PL
 |
EN
Preferencje help
Polish version English version Język
Widoczny [Schowaj] Abstrakt
Liczba wyników

Znaleziono wyników: 16

Liczba wyników na stronie
first rewind previous Strona / 1 next fast forward last
Wyniki wyszukiwania
Wyszukiwano:
w słowach kluczowych:  spectrogram
help Sortuj według:

help Ogranicz wyniki do:
first rewind previous Strona / 1 next fast forward last
1
Content available remote Acoustic-Based Drone Detection Using Neural Networks – A Comprehensive Analysis
Paszkowski Waldemar, Gola Arkadiusz, Świć Antoni
Advances in Science and Technology. Research Journal
|
2024
|
Vol. 18, no 1
36--47
EN
The article presents and describes the implementation of research on the detection of a drone in an urban environment using of the sound features. The methods of drone detection were recognized on the basis of modeling and evaluation of the features of the audio and acoustic signal. The authors proposed the use of a neural network model for the needs of drone detection taking into account acoustic measurements in an anechoic chamber and in an urban environment. The final part presents the obtained results of the drone detection. For the purposes of detection, a neural network model was used in order to recognize the obtained images of the spectograms of sound sources.
2
Content available Analiza porównawcza jakości zarejestrowanego dźwięku w funkcji różnych formatów zapisu
Król Andrzej, Szymczyk Tomasz
Journal of Computer Sciences Institute
|
2022
|
Vol. 24
189--194
PL
W artykule przeanalizowano jakość koderów: mp3, AAC, wma i OGG Vorbis. Do przeprowadzenia badania ilościo-wego wykorzystano autorską metodę graficzną. Polega ona na porównaniu liczby pikseli (reprezentujących dane) po-między spektrogramem pliku wav, a spektrogramami plików skompresowanych różnymi kodekami i przepływnościami. Wykazano, iż najwięcej danych z nieskompresowanej próbki wav zachowuje koder Ogg Vorbis we wszystkich bada-nych przepływnościach (128KBit/s, 160KBit/s, 320KBit/s).
EN
In article, the quality of the following encoders was analyzed: mp3, AAC, wma and OGG Vorbis. An original graphic method was used to carry out the quantitative research. It consists in comparing the number of pixels (representing data) between the spectrogram of a wav file and the spectrograms of files compressed with different codecs and bit rates. It has been shown that the Ogg Vorbis encoder retains the most data from the uncompressed wav sample in all tested bit rates (128KBit/s, 160KBit/s, 320KBit/s).
3
Analiza modeli splotowych w zadaniach klasyfikacji dźwieków środowiskowych
Piczak Karol J.
Przegląd Telekomunikacyjny + Wiadomości Telekomunikacyjne
|
2020
|
nr 7-8
366-369, CD
PL
Splotowe sieci neuronowe są obecnie popularnym narzędziem wykorzystywanym w rozpoznawaniu dźwięków środowiskowych. Na skuteczność ich działania wpływa wiele potencjalnych czynników. Niniejszy referat przedstawia podsumowanie wyników uzyskanych w rozprawie doktorskiej autora w zakresie analizy wrażliwości modeli splotowych na dobierane wartości hiperparametrów. W szczególności zastosowanie techniki dropout okazuje się mieć znaczący wpływ na funkcjonowanie tego typu modeli.
EN
Convolutional neural networks are a popular tool used in environmental sound recognition tasks. Their performance depends on multiple factors. This paper presents a summarized extract from author’s PhD dissertation on analyzing the sensitivity of convolutional models to hyperparameter values. In particular, dropout happens to play an important role in these kinds of models.
4
Content available Analysis of measurement conditions influence on the magnetic Barkhausen noise phenomenon
Maciusowicz Michał, Psuj Grzegorz
Welding Technology Review
|
2019
|
R. 91, nr 12
17--24
EN
The effectiveness of the magnetic Barkhausen noise method (MBN), used for non-destructive testing of ferromagnetic materials, depends to a large extent on a number of factors determining the measurement conditions. The use of conditions allowing the highest possible level of discrimination between the various states of the materials state is of highest importance. Therefore, this paper presents an analysis of the impact of measurement conditions on Barkhausen noise signals observed for various states of the material conditions. Taking into consideration the stochastic nature of MBN and the complex characterization of its changes, the analysis was based on the time-frequency representation of the MBN signal. The paper presents selected distributions achieved using two transformation methods. In addition, the extraction methods of features allowing the quantification of complex information were given. Finally, the discrimination ability for a number of parameters and features of MBN signals were determined and the obtained results were discussed.
PL
Skuteczność metody magnetycznego szumu Barkhausena MBN (ang. Magnetic Barkhausen Noise), wykorzystywanej do badań nieniszczących materiałów ferromagnetycznych, zależy w dużej mierze od szeregu czynników określających warunki pomiarowe. Kluczowe znaczenie ma zastosowanie warunków umożliwiających najwyższy możliwy poziom dyskryminacji między różnymi stanami badanych materiałów. W związku z tym w niniejszej pracy przedstawiono analizę wpływu warunków pomiaru na sygnały szumu Barkhausena rejestrowane dla różnych stanów badanego materiału. Mając na uwadze stochastyczną naturę szumu MBN i złożoną charakterystykę jego zmian, analizę przeprowadzono na podstawie reprezentacji czasowo-częstotliwościowej sygnału MBN. W pracy zaprezentowano wybrane rozkłady z zastosowaniem dwóch metod transformacji. Ponadto przybliżono metody ekstrakcji cech umożliwiające kwantyfikację złożonej informacji. Na koniec określono poziomy rozróżnialności dla szeregu parametrów i cech sygnałów MBN oraz omówiono uzyskane wyniki.
5
Content available Analysis and comparison of vibration signals from internal combustion engine acquired using piezoelectric and MEMS accelerometers
Bismor Dariusz
Vibrations in Physical Systems
|
2019
|
Vol. 30, nr 1
art. no. 2019112
EN
Condition monitoring of vehicles with internal combustion engine is of immense importance due to high number of vehicles with such engines and their importance to transport and economy. As many persons use a vehicle which is old and inexpensive, a condition monitoring system designed for such vehicles cannot be expensive. Unfortunately, condition monitoring of engines is usually based on the use of vibration signals, which are acquired by accelerometers. Piezoelectric accelerometers are the most commonly used for this purpose, and such accelerometers are not cheap. However, an alternative exists in the form of microelectromechanical systems (MEMS) accelerometers, which are much cheaper, but have narrower frequency characteristics. This paper describes preliminary results of a research on feasibility of use of MEMS accelerometers for condition monitoring and failure detection in internal combustion engines.
6
Content available Performance evaluation of deep neural networks applied to speech recognition : RNN, LSTM and GRU
Shewalkar Apeksha
Journal of Artificial Intelligence and Soft Computing Research
|
2019
|
Vol. 9, No. 4
235--245
EN
Deep Neural Networks (DNN) are nothing but neural networks with many hidden layers. DNNs are becoming popular in automatic speech recognition tasks which combines a good acoustic with a language model. Standard feedforward neural networks cannot handle speech data well since they do not have a way to feed information from a later layer back to an earlier layer. Thus, Recurrent Neural Networks (RNNs) have been introduced to take temporal dependencies into account. However, the shortcoming of RNNs is that long-term dependencies due to the vanishing/exploding gradient problem cannot be handled. Therefore, Long Short-Term Memory (LSTM) networks were introduced, which are a special case of RNNs, that takes long-term dependencies in a speech in addition to shortterm dependencies into account. Similarily, GRU (Gated Recurrent Unit) networks are an improvement of LSTM networks also taking long-term dependencies into consideration. Thus, in this paper, we evaluate RNN, LSTM, and GRU to compare their performances on a reduced TED-LIUM speech data set. The results show that LSTM achieves the best word error rates, however, the GRU optimization is faster while achieving word error rates close to LSTM.
7
Content available remote Use of incremental decomposition and spectrogram in vibroacoustic signal analysis in knee joint disease examination
Dołęgowski M., Szmajda M., Bączkowicz D.
Przegląd Elektrotechniczny
|
2018
|
R. 94, nr 7
162--166
EN
This paper presents two methods that enable indication of chosen dysfunction of knee joint: different stages of chondromalacia and osteoarthritis. The incremental decomposition of voltage in time and spectrogram were used. Both methods enable detection and identification of particular stages of knee joint dysfunction.
PL
W artykule zaprezentowano dwie metody umożliwiające wykrywanie wybranych dysfunkcji stawu kolanowego: różnych stadiów chondromalacji oraz choroby zwyrodniowej. Zastosowano metody badania rozkładów przyrostów przebiegów oraz spektrogram. Obydwie metody umożliwiają detekcję oraz identyfikację poszczególnych schorzeń.
8
Content available Advanced time-frequency representation in voice signal analysis
Mika Dariusz, Józwik Jerzy
Advances in Science and Technology. Research Journal
|
2018
|
Vol. 12, no 1
251--259
EN
The most commonly used time-frequency representation of the analysis in voice signal is spectrogram. This representation belongs in general to Cohen’s class, the class of time-frequency energy distributions. From the standpoint of properties of the resolution, spectrogram representation is not optimal. In Cohen class representations are known which have a better resolution properties. All of them are created by smoothing the Wigner-Ville’a distribution characterized by the best resolution, however, the biggest harmful interference. The used smoothing functions decide about a compromise between the properties of resolution and eliminating harmful interference term. Another class of time-frequency energy distributions is the affine class of distributions. From the point of view of readability of analysis of the best properties are known so called Redistribution of energy caused by the use of a general methodology referred to as reassignment to any time-frequency representation. Reassigned distributions efficiently combine a reduction of the interference terms provided by a well adapted smoothing kernel and an increased concentration of the signal components.
9
Content available Differentiation of random structure properties using wavelet analysis of backscattered ultrasound
Gambin B., Wojcik J., Doubrovina O.
Hydroacoustics
|
2016
|
Vol. 19
121--128
EN
The aim of this work was to find the differences between random media by analyzing the properties of the ultrasound signals backscattered from the inhomogeneities. A numerical model is used to generate two types of random media. The first has the randomness in scatterers’ positions and the second has the randomness in the size and acoustical properties of scatterers. The numerical model of wave scattering has been used to simulate the RF (radio frequency) signals caused by the incident pulse traveling as a plane wave. The markers of randomness type differences between the scattering media were obtained with the help of the spectral and wavelet analysis. The effect of differences in randomness type is more spectacular when the wavelet analysis is performed.
10
Content available Spectral methods in Polish emotional speech recognition
Powroźnik P., Czerwiński D.
Advances in Science and Technology. Research Journal
|
2016
|
Vol. 10, nr 32
73--81
EN
In this article the issue of emotion recognition based on Polish emotional speech signal analysis was presented. The Polish database of emotional speech, prepared and shared by the Medical Electronics Division of the Lodz University of Technology, has been used for research. Speech signal has been processed by Artificial Neural Networks (ANN). The inputs for ANN were information obtained from signal spectrogram. Researches were conducted for three different spectrogram divisions. The ANN consists of four layers but the number of neurons in each layer depends of spectrogram division. Conducted researches focused on six emotional states: a neutral state, sadness, joy, anger, fear and boredom. The averange effectiveness of emotions recognition was about 80%.
11
Content available Perception of Mixture of Musical Instruments with Spectral Overlap Removed
Kleczkowski P.
Archives of Acoustics
|
2012
|
Vol. 37, No. 3
355-363
EN
The issue of auditory segregation of simultaneous sound sources has been addressed in speech research but was given less attention in musical acoustics. In perception of concurrent speech, or speech with noise, the operation of time-frequency masking was often used as a research tool. In this work, an ex- tension of time-frequency masking, leading to the removal of spectro-temporal overlap between sound sources, was applied to musical instruments playing together. The perception of the original mixture was compared with the perception of the same mixture with all spectral overlap electronically removed. Ex- periments differed in the method of listening (headphones or a loudspeaker), sets of instruments mixed, and populations of participants. The main findings were: (i) in one of the experimental conditions the re- moval of spectro-temporal overlap was imperceptible, (ii) perception of the effect increased when removal of spectro-temporal overlap was performed in larger time-frequency regions rather than in small ones, (iii) perception of the effect decreased in loudspeaker listening. The results support both the multiple looks hypothesis and the "glimpsing" hypothesis known from speech perception.
12
Content available Measurement of instantaneous shaft speed by advanced vibration signal processing-application to wind turbine gearbox
Zimroz R., Urbanek J., Barszcz T., Bartelmus W., Millioz F., Martin N.
Metrology and Measurement Systems
|
2011
|
Vol. 18, nr 4
701-711
EN
Condition monitoring of machines working under non-stationary operations is one of the most challenging problems in maintenance. A wind turbine is an example of such class of machines. One of effective approaches may be to identify operating conditions and investigate their influence on used diagnostic features. Commonly used methods based on measurement of electric current, rotational speed, power and other process variables require additional equipment (sensors, acquisition cards) and software. It is proposed to use advanced signal processing techniques for instantaneous shaft speed recovery from a vibration signal. It may be used instead of extra channels or in parallel as signal verification.
13
Wykorzystanie metody mikroanalizy rentgenowskiej EDS do identyfikacji pierwiastków na przykładzie materiału łopatki turbiny silnika lotniczego
Lisiecki J.
Prace Naukowe Instytutu Technicznego Wojsk Lotniczych. Technika Lotnicza
|
2006
|
nr 20
57-67
PL
W artykule przedstawiono podstawowe zalety i problemy wynikające ze stosowania analizatorów EDS do identyfikacji pierwiastków w materiale badanych próbek. Omówiono różnice pomiędzy analizatorami EDS i WDS ze względu na zasadę działania i rozdzielczość. Przedstawiono przykład zastosowania analizatora EDS do mikroanalizy chemicznej materiału łopatki turbiny silnika lotniczego po eksploatacji.
EN
The most essential merits and drawbacks resulting from applying the EDS detectors to identity elements in the materiaIs of specimens under examination have been presented. Differences between the EDS and WDS detectors due to both the principle of operation and resolution have been discussed. An instance of applying the EDS detector to microchemical analysis of the material of an aircraft turbine blade withdrawn from service has been discussed.
14
Content available remote Analiza wyników obliczeń dynamicznych gruntu przy użyciu transformaty falkowej
Wrana B., Borowiec A.
Zeszyty Naukowe Politechniki Rzeszowskiej. Mechanika
|
2005
|
z. 65 [222]
375-382
EN
The paper deals with analysis of two phase soil layer results by mean of the wavelet transform. The layer is modeled as Biot's porous media with modern contribution and u - p simplification. Signal analysis conducted under Matlab proves that process is non-stationary. Classical signal processing tools such as Fourier transform and spectrogram are insufficient. The wavelet analysis was used instead. Frequencies spectrum calculated using Morlet's wave has similar values as those obtained from Fourier. Thanks to wavelet transform we gained the possibility of establishing the time when the dominant frequency (highest peak) was reached.
15
Obrazy widmowe testów lingwistycznych używanych w audiometrii słownej
Wykowska M., Bieda R.
Mechanika / Akademia Górniczo-Hutnicza im. St. Staszica
|
2003
|
T. 22, z. 4
559-578
PL
Podstawowym celem pracy było utworzenie spektrogramów wybranych testów lingwistycznych. Jako kryterium badań przyjęto stopień ich trudności, relacje pomiędzy elementami składowymi poszczególnych list testowych oraz zróżnicowanie głosów lektora ze względu na płeć. Listy wybrano losowo ze względu na obszerność materiału. Stosowane testy słowne aby mogły spełniać swoją rolę, powinny być zrównoważone pod wieloma względami. Za pomocą otrzymanych obrazów widmowych analizowano ww. aspekty. Został też podany przegląd istniejących testów. Przedstawiono także algorytm postępowania przy otrzymywaniu spektrogramów.
EN
The aim of the work conducted has been to create spectrograms of selected linguistic tests. Level of difficulty, relations between the elements consisting on testing lists and the range of lectures' sounds with respect to their gender have been chosen as the criteria for research. The lists were chosen randomly due to the scope of material. Verbal tests applied should be balanced in several aspects so that they fulfil their role. The above mentioned aspects were analysed using spectral pictures obtained. The overview of existing tests is presented as well. Finally, the procedure of obtaining spectrograms has been shown.
16
Content available remote Spectral Analysis of the Mass Irregularity of Slivers Using Uster Tester 3
Germanova-Krasteva D.
Fibres & Textiles in Eastern Europe
|
2003
|
Vol. 11, no. 3 (42)
25--27
EN
A method of calculation for determining the spectral characteristics of wavelengths and amplitudes of assessed sliver irregularities has been developed. Spectrograms from the Uster Tester 3 apparatus for irregularity measurement were utilised. The absolute and relative amplitude values were determined. A common formula for fast amplitude calculation is proposed.
PL
Opracowano metodę obliczeniową dla określania charakterystyki spektralnej długości fal i wielkości amplitud nierównomierności masy liniowej taśm. Dla celów metody wykorzystywane są spektrogramy pochodzące z aparatu lister Tester 3, służącego do pomiaru masy liniowej. Metoda pozwala na określanie bezwzględnych i względnych wartości amplitud. Zaproponowano postępowanie dla szybkiego obliczania wartości amplitud.
first rewind previous Strona / 1 next fast forward last
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.