Wyniki wyszukiwania - BazTech

1

Hierarchical Classification of Environmental Noise Sources Considering the Acoustic Signature of Vehicle Pass-Bys

Valero X., Alias F.

Archives of Acoustics

|

2012

|

Vol. 37, No. 4

423-434

EN

This work is focused on the automatic recognition of environmental noise sources that affect humans’ health and quality of life, namely industrial, aircraft, railway and road traffic. However, the recognition of the latter, which have the largest influence on citizens’ daily lives, is still an open issue. Therefore, although considering all the aforementioned noise sources, this paper especially focuses on improving the recognition of road noise events by taking advantage of the perceived noise differences along the road vehicle pass-by (which may be divided into different phases: approaching, passing and receding). To that effect, a hierarchical classification scheme that considers these phases independently has been implemented. The proposed classification scheme yields an averaged classification accuracy of 92.5%, which is, in absolute terms, 3% higher than the baseline (a traditional flat classification scheme without hierarchical structure). In particular, it outperforms the baseline in the classification of light and heavy vehicles, yielding a classification accuracy 7% and 4% higher, respectively. Finally, listening tests are performed to compare the system performance with human recognition ability. The results reveal that, although an expert human listener can achieve higher recognition accuracy than the proposed system, the latter outperforms the non-trained listener in 10% in average.

2

Genetyczna klasyfikacja dźwięków i struktura warstwy dźwiękowej w subpolarnym krajobrazie Spitsbergenu

Rodzik Jan

Prace Komisji Krajobrazu Kulturowego

|

2008

|

nr 11

74--85

PL

Warstwa dźwiękowa stanowi uzupełnienie krajobrazu wizualnego i jest nowym przedmiotem badań, głownie w kontekście komfortu życia człowieka (Bernat, 2008). W środowisku nie zasiedlonym trwale przez człowieka, m.in. w obszarach polarnych, badania dźwięku są na etapie pionierskim (Quin, 2002/2003). Naukowych badań akustycznych nie prowadzono także na jednym z najlepiej poznanych lądow arktycznych, jakim jest Spitsbergen – od kilkudziesięciu lat teren wszechstronnych badań ekspedycji z rożnych krajów świata, w tym licznych wypraw polskich. Różnorodne, niekiedy bardzo intensywne, dźwięki towarzyszą tu dynamicznym procesom przyrodniczym, toteż strona akustyczna pełni ważną rolę w popularnych relacjach z wypraw badawczych, zwłaszcza wypraw pionierskich (Czeppe, 1958; Jahn, 1958; Rożycki, 1959; Birkenmajer, 1975). Naturalny krajobraz Spitsbergenu jest wyjątkowo zróżnicowany i mobilny (Czeppe, 1966; King 1994; Ziaja, 1994). Jest on przez to atrakcyjny także dla eksploracji turystycznych, sportowych oraz rekreacyjnych, ograniczanych przez administrację norweską poprzez wprowadzanie rygorystycznie przestrzeganych przepisów, dotyczących terytorialnych form ochrony przyrody (Mehlum, 1989; Schramm, 1994). W pobliżu stałych osiedli i stacji badawczych daleko posunęła się jednak dewastacja środowiska naturalnego (Krzyszowska, 1981), a związane z funkcjonowaniem tych obiektów dźwięki są trwałym, o wielokilometrowym nawet zasięgu, składnikiem krajobrazu. Istnienie stacji badawczych skłania do prowadzenia monitoringowych badań akustycznych. Podstawą do podjęcia badań może być genetyczna klasyfikacja dźwięków, sporządzona na podstawie ich znajomości, nabytej podczas kilku wy praw naukowych na Spitsbergen: trzech całorocznych ekspedycji Instytutu Geofizyki Polskiej Akademii Nauk, zimujących w Polskiej Stacji Polarnej w Hornsundzie (1982/83, 1992/93 i 1994/95) oraz trzech pierwszych, sezonowych wypraw letnich (1986-88) Instytutu Nauk o Ziemi Uniwersytetu Marii Curie-Skłodowskiej w Lublinie do Bellsundu. Mimowolnej „rejestracji” dźwięków dokonywano podczas obserwacji meteorologicznych, badań geomorfologicznych oraz udziału w rożnych pracach badawczych i logistycznych na lądzie, lodowcach i na morzu. Wykorzystano także opisy dźwięków zawarte w relacjach z innych wypraw, zwłaszcza z geologicznych eksploracji wnętrza Spitsbergenu (Rożycki, 1959; Birkenmajer, 1975).

EN

Spitsbergen is a mountain island located in significantly ice-covered Norwegian archipelago Svalbard. In spite of polar location, its climate is not very frosty but subpolar with ocean features. To a considerable degree, Icelandic Low and “warm” West Spitsbergen Current shape here thermal and rain conditions. Thanks to that, Greenland Sea nearby Spitsbergen shore doesn’t freeze whereas drift ice often appears here (Fig. 1). Dynamics of atmosphere, litosphere, cryosphere, hydrosphere, biosphere and human activity finds reflection in rich sound layer of landscape. On a basis of natural sounds’ origin, groups of them were distinguished: atmospheric, niveogenic, glacigenic, hydrogenic, zoogenic, anthropogenic and technical (Table 1). Sounds of resembling tone may be produced by variable components of geographical environment and on the other hand – variable sounds may derive from one source. Their diversity and intensity indicate not only on type and dynamics of home processes but also on a source of energy. In the Spitsbergen landscape, sounds are an effect of influence of: gravity (falling and flow), wind (causing a direct collision of centers and objects or through waving), living organisms (movement or activity of vocal system) and sudden blowing of carbohydrates and explosive agents (Fig. 2). Structure of Spitsbergen sound layer is characterized by distinct annual rhythm. Short summer period, connected with thawing of snow and ice cover and water flow, calving of glaciers, especially with functioning of numerous bird’s colony, stands out with sounds resource. In this part of Spitsbergen landscape, six zones in two parallel, arrange in tiers can be distinguished. Periglacial system build following zones: shore, tundra (plains) and subnival, including unfrozen mountain slopes. In a glacial system, these are zones: paraglacial (in front of glaciers), glacial (glaciers) and nival, including nunataks and frozen mountains’ slopes. “Openness” of this terrain causes that some sounds have over zonal extent but generally each zone has different sound layer (Table 2).

3

Sound Isolation by Harmonic Peak Partition for Music Instrument Recognition

Zhang X., Raś Z.W.

Fundamenta Informaticae

|

2007

|

Vol. 78, nr 4

613-628

EN

Identification of music instruments in polyphonic sounds is difficult and challenging, especially where heterogeneous harmonic partials are overlapping with each other. This has stimulated the research on sound separation for content-based automatic music information retrieval. Numerous successful approaches on musical data feature extraction and selection have been proposed for instrument recognition in monophonic sounds. Unfortunately, none of those algorithms can be successfully applied to polyphonic sounds. Based on recent successful in sound classification of monophonic sounds and studies in speech recognition, Moving Picture Experts Group (MPEG) standardized a set of features of the digital audio content data for the purpose of interpretation of the information meaning. Most of them are in a form of large matrix or vector of large size, which are not suitable for traditional data mining algorithms; while other features in smaller size are not sufficient for instrument recognition in polyphonic sounds. Therefore, these acoustical features themselves alone cannot be successfully applied to classification of polyphonic sounds. However, these features contain critical information, which implies music instruments' signatures. We have proposed a novel music information retrieval system with MPEG-7-based descriptors and we built classifiers which can retrieve the important time-frequency timbre information and isolate sound sources in polyphonic musical objects, where two instruments are playing at the same time, by energy clustering between heterogeneous harmonic peaks.

4

Musical Sound Classification based on Wavelet Analysis

Wieczorkowska A.

Fundamenta Informaticae

|

2001

|

Vol. 47, nr 1,2

175-188

EN

Contents-based searching through audio data is basically restricted to metadata, which are attached manually to the file. Otherwise, users have to look for the specific musical information alone. Nevertheles, when classifiers based on descriptors extracted from sounds analytically are used, automatic classification can be in some cases possible. For instance, wavelet analysis can be used as a basis for automatic classification of audio data. In this paper, classification of musical instrument sounds based on wavelet parameterization is described. Decision trees and rough set based algorithms are used as classification tools. The parameterization is very simple, but the efficiency of classification proves that automatic classification of these sounds is possible.