An Ensemble One Dimensional Convolutional Neural Network with Bayesian Optimization for Environmental Sound Classification

Ragab, Mohammed Gamal; Abdulkadir, Said Jadid; Aziz, Norshakirah; Alhussian, Hitham; Bala, Abubakar; Alqushaibi, Alawi

doi:10.3390/app11104660

Cited by 26 publications

(4 citation statements)

References 94 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The final model achieved a 91% accuracy rate on the test dataset, which suggests that the method effectively categorizes urban sounds. Similarly, [13] presents a novel approach for classifying environmental sounds, which involves using a one-dimensional CNN along with Bayesian optimization and ensemble learning. The proposed end-to-end model directly extracts feature representations from audio signals through…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Preventing Crimes Through Gunshots Recognition Using Novel Feature Engineering and Meta-Learning Approach

Raza,

Rustam,

Mallampati

et al. 2023

IEEE Access

View full text Add to dashboard Cite

Gunshot sounds are common in crimes, particularly those involving threats, harassment, or killing. The gunshot sounds in crimes can create fear and panic among victims, often leading to psychological trauma. Gunshot sounds are associated with a significant mortality rate, especially in cases of gun violence. The sound of gunshots can serve as evidence in criminal investigations, allowing law enforcement officials to determine the number of shots fired, the caliber of the gun used, and the distance from which the shots were fired. Efficient gunshot detection is necessary to address the issue of gun violence in society. This study aims to detect gunshot sounds using an efficient approach to prevent crimes. The frequency-time domain spectrum analysis is performed to understand the patterns of signals related to each target class. A novel Discrete Wavelet Transform Random Forest Probabilistic (DWT-RFP) feature engineering approach is proposed, which takes Mel-frequency cepstral coefficients (MFCC) extracted from gunshot sound data as input for feature extraction. A novel meta-learning-based Meta-RF-KN (MRK) is proposed to detect gunshots based on newly created ensemble features with a DWT-RFP approach. For experiments, the gunshot sounds dataset containing 851 audio clips collected from public videos on YouTube from eight kinds of gun models, is used. Advanced machine learning and deep learning techniques are applied in comparison to evaluate the performance of the proposed approach. Extensive experiments show that the proposed MRK approach achieves 99% k-fold accuracy for detecting gunshots and outperforms state-of-the-art approaches. The proposed approach can potentially be used for accurate gunshot detection and to help prevent crimes.

show abstract

Section: Related Workmentioning

confidence: 99%

“…AReN Accuracy: 91% [12] 2021 Short audio signals and spectrograms. Convolutional neural network Accuracy: 91% [13] 2021 UrbanSound8K Convolutional neural network Accuracy: 94% [14] 2022 AudioSet and the NIJ Grant 2016-DN-BX-0183 gunshot dataset.…”

Section: Refmentioning

confidence: 99%

Preventing Crimes Through Gunshots Recognition Using Novel Feature Engineering and Meta-Learning Approach

Raza,

Rustam,

Mallampati

et al. 2023

IEEE Access

View full text Add to dashboard Cite

show abstract

“…(Sheltami et al 2016) categorize the leak detection and localization methods as hardware-based, software-based, and non-technical based on the degree of human involvement and level of automation. Under software-based methods for leak detection, applied algorithms are Support Vector Machine (SVM) (Lang et al 2017;Shravani et al 2019a, b), Logistic Regression (LR) (Shravani et al 2019b), Decision Tree (DT) (Shravani et al 2019b;Fereidooni et al 2021), Random Forest (RF) (Fereidooni et al 2021), Naïve Bayes (NB) (Shravani et al 2019b;Fereidooni et al 2021), K Nearest Neighbors (KNN) (Fereidooni et al 2021), Multi-Layer Perceptron (MLP) (Shravani et al 2019a, b), and Convolutional Neural Network (CNN) (Kang et al 2018;Zhou et al 2019Zhou et al , 2021Gautam and Singh 2020;Ragab et al 2021;Massaro et al 2021;Yang et al 2022;Man et al 2022) and for leak localization, mainly applied algorithm is cross-correlation (Lang et al 2017;Kothandaraman et al 2020aKothandaraman et al , b, 2022. Shravani et al tested different algorithms to detect leaks to identify the perfect algorithm for flow rate sensor data.…”

Section: Introductionmentioning

confidence: 99%

Water Pipeline Leak Detection and Localization With 1DCNN and Novel Adaboost Classifier

Kothandaraman

Rajasekaran

Pua

2023

Preprint

View full text Add to dashboard Cite

A pipeline leak localization technique is crucial in a structural health monitoring system to prevent water wastage at an early stage. Recent advancements include machine learning and deep learning techniques to detect a leak with great accuracy, and cross-correlation is the widely used tool for localization. All the existing algorithms use two architectures to detect and localize a leak. The main aim of this paper is to propose a standalone architecture for leak detection and localization with 1DCNN and novel AdaBoost architecture. The data collected from a real-time pipeline setup is redundant, very dynamic, and contains a substantial amount of surrounding noise. So an appropriate feature extraction technique helps to capture essential information from pipeline data. In the proposed method, the 1DCNN extracts the essential features from the data collected from the real-time pipeline setup located in UTAR, Malaysia, using an Acousto-optic vibration sensor. The proposed novel AdaBoost architecture localizes the leak with the help of the features extracted using 1DCNN. The performance metrics used in this paper to evaluate the proposed method are training accuracy, validation accuracy, precision, recall, F1 score, and R-squared, and the results are 99.23, 97.26, 98.18, 98.19, 98.14, and 98.55, respectively which is better compared to the implemented existing method. The results show that the proposed 1DCNN-novel AdaBoost localizes the leak with high training and validation accuracy and performs well in other metrics.

show abstract

“…The frequently applied machine learning methods include K-Nearest Neighbor (k-NN) [1,2,3,4,5,6], Support Vector Machine (SVM) [7,8,9], Long Short-Term Memory (LSTM) [10,11,12,13,14,15], and Convolutional Neural Network (CNN) and its variants [4,16,17,18,19,20,21,22,23,24], and ensemble models [25,26,27,28,29,30,31,18]. However, a majority of research and experiments done within the field of musical instrument recognition or music classification are targeted at those belonging to western culture, mostly of European and North American origin.…”

Section: Introductionmentioning

confidence: 99%

Deep Chnet 1-D CNN using raw audio waveform for recognizing traditional Chinese musical instrument

2023

ACE

View full text Add to dashboard Cite

The classification of music information using various deep learning models is increasingly popular in the field of Music Information Retrieval research. However, as most proposed works focus on western music and musical instruments, little attention is given to traditional Chinese music. This paper proposes a 1-D Convolutional Neural Network (1-D CNN) using only raw audio waveform as input, to undertake the task of traditional Chinese musical instruments classification. This paper starts with a review of the current state of research on the related field, then discuss the proposed model and its data in detail, followed by its performance metrics and then a conclusion on the experiment. The result shows that 1-D CNN provides competitive and even superior results when compared to its 2-D versions as well as when compared to traditional models.

show abstract

An Ensemble One Dimensional Convolutional Neural Network with Bayesian Optimization for Environmental Sound Classification

Cited by 26 publications

References 94 publications

Preventing Crimes Through Gunshots Recognition Using Novel Feature Engineering and Meta-Learning Approach

Preventing Crimes Through Gunshots Recognition Using Novel Feature Engineering and Meta-Learning Approach

Water Pipeline Leak Detection and Localization With 1DCNN and Novel Adaboost Classifier

Deep Chnet 1-D CNN using raw audio waveform for recognizing traditional Chinese musical instrument

Contact Info

Product

Resources

About