Effect of Sampling Rate Reduction and Signal Filtering for Gunfire Sound Classification with Spectral Vector using ANN
Article Sidebar
Main Article Content
Abstract
To identify the type of the shooting gun, there are many important parameters to be considered. One key feature is the gunfire sound. It has been shown [1, 2] that applying the Artificial Neural Network or ANN to the gunfire sound database, the gunfire sound classification can be obtained. The input data to the ANN model was the frequency components in the frequency range at which occupying by the gunfire sound. It is seen that if the number of input frequency components is large, a significant computational resource is required. To reduce such resource so that the whole classification process can be done in a small device; for example, in a mobile phone, the sampling frequency is considered. In this research, the sampling frequency for transforming the input gunfire sound into a digital signal is varied from 44.1 kHz down to 4.41 kHz, so that the effect of the sampling frequency on the gunfire sound classification can be studied. 6 different types of gunfire sound are considered. In order to determine the effectiveness of the classification process, noise is also added to the gunfire sound samples; thus, different values of signal-to-noise ratio are considered. Additionally, the effect of applying different types of signal filtering on the gunfire sound classification is taken into account. It is found that only reducing the sampling frequency on the input gunfire sound signal does not deliver a good performance in terms of gunfire sound classification. To obtained a good classification accuracy. signal filtering has to also be applied to the process. With Chebyshev Type II filter and 4.41 kHz sampling frequency, the obtained classification accuracy is all 100% for the practical range of SNR; that is, between 20 dB down to 5 dB. This impressive classification accuracy comes with a huge reduction on the computational resource; that is, 10 times reduction; since the sampling frequency is reduced from 44.1 kHz to 4.41 kHz. The findings from this work can be certainly applied to the gunfire sound classification system with limited computational resource in order to obtain high classification accuracy.
Article Details
This journal provides immediate open access to its content on the principle that making research freely available to the public supports a greater global exchange of knowledge.
- Creative Commons Copyright License
The journal allows readers to download and share all published articles as long as they properly cite such articles; however, they cannot change them or use them commercially. This is classified as CC BY-NC-ND for the creative commons license.
- Retention of Copyright and Publishing Rights
The journal allows the authors of the published articles to hold copyrights and publishing rights without restrictions.
References
[2] S. Tangkawanita, C. Pinthong, and S. Kanprachar, “Spectral Vector Design for Gunfire Sound Classification System with a Smartphone using ANN,” International Symposium on Wireless Personal Multimedia Communications (WPMC), 2018 21st Joint International 2018., pp. 421 – 426, 2018.
[3] “Boomerang (countermeasure)”, assessed July 30, 2019, https://en.wikipedia.org/wiki/ Boomerang_(countermeasure).
[4] Robert C. Maher, “Acoustical Characterization of Gunshots,” Workshop on Signal Processing Applications for Public Security and Forensics, 2007 International 2007., pp. 1-5, 2007.
[5] Robert C. Maher, “Modeling and Signal Processing of Acoustic Gunshot Recordings,” 12th Digital Signal Processing Workshop & 4th IEEE Signal Processing Education Workshop, 2006 International 2006., pp. 257-261, 2006.
[6] Héctor A. Sánchez-Hevia, David Ayllón, Roberto Gil-Pita, and Manuel Rosa-Zurera, “Maximum Likelihood Decision Fusion for Weapon Classification in Wireless Acoustic Sensor Networks,” IEEE/ACM Transactions on Audio, Speech, and Language Processing, 2017 Volume: 25, Issue: 6, 2017.
[7] Alfonso Chacon-Rodriguez, Pedro Julian, Liliana Castro, Pablo Alvarado, and Néstor Hernandez, “Evaluation of Gunshot Detection Algorithms,” IEEE Transactions on Circuits and Systems I, 2011 Volume: 58, Issue: 2, 2011.
[8] Ângelo Marcio Cardoso Ribeiro Borzino, José Antonio Apolinário, and Marcello Luiz Rodrigues de Campos, “Consistent DOA estimation of heavily noisy gunshot signals using a microphone array,” IET Radar, Sonar & Navigation, 2016 Volume: 10, Issue: 9, 2016.
[9] Sébastien Hengy, Pascal Duffner, Sébastien DeMezzo, Stéphane Heck , Laurent Gross, and Pierre Naz, “Acoustic shooter localisation using a network of asynchronous acoustic nodes,” IET Radar, Sonar & Navigation, 2016 Volume: 10, Issue: 9, 2016.
[10] Jemin George, and Lance M. Kaplan, “Shooter localization using soldier-worn gunfire detection systems,” International Conference on Information Fusion, 2011 Joint 14th International 2011., 2011.
[11] David Grasing and Sachi Desai, “Data fusion methods for small arms localization solutions,” International Conference on Information Fusion, 2012 Joint 15th International 2012., 2012.
[12] Foad Ghaderi, Hamid R. Mohseni, and Saeid Sanei, “Localizing Heart Sounds in Respiratory Signals Using Singular Spectrum Analysis,” IEEE Transactions on Biomedical Engineering, 2011 Volume: 58, Issue: 12, 2011.
[13] V. Elamaran, N. Arunkumar, Ahmed Faeq Hussein, Mario Solarte, and Gustavo Ramirez-Gonzalez, “Spectral Fault Recovery Analysis Revisited With Normal and Abnormal Heart Sound Signals,” IEEE Access, 2018 Volume: 6, 2018.
[14] Li Tan and Jean Jiang, Digital signal processing: fundamentals and applications, Amsterdam: Elsevier, 2013.
[15] B.P. Lathi, Roger Green, Essentials of digital signal processing New York, N.Y.: Cambridge University Press, 2014.
[16] K. Rao, et al., Fast Fourier Transform - Algorithms and Applications (Signals and Communication Technology), New York, Springer, 2010.
[17] Broughton, Allen S., Discrete Fourier analysis and wavelets, New Jersey: John Wiley & Sons, 2009.
[18] Michael Parker, Digital signal processing everything you need to know to get started, Burlington, MA: Newnes/Elsevier, 2010.
[19] Sanjit K. Mitra, Digital signal processing: a computer based approach, Boston: McGraw-Hill, 2011.
[20] Sen M. Kuo, Bob H. Lee and Wenshun Tian, Real-time digital signal processing: fundamentals, implementations and applications, Chichester; New York: Wiley, 2013.
[21] Rajiv J. Kapadia., Digital filters: theory, application and design of modern filters, Weinheim: Wiley-VCH, 2012.
[22] Hercules G. Dimopoulos, “Optimal Use of Some Classical Approximations in Filter Design,” IEEE Transactions on Circuits and Systems II: Express Briefs, 2007 Volume: 54, Issue: 9, 2007.
[23] Fei Xiao, “Fast Design of IIR Digital Filters With a General Chebyshev Characteristic,” IEEE Transactions on Circuits and Systems II: Express Briefs, 2014 Volume: 61, Issue: 12, 2014.
[24] S. Haykin., Neural Networks: A Comprehensive Foundation (First Edition) UK: Prentice – Hall Inc., Pages 156 - 202, 2000.
[25] D. Graupe, Principles of Artificial Neural Networks: Advanced Series on Circuit and Systems Vol-6 (Second Edition) Singapore: World Scientific Publishing companies, Pages 1 – 94, 2007.
[26] V. Sze, et al., “Efficient Processing of Deep Neural Networks: A Tutorial and Survey,” Proceedings of the IEEE 105(12), 2295-2329 (2017).