Pattern recognition applied to seismic signals of the Llaima volcano (Chile): An analysis of the events' features

Curilem, Millaray; Vergara, Jorge R.; Martín, César San; Fuentealba, Gustavo; Cardona, Carlos; Huenupán, Fernando; Chacón, Máx; Khan, Muhammad Salman; Hussein, Walid; Yoma, Néstor Becerra

doi:10.1016/j.jvolgeores.2014.06.004

Cited by 39 publications

(13 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…LP is in the range of [1][2][3][4][5] Hz, VT is in [2][3][4][5][6][7][8][9][10][11][12][13][14][15] Hz and TR is in [1][2][3][4][5]Hz. In addition to this, additive noise in sensors have a broad frequency range.…”

Section: Resultsmentioning

confidence: 99%

“…The number of neurons on the input layer is 2000 while in the hidden layer the number of neurons was varied from 50 to 200 in step of 50. The data used corresponds to 268 events, 200 were used for training and 76 for validation, which is considerably less amount of training data compared with the size of the training set used in previous papers [8,10,42]. Each event was coded into a binary representation, namely, the DNNs has 3 outputs, each one representing a different event (LP=[1 0 0], TR=[0 1 0] and VT=[0 0 1)].…”

Section: Resultsmentioning

confidence: 99%

“…These seismic sensors are equipped with seismometers with a bandwidth of 50 Hz in X, Y and Z directions. The data recorded about the seismic events is sent in real time to the OVDAS [8].…”

Section: Introductionmentioning

confidence: 99%

“…Events caused by the fracture of brittle material due to changes produced by the flow of magma or fluids are called Volcano-Tectonic (VT). Oscil- sensors [8] lating changes in the pressure produced by the dynamics of the fluids inside the ducts or small fissures are denominated Long Period (LP) and TRemor (TR) events.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Evaluation of Llaima volcano activities for localization and classification of LP, VT and TR events

Firoozabadi

Seguel

Soto

et al. 2017

Journal of Electrical Engineering

Self Cite

View full text Add to dashboard Cite

Evaluation of seismic signals is one of the most important research topics on Volcanology. Volcanoes have daily activity; therefore, high speed evaluation of recorded signals is a challenge for improving the study of the natural phenomena occurring inside these natural formations. The aim of this paper is the evaluation (denoising, localization and classification) and analysis of Llaima volcano activities, one of the most actives volcanoes in South America. Different already proposed methods, such as, Butterworth, Spectral Subtraction (SS) and Wiener Filter (WF) are compared to the proposed Modified Spectral Subtraction (MSS) and Modified Wiener Filter (MWF) to find the best method for denoising the volcano signals. Then, event localization based on received signals of volcano is performed. In this step, Time Delay Estimation (TDE)-based method is used on data acquired from 3 mechanical sensors located in the volcano area. The proposed method is used to estimate the area for event location. The proposed denoising methods make the starting point for the event more evident to increase the localization accuracy for events where the starting point is difficult to find. In the last step, a method based on the novel DNN technique is proposed to classify the three main events occurring in the Llaima volcano (TR (Tremor), LP (Long Period) and VT (Volcano Tectonic)).

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Resultsmentioning

confidence: 99%

“…These seismic sensors are equipped with seismometers with a bandwidth of 50 Hz in X, Y and Z directions. The data recorded about the seismic events is sent in real time to the OVDAS [8].…”

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Evaluation of Llaima volcano activities for localization and classification of LP, VT and TR events

Firoozabadi

Seguel

Soto

et al. 2017

Journal of Electrical Engineering

Self Cite

View full text Add to dashboard Cite

show abstract

“…Considering that the sampling frequency is 100Hz, the [0-50 Hz] bandwidth was divided in 16 uniform bandpass filters. The features considered are: five linear prediction filter coefficients (LPC) (Esposito et al, 2013); twenty Cepstral coefficients (Beyreuther et al, 2012) estimated with the 16 uniform bandpass filters mentioned above; and the energy in the 5 th wavelet band (1.56-3.125 Hz) (Curilem et al, 2014) obtained with the wavelet transform using a Daubechies mother type five and five decomposition levels. This feature was computed as the ratio between the sum of the components of the 5th wavelet band over the sum of all the wavelet components (in all the bands).…”

Section: Feature Extractionmentioning

confidence: 99%