Pattern recognition based on pulse scanning imaging and convolutional neural network for vibrational events in Φ-OTDR

Linear infrastructures, such as railways, tunnels, and pipelines, play essential roles in economic and social development worldwide. However, under the influence of geohazards, earthquakes, and human activities, linear infrastructures face the potential risk of damage and may not function properly. Current monitoring systems for linear infrastructures are mainly based on non-contact detection (InSAR, UAV, GNSS, etc.) and geotechnical instrumentation (extensometers, inclinometers, tiltmeters, piezometers, etc.) techniques. Regarding monitoring sensitivity, frequency, and coverage, most of these methods have some shortcomings, which make it difficult to perform the accurate, real-time, and comprehensive monitoring of linear infrastructures. Distributed acoustic sensing (DAS) is an emerging sensing technology that has rapidly developed in recent years. Due to its unique advantages in long-distance, high-density, and real-time monitoring, DAS arrays have shown broad application prospects in many fields, such as oil and gas exploration, seismic observation, and subsurface imaging. In the field of linear infrastructure monitoring, DAS has gradually attracted the attention of researchers and practitioners. In this paper, recent research and the development activities of applying DAS to monitor different types of linear infrastructures are critically reviewed. The sensing principles are briefly introduced, as well as the main features. This is followed by a summary of recent case studies and some critical problems associated with the implementation of DAS monitoring systems in the field. Finally, the challenges and future trends of this research area are presented.

show abstract

“…In recent years, many successful trials have been performed [ 87 , 88 , 89 , 90 ]. In 2015, Wu et al applied DAS to pipeline leakage monitoring.…”

Section: Applications Of Das In Linear Infrastructure Monitoringmentioning

confidence: 99%

Distributed Acoustic Sensing for Monitoring Linear Infrastructures: Current Status and Trends

Zhu

Liu

Wang

et al. 2022

Sensors

View full text Add to dashboard Cite

show abstract

“…It has significant research value for the collection, extraction, positioning and recognition of conventional vibration signals [1,2]. In recent years, domestic and foreign scholars mainly identify different types of vibration events by extracting the time domain and frequency domain information of vibration signals [3,4]. Based on the φ-OTDR, AKTAS M and others adopted a method that combines short-time Fourier transform (STFT) with 5-layer neural network, and the classification accuracy of shovel digging, pick digging, strong wind blowing and other classic events exceeded 93% [5].…”

Section: Introductionmentioning

confidence: 99%

Research on vibration pattern recognition based on phase‐sensitive optical time domain reflectometry and voting fully convolution neural networks

Liao,

Li,

Gong

2024

IET Optoelectronics

View full text Add to dashboard Cite

A method that combines phase‐sensitive optical time domain reflectometry with deep learning to construct new voting fully convolution neural networks (VoteFCNs) is proposed. Compared to the traditional convolutional network, the VoteFCN can be input with data of random size and requires less parameters so that the training speed can be improved greatly. The recognition results can be more accurate and more reliable if we use classification voting count and average recognition rate as the criteria to judge network training quality. At last, the training and identification were conducted by simulating such several disturbance events: walking, raining, climbing fence, hammering the ground optical fibre and normal outdoor environments. The results show that the average test accuracy of this method is about 93.4%.

show abstract

“…When the morphologic features of time-space domain signals are extracted and obtained in the experiment, the relevance vector machine (RVM) can be used to achieve a relatively short recognition time of below 1 s for vibration behavior [ 10 ]. In addition, convolutional neural networks (CNNs) can effectively classify a single vibration activity image without expert knowledge [ 11 , 12 ].…”

Section: Introductionmentioning

confidence: 99%

Real-Time Φ-OTDR Vibration Event Recognition Based on Image Target Detection

Yang

Zhao

Chen

2022

Sensors

View full text Add to dashboard Cite

Accurate and fast identification of vibration signals detected based on the phase-sensitive optical time-domain reflectometer (Φ-OTDR) is crucial in reducing the false-alarm rate of the long-distance distributed vibration warning system. This study proposes a computer vision-based Φ-OTDR multi-vibration events detection method in real-time, which can effectively detect perimeter intrusion events and reduce personnel patrol costs. Pulse accumulation, pulse cancellers, median filter, and pseudo-color processing are employed for vibration signal feature enhancement to generate vibration spatio-temporal images and form a customized dataset. This dataset is used to train and evaluate an improved YOLO-A30 based on the YOLO target detection meta-architecture to improve system performance. Experiments show that using this method to process 8069 vibration data images generated from 5 abnormal vibration activities for two types of fiber optic laying scenarios, buried underground or hung on razor barbed wire at the perimeter of high-speed rail, the system mAP@.5 is 99.5%, 555 frames per second (FPS), and can detect a theoretical maximum distance of 135.1 km per second. It can quickly and effectively identify abnormal vibration activities, reduce the false-alarm rate of the system for long-distance multi-vibration along high-speed rail lines, and significantly reduce the computational cost while maintaining accuracy.

show abstract

Pattern recognition based on pulse scanning imaging and convolutional neural network for vibrational events in Φ-OTDR

Cited by 19 publications

References 21 publications

Distributed Acoustic Sensing for Monitoring Linear Infrastructures: Current Status and Trends

Distributed Acoustic Sensing for Monitoring Linear Infrastructures: Current Status and Trends

Research on vibration pattern recognition based on phase‐sensitive optical time domain reflectometry and voting fully convolution neural networks

Real-Time Φ-OTDR Vibration Event Recognition Based on Image Target Detection

Contact Info

Product

Resources

About