Abstract
The paper deals with the detection of abnormal situations via captured sound processing. Different settings of feature extraction algorithms were realized and evaluated. Chosen feature sets were used for building the effective parametric representation for gun shots and breaking glass. This way two types of high dimensional feature supervectors were created in regard to the best individual settings of each feature extraction algorithm. For improving the recognition rate Minimum Redundancy Maximum Relevance (MRMR) and Joint Mutual Information (JMI) feature selection algorithms were also applied. They were used for the selection of superior features and for the creation of n-dimensional feature supervectors. The investigation of the appropriate dimension of feature supervectors was performed too. The framework for recognition of potentially dangerous acoustic events such as breaking glass and gun shots, based on the MRMR and JMI selected feature supervector through Hidden Markov Models based classification is proposed in the paper.
Similar content being viewed by others
Notes
K low = floor(62,5/ΔF), where ΔF is interval between two FFT bins
References
Bach JH, Kayser H, Anemller J (2012) Audio classification and localization for incongruent event detection. Detection and identification of rare audiovisual cues, studies in computational intelligence, vol 384, pp 39–46
Barras C, Geoffrois E, Wu Z, Liberman M (1998) Transcriber: a free tool for segmenting, labeling and transcribing speech. In: First international conference on language resources and evaluation (LREC), pp 1373–1376
Brown G, Pocock A, Zhao MJ, Luján M (2012) Conditional likelihood maximisation: a unifying framework for information theoretic feature selection. J Mach Learn Res 13:27–66
Cruz R, Ortiz A, Barbancho AM, Barbancho I (2012) Unsupervised Classification of Audio Signals by Self-Organizing Maps and Bayesian Labeling, Hybrid Artificial Intelligent Systems. Lecture notes in computer science, vol 7208, pp 61–70
Ding C, Peng H (2003) Minimum redundancy feature selection from microarray gene expression data. In: Bioinformatics conference, pp 523–528
Han J, Kamber M (2006) Data mining: concepts and techniques, 2nd edn. Morgan Kaufmann publisher. ISBN 10:1-55860-901-6
INDECT project [online] (2000) Available at: http://www.indect-project.eu/
Jain AK, Duin RPW, Mao J (2000) Statistical pattern recognition: A review. Trans Pattern Anal Mach Intell 22(1):4–32
Kiktova E, Lojka M, Pleva M, Juhar J, Cizmar A (2013) Comparison of different feature types for acoustic event detection system. In: Communications in Computer And Information Science: Multimedia Communications, Services and Security, vol 368, pp 288–297
Kim HG, Moreau N, Sikora T (2005) MPEG-7 audio and beyond: Audio content indexing and retrieval. Wiley, New York, pp 304
Kotus J, Lopatka K, Czyzewski A (2012) Detection and localization of selected acoustic events in acoustic field for smart surveillance applications. Multimed Tools Appl. doi:10.1007/s11042-012-1183-0
Kotus J, Lopatka K, Kopaczewski K, Czyzewski A (2010) Automatic Audio-Visual Threat Detection. In: Proceedings of the IEEE international conference on multimedia communications, services and security mCSS 2010, pp 140–144
Latané B, Darley JM (1969) Bystander apathy. Am Sci 57(2):244–268
Lin C-H, Tu M-C, Chin Y-H, Liao W-J, Hsu C-S, Lin S-H, Wang J-C, Wang J-F (2012) SVM-Based Sound Classification Based on MPEG-7 Audio LLDs and Related Enhanced Features, Convergence and Hybrid Information Technology Communications in Computer and Information Science, vol 310, pp 536–543
Molina LC, Belanche L, Nebot A (2002) Feature selection algorithms: a survey and experimental evaluation. In: Proceedings EEE international conference on data mining, pp 306–313
Peng H, Long F, Ding C (2005) Feature selection based on mutual information: criteria of max-dependency, max-relevance, and min-redundancy. IEEE Trans Pattern Anal Mach Intell 27(8):1226–1238
Pleva M, Vozarikova E, Dobos L, Cizmar A (2011) The joint database of audio events and backgrounds for monitoring of urban areas. JEEE 4(1):185–188
Psutka J, Müller L, Psutka JV (2001) Comparison of MFCC and PLP parametrizations in the speaker independent continuous speech recognition task. In: Eurospeech, pp 1813–1816
Pudil P, Novovicova J, Kittler J (1994) Floating Search Methods in Feature Selection. Pattern Recogn Lett 15(11):1119–1125
Saeys Y, Inza I, Larranaga P (2007) A review of feature selection techniques in bioinformatics. Bioinformatics 23(19):2507–2517
Sun Y (2007) Iterative RELIEF for feature weighting: algorithms, theories, and applications. IEEE Trans. Pattern Anal. Mach. Intell. 29(6):1035–1051
Tourassi GD, Frederick ED, Markey MK, Floyd CE (2001) Application of the mutual information criterion for feature selection in computer-aided diagnosis. Med Phys 28(12):2394–2402
Tzanetakis G, Cook P (2002) Musical genre classification of audio signals. IEEE Trans Speech Audio Process 10(5):293–302
Vavrek J, Vozarikova E, Pleva M, Juhar J (2012) Broadcast news audio classification using SVM binary trees, In: TSP 2012, pp 469–473
Vozarikova E, Juhar J, Cizmar A (2011) Acoustic Events Detection Using MFCC and MPEG-7 Descriptors. In: Communications in Computer And Information Science: Multimedia Communications, Services and Security, vol 149, pp 191–197
Vozarikova E, Juhar J, Cizmar A (2011) Study of audio spectrum flatness for acoustic events recognition. In: Digital Technologies 2011, pp 295–298. Žilina
Vozarikova E, Lojka M, Juhar J, Cizmar A (2012) Performance of Basic Spectral Descriptors and MRMR Algorithm to the Detection of Acoustic Events. In: Communications in Computer and Information Science : Multimedia Communications, Services and Security, No. 287, pp 350–359
Willett P (1988) Recent trends in hierarchic document clustering: a critical review. Inform Process Manag 24:577–597
Yang H, Van Vuuren S, Sharma S, Hermansky H (2000) Relevance of time-frequency features for phonetic and speaker-channel classification. Speech Commun 31(1):35–50
Yang HH, Moody J (1999) Feature selection based on the join mutual information. In: Proc: Conf. Advances in Intelligent Data Analysis, Computational Intelligence Methods, and Applications
Young S et al (2006) The HTK Book, Cambridge University, pp. 368
Zhu Y, Ming Z, Huang Q (2007) SVM-based audio classification for content-based multimedia retrieval. In: Multimedia Content Analysis and Mining, Lecture Notes in Computer Science, vol 4577, pp 474–482
Acknowledgements
This work has been performed partially in the framework of the EU ICT Project INDECT (FP7 - 218086) (25 %) and under research project VEGA 1/0386/12 (25 %) supported by the Ministry of Education of Slovak Republic and under research project ITMS-26220220155 (50 %) supported by the Research & Development Operational Programme funded by the ERDF.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Kiktova-Vozarikova, E., Juhar, J. & Cizmar, A. Feature selection for acoustic events detection. Multimed Tools Appl 74, 4213–4233 (2015). https://doi.org/10.1007/s11042-013-1529-2
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11042-013-1529-2