were signals which were held up as a models for other signals and the signals obtained from model signals by adding random background noise. The later were put under discrete wavelet transformation in order to eliminate the influence of the random factor. The aim of the examination was the verification of the magnitude of the noise influence restoring the original model signal.

17

Transformata falkowa w diagnostyce urządzeń technicznych

Józefczyk I.

Problemy Eksploatacji

|

2008

|

nr 4

81-90

PL

Artykuł jest poszukiwaniem odpowiedzi na pytanie: jakie warunki musi spełniać falka, aby sygnał poddany transformacji falkowej nadawał się do zdekodowania informacji diagnostycznej. Obróbce poddano dwa sygnały symulacyjne (stacjonarny i niestacjonarny) oraz sygnał pobrany z urządzenia mechanicznego. W oparciu o uzyskane rezultaty potwierdzona została hipoteza mówiąca, iż bardzo istotny jest dobór falki. Podkreślono także przydatność transformacji falkowej jako narzędzia do obróbki wstępnej, dla celów diagnostyki urządzeń technicznych.

EN

The article attempts to define the conditions that must be fulfilled by the wavelet. These conditions ensure that after the wavelet transformation, the signal will be able to decode diagnostics information. The transformation embraced two simulation signals (stationary and non-stationary) and a signal from a mechanical appliance. Based on the results, the thesis attaching a great importance to wavelet selection has been corroborated. Also, emphasis was put on wavelet transformation as a useful tool for initial transformation.

18

Porównanie jakości radiometrycznej zdjęć lotniczych wykonanych kamerą analogową i cyfrową

Pyka K.

Archiwum Fotogrametrii, Kartografii i Teledetekcji

|

2008

|

Vol. 18b

537--545

PL

Dla potrzeb badań zebrano materiał składający się z par zdjęć zarejestrowanych kamerą analogową (LMK1000) i cyfrową (DMC). Było to zdjęcia wyselekcjonowane z wykonanych uprzednio prac fotolotniczych, przy czym materiał dobierano tak, aby w miarę możliwości porównywać zdjęcia wykonane w podobnym okresie wegetacyjnym i posiadające zbliżoną rozdzielczość geometryczną. Na zdjęciach wybierano fragmenty o jednolitym użytkowaniu terenu, np. budynki, parkingi, pola, lasy. W ten sposób zgromadzono materiał badawczy liczący 25 par obrazów o rozmiarach 1024 x 1024 pikseli. Jako metodę badań jakości radiometrycznej obrazów wybrano analizę ich transformat falkowych. Na podstawie analizy równania zachowania wariancji względnej stwierdzono następujące prawidłowości: (1) w obrazach z kamery cyfrowej względna wariancja detali sukcesywnie rośnie wraz ze wzrostem poziomu dekompozycji, (2) w obrazach z kamery analogowej względna wariancja maleje pomiędzy 1. i 2. poziomem rozdzielczości a potem powoli rośnie lub jest stabilna. Sukcesywny wzrost wariancji detali, obserwowany dla obrazów z kamery cyfrowej, świadczy o bardzo niskim poziomie szumów przypadkowych. Z kolei niestabilne zmiany wariancji dla obrazów z kamery analogowej dowodzą wysokiej zawartości szumów.

EN

A set of aerial images taken by two cameras, analogue (LMK 1000) and digital photogrammetric (DMC), was used to compare radiometric quality of images obtained. Pairs consisting of analogue and digital images showing corresponding fragments of identical contents were selected for comparisons. As far as possible, fragments showing homogenous land use, e.g. buildings, parking lots, fields, forests, were selected, which made it possible to observe how the land use affected the wavelet transform. A total of 25 and 9 fragments of medium- and small-scale images, respectively were subjected to comparisons. The fragments selected were 1024 * 1024 pixels in size. The wavelet transformation was chosen as a method with which to compare radiometric image quality. Analysis of the equation for relative image variance preservation allowed to reveal the following patterns: (1) the relative variance of details in digital camera images was found to increase with the decomposition level, (2) the relative variance in analogue camera images was observed to decrease between the first and the second decomposition level. In all the cases examined, the digital camera produced better parameters of noise evaluation. The DMC images contained several times less random noise than those taken with the analogue camera. The study confirmed that it was possible to define the noise content indicators by analysing the wavelet detail coefficients.

19

Transformata falkowa wybranych sygnałów diagnostycznych

Józefczyk I.

Problemy Eksploatacji

|

2007

|

nr 1

167-177

PL

W artykule przedstawiono dyskretną transformcję falkową (DWT) kilku sygnałów symulacyjnych z zastosowaniem falek, których użyteczność potwierdziły wcześniejsze badania. Przedmiotem badań były sygnały o znanej strukturze częstotliwościowej - wzorce oraz sygnały powstałe z wzorca poprzez dodanie tła losowego - zakłóceń. Te drugie poddawano transformacji falkowej, chcąc w ten sposób wyeliminować czynnik losowy. Przedstawiono głównie wyniki tych falek, które najlepiej odzwierciedlają zaprogramowane informacje. Uzyskane rezultaty wskazują na konieczność dobrego doboru falki (najlepsze wyniki uzyskano dla falki dmey) oraz kryterium stopu.

EN

The paper presents the discrete wavelet transform (DWT) of a few simulated signals using wavelets whose usefulness was confirmed by earlier studies. The subjects of the examination were pure signals of known frequency and signals distorted with random noise. The latter were put under discrete wavelet transformation in order to eliminate the influence of random noise. Only those results which show programmed information in the best way were presented. The obtained results point at the necessity of good selection of wavelet and stop criterion. The best results were achieved using dmey wavelet.

20

Recognition of handwritten digits based on complex discrete wavelet transform

Maszewski M.

Image Processing & Communications

|

2007

|

Vol. 12, no 1

29-45

EN

In this article the basic concept of handwritten digits recognition system based on some different variants of Complex Discrete Wavelet Transform (CDWT) is presented. The achieved results and accuracy of practical implementations of the proposed system are discussed.