Pattern recognition based on time-frequency analysis and convolutional neural networks for vibrational events in φ-OTDR

Chengjin Xu; Junjun Guan; Ming Bao; Jiangang Lu; Wei Ye

doi:10.1117/1.OE.57.1.016103

3 January 2018 Pattern recognition based on time-frequency analysis and convolutional neural networks for vibrational events in φ-OTDR

Chengjin Xu, Junjun Guan, Ming Bao, Jiangang Lu, Wei Ye

Author Affiliations +

Funded by: Chinese Academy of Sciences, Natural Science Foundation of China, National Key Research and Development Plan of China

Optical Engineering, Vol. 57, Issue 1, 016103 (January 2018). https://doi.org/10.1117/1.OE.57.1.016103

Abstract

Based on vibration signals detected by a phase-sensitive optical time-domain reflectometer distributed optical fiber sensing system, this paper presents an implement of time-frequency analysis and convolutional neural network (CNN), used to classify different types of vibrational events. First, spectral subtraction and the short-time Fourier transform are used to enhance time-frequency features of vibration signals and transform different types of vibration signals into spectrograms, which are input to the CNN for automatic feature extraction and classification. Finally, by replacing the soft-max layer in the CNN with a multiclass support vector machine, the performance of the classifier is enhanced. Experiments show that after using this method to process 4000 vibration signal samples generated by four different vibration events, namely, digging, walking, vehicles passing, and damaging, the recognition rates of vibration events are over 90%. The experimental results prove that this method can automatically make an effective feature selection and greatly improve the classification accuracy of vibrational events in distributed optical fiber sensing systems.

Citation Download Citation

Chengjin Xu, Junjun Guan, Ming Bao, Jiangang Lu, and Wei Ye "Pattern recognition based on time-frequency analysis and convolutional neural networks for vibrational events in φ-OTDR," Optical Engineering 57(1), 016103 (3 January 2018). https://doi.org/10.1117/1.OE.57.1.016103

Received: 13 July 2017; Accepted: 7 December 2017; Published: 3 January 2018

ACCESS THE FULL ARTICLE

JOURNAL ARTICLE
7 PAGES

DOWNLOAD PAPER SAVE TO MY LIBRARY

GET CITATION

CITATIONS

Cited by 64 scholarly publications.

Explore citations on Lens.org

RIGHTS & PERMISSIONS

Get copyright permission Get copyright permission on Copyright Marketplace

KEYWORDS

Time-frequency analysis

Optical fibers

Signal processing

Sensing systems

Interference (communication)

Convolutional neural networks

Pattern recognition

Show All Keywords

RELATED CONTENT

High-precision snore detection method based on deep learning
Proceedings of SPIE (December 16 2022)

Snoring detection method in sleep based on MBAM-ResNet
Proceedings of SPIE (October 20 2022)

Matrix factorization to time-frequency distribution for structural health monitoring
Proceedings of SPIE (April 20 2016)

Multiple disturbance detection and intrusion recognition in distributed acoustic sensing
Proceedings of SPIE (December 12 2018)

Intrusion signal extraction and recognition for optical fiber perimeter system...
Proceedings of SPIE (December 18 2019)

Pattern Recognition Using Wigner Distribution Function
Proceedings of SPIE (April 15 1983)

Pattern recognition under translation and scale changes
Proceedings of SPIE (October 24 1997)

Subscribe to Digital Library

Receive Erratum Email Alert

Show All Keywords

Keywords/Phrases

Search In:

Publication Years