Estimation and interpolation of underwater low frequency ambient noise spectrum using artificial neural networks

Ramji, Shyam; Latha, G.; Ramakrishnan, S.

doi:10.1016/j.apacoust.2009.01.006

Cited by 7 publications

(6 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Furthermore, the results acknowledge qualitative contribution of non-anthropogenic sources such as sea surface agitation and wave-breaking along the shorelines (Manasseh et al, 2006;Cho and Choi, 2010) on the background noise in the sea. The results therefore demonstrates a capacity of the joint neural network analysis to extract the connections between the variables of different types, already found in other oceanographic studies (Ramji et al, 2009;Mihanović et al, 2011;.…”

Section: Discussionmentioning

confidence: 74%

“…In our case, by measuring and forecasting the distribution of small recreational boats and fishing (TW_GN) vessels, one can try to associate them to a SOM solution, and then to use the corresponding SAN distribution as a forecast. This was already found to be feasible approach in SAN studies (Ramji et al, 2009), as well in various oceanographic studies (e.g., Mihanović et al, 2011). To achieve that, it is necessary to have a noteworthy amount of the SAN and vessel presence data, both in time and space, to be used in the training of the neural networks and mapping of SOM solutions.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Mapping underwater sound noise and assessing its sources by using a self-organizing maps method

Rako¹,

Vilibić

Mihanović

2013

The Journal of the Acoustical Society of America

View full text Add to dashboard Cite

This study aims to provide an objective mapping of the underwater noise and its sources over an Adriatic coastal marine habitat by applying the self-organizing maps (SOM) method. Systematic sampling of sea ambient noise (SAN) was carried out at ten predefined acoustic stations between 2007 and 2009. Analyses of noise levels were performed for 1/3 octave band standard centered frequencies in terms of instantaneous sound pressure levels averaged over 300 s to calculate the equivalent continuous sound pressure levels. Data on vessels' presence, type, and distance from the monitoring stations were also collected at each acoustic station during the acoustic sampling. Altogether 69 noise surveys were introduced to the SOM predefined 2 × 2 array. The overall results of the analysis distinguished two dominant underwater soundscapes, associating them mainly to the seasonal changes in the nautical tourism and fishing activities within the study area and to the wind and wave action. The analysis identified recreational vessels as the dominant anthropogenic source of underwater noise, particularly during the tourist season. The method demonstrated to be an efficient tool in predicting the SAN levels based on the vessel distribution, indicating also the possibility of its wider implication for marine conservation.

show abstract

Section: Discussionmentioning

confidence: 74%

Section: Discussionmentioning

confidence: 99%

Mapping underwater sound noise and assessing its sources by using a self-organizing maps method

Rako¹,

Vilibić

Mihanović

2013

The Journal of the Acoustical Society of America

View full text Add to dashboard Cite

show abstract

“…LPC [7] was applied to the scattered signal of each object and the coefficients are extracted for each type. LPC is implemented with the linear combination of past samples and can be written as [7] ] A multi layer feed forward neural network [8], [9] has been used for classification. A multilayer feed forward neural network is an interconnection of perceptrons in which data and calculations flow in a single direction, from the input data to the outputs.…”

Section: Fig 3: Feature Selection and Classification Schemementioning

confidence: 99%

Acoustic scattering of underwater targets

Malarkodi

Manamalli

Kavitha

et al. 2013

2013 Ocean Electronics (SYMPOL)

Self Cite

View full text Add to dashboard Cite

The objective of this paper is to provide feature extraction algorithm for underwater targets. The targets are homogeneous elastic bodies of finite dimensions. The targets considered are a brass sphere, a PVC sphere, a brass cylinder, a PVC cylinder, concrete block and MS cylinder of different dimensions. The incident acoustic signal used was a linear frequency modulated (LFM) signal of finite duration with the signal bandwidth of 40 kHz to 80 kHz. The scattered acoustic signal from the targets are recorded and processed for feature selection. The scattered signals were analysed using power spectrum analysis, Linear Predictive Coding and Auto Regressive (AR) modelling , and its statistical features are extracted for all the targets. The nature of the backscattered signal for the underwater targets is also explained. The extracted features are passed into the feed forward neural network (FFNN) classifier. FFNN was used to identify the targets of six classes, to check the validity of extracting the feature of the targets. The result of the neural network shows that this feature extraction algorithm could enhance the fractal features of the signals and reduce the number of dimensions of the feature space and prove that it can efficiently classify underwater targets. A comprehensive study was then carried out to compare the classification performance by using these data sets in terms of performance analysis like specificity and sensitivity.

show abstract

“…Neural networks have been widely used in chemometrics to replace traditional multivariate calibration methods based on models of multiple linear regression because they are able to efficiently map and extract nonlinear, noisy, or incomplete relationships from the study data (Ramji et al, 2009;Chakraborty and Sahu, 2014). Artificial neural networks (ANN) come from artificial intelligence, a branch of computing science that tries to understand and model the behavior of the human brain (Haykin, 1999).…”

Section: Introductionmentioning

confidence: 99%

Quantification of whey in fluid milk using confocal Raman microscopy and artificial neural network

Rocha

Paiva

Anjos

et al. 2015

Journal of Dairy Science

View full text Add to dashboard Cite

In this work, we assessed the use of confocal Raman microscopy and artificial neural network as a practical method to assess and quantify adulteration of fluid milk by addition of whey. Milk samples with added whey (from 0 to 100%) were prepared, simulating different levels of fraudulent adulteration. All analyses were carried out by direct inspection at the light microscope after depositing drops from each sample on a microscope slide and drying them at room temperature. No pre- or posttreatment (e.g., sample preparation or spectral correction) was required in the analyses. Quantitative determination of adulteration was performed through a feed-forward artificial neural network (ANN). Different ANN configurations were evaluated based on their coefficient of determination (R2) and root mean square error values, which were criteria for selecting the best predictor model. In the selected model, we observed that data from both training and validation subsets presented R2>99.99%, indicating that the combination of confocal Raman microscopy and ANN is a rapid, simple, and efficient method to quantify milk adulteration by whey. Because sample preparation and postprocessing of spectra were not required, the method has potential applications in health surveillance and food quality monitoring.

show abstract

Estimation and interpolation of underwater low frequency ambient noise spectrum using artificial neural networks

Cited by 7 publications

References 22 publications

Mapping underwater sound noise and assessing its sources by using a self-organizing maps method

Mapping underwater sound noise and assessing its sources by using a self-organizing maps method

Acoustic scattering of underwater targets

Quantification of whey in fluid milk using confocal Raman microscopy and artificial neural network

Contact Info

Product

Resources

About