Segmentation and characterization of acoustic event spectrograms using singular value decomposition

Mulimani, Manjunath; Koolagudi, Shashidhar G.

doi:10.1016/j.eswa.2018.12.004

Cited by 29 publications

(8 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Geiger & Helwani (2015) used Gabor filterbank features and Gaussian Mixture Models for event detection. Mulimani & Koolagudi (2019) used a singular value decomposition method for extracting acoustic event specific features from spectrogram. Theses features are used as inputs to a Support Vector Machine (SVM) classifier.…”

Section: Introductionmentioning

confidence: 99%

End-to-end environmental sound classification using a 1D convolutional neural network

Abdoli

Cardinal

Koerich

2019

Expert Systems with Applications

256

128

View full text Add to dashboard Cite

In this paper, we present an end-to-end approach for environmental sound classification based on a 1D Convolution Neural Network (CNN) that learns a representation directly from the audio signal. Several convolutional layers are used to capture the signal's fine time structure and learn diverse filters that are relevant to the classification task. The proposed approach can deal with audio signals of any length as it splits the signal into overlapped frames using a sliding window. Different architectures considering several input sizes are evaluated, including the initialization of the first convolutional layer with a Gammatone filterbank that models the human auditory filter response in the cochlea. The performance of the proposed end-to-end approach in classifying environmental sounds was assessed on the UrbanSound8k dataset and the experimental results have shown that it achieves 89% of mean accuracy. Therefore, the propose approach outperforms most of the state-of-the-art approaches that use handcrafted features or 2D representations as input. Furthermore, the proposed approach has a small number of parameters compared to other architectures found in the literature, which reduces the amount of data required for training.

show abstract

Section: Introductionmentioning

confidence: 99%

End-to-end environmental sound classification using a 1D convolutional neural network

Abdoli

Cardinal

Koerich

2019

Expert Systems with Applications

256

128

View full text Add to dashboard Cite

show abstract

“…There are different types of feature selection algorithm including principal component analysis, singular value decomposition, non-negative matrix factorization, latent semantic analysis, and locality preserving projections. In this regard, singular value decomposition is preferred over other feature extraction algorithms because it proved its efficiency in dealing with a wide range of engineering applications including forecasting weekly solar radiation ( 59 ), streamflow forecasting ( 60 ), and acoustic event classification ( 61 ). Additionally, it is characterized by its low computational complexity ( 62 , 63 ).…”

Section: Proposed Methodsmentioning

confidence: 99%

Hybrid Elman Neural Network and an Invasive Weed Optimization Method for Bridge Defect Recognition

Abdelkader

Moselhi

Marzouk

et al. 2020

Transportation Research Record: Journal of the Transportation R

View full text Add to dashboard Cite

Existing bridges are aging and deteriorating, raising concerns for public safety and the preservation of these valuable assets. Furthermore, the transportation networks that manage many bridges face budgetary constraints. This state of affairs necessitates the development of a computer vision-based method to alleviate shortcomings in visual inspection-based methods. In this context, the present study proposes a three-tier method for the automated detection and recognition of bridge defects. In the first tier, singular value decomposition ([Formula: see text]) is adopted to formulate the feature vector set through mapping the most dominant spatial domain features in images. The second tier encompasses a hybridization of the Elman neural network ([Formula: see text]) and the invasive weed optimization (I[Formula: see text]) algorithm to enhance the prediction performance of the ENN. This is accomplished by designing a variable optimization mechanism that aims at searching for the optimum exploration–exploitation trade-off in the neural network. The third tier involves validation through comparisons against a set of conventional machine-learning and deep-learning models capitalizing on performance prediction and statistical significance tests. A computerized platform was programmed in C#.net to facilitate implementation by the users. It was found that the method developed outperformed other prediction models achieving overall accuracy, F-measure, Kappa coefficient, balanced accuracy, Matthews’s correlation coefficient, and area under curve of 0.955, 0.955, 0.914, 0.965, 0.937, and 0.904, respectively as per cross validation. It is expected that the method developed can improve the decision-making process in bridge management systems.

show abstract

“…ZCF is the number of times the singular values in the first column of the matrix U , crosses zero. First column of U corresponds to the frequency information corresponding to the MSV [10]. It indicates the variation in the frequency component of the signal.…”

Section: Singular Value Decompositionmentioning

confidence: 99%

“…Increase in RMS and MAV, decrease in MNF and MDF during fatigue has been reported [6]. In recent researches, singular value features have been used in various non‐stationary signals such as vibration signal, EEG signals, acoustic signals to decompose TF matrix to one dimensional matrix for feature extraction [10–12].…”

Section: Introductionmentioning

confidence: 99%

Muscle fatigue analysis in biceps brachii surface electromyography signals using synchrosqueezed Morlet wavelet and singular value decomposition

Hari

Venugopal

Swaminathan

2020

Electronics Letters

View full text Add to dashboard Cite

Muscle fatigue during isometric contraction of biceps brachii is analysed using synchrosqueezed continuous wavelet transform with Morlet wavelet and singular value decomposition (SVD) features. The recorded surface electromyography signals are decomposed to time frequency matrix using Morlet wavelet and the characteristics are extracted using singular value features such as maximum singular value and zero crossing frequency. The percentage difference in feature values for each segment with the progression of fatigue is calculated. Results show that the recorded signals are complex, non‐stationary, multicomponent, and random in nature. Maximum singular value represents the non‐stationarity of a signal, with an increasing trend towards the fatigue condition. Zero crossing frequency represents the complexity or randomness in the signals and it decreases with the progression of fatigue. It is found that both the features are statistically significant with p < 0.01. It appears that the synchrosqueezed continuous wavelet transform and singular value features are able to analyse fatigue in surface electromyography signals.

show abstract

Segmentation and characterization of acoustic event spectrograms using singular value decomposition

Cited by 29 publications

References 27 publications

End-to-end environmental sound classification using a 1D convolutional neural network

End-to-end environmental sound classification using a 1D convolutional neural network

Hybrid Elman Neural Network and an Invasive Weed Optimization Method for Bridge Defect Recognition

Muscle fatigue analysis in biceps brachii surface electromyography signals using synchrosqueezed Morlet wavelet and singular value decomposition

Contact Info

Product

Resources

About