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

Nachuan Yang,Yongjun Zhao,Jinyang Chen

doi:10.3390/s22031127

Abstract

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.

Highlights

Received: 18 January 2022Phase-sensitive optical time-domain reflectometer (Φ-OTDR) is a simple and effective method to measure single-mode fiber vibration [1,2], which has the advantages of distributed detection capability up to 100 km at a single station [3], as well as high resolution, anti-electromagnetic interference, corrosion resistance, low energy loss, flame and explosion resistance [4], and can be combined with Raman pumping amplification to achieve long-range distributed abnormal vibration location inspection [5], which makes Φ-OTDR extensively employed in perimeter security and oil pipeline monitoring.the long-range detection and high sensitivity of Φ-OTDR will inevitably result in high nuisance-alarm rates (NARs)
The contributions of this study are: (1) The effect of parameter adjustment in the signal pre-processing module on inspection is illustrated from the perspective of statistical theory; (2) An improved You Only Look Once (YOLO) model for high-speed railway perimeter abnormal vibration detection is proposed to achieve single detection of multiple vibration targets; (3) The effect of different feature extraction networks on the detection performance of the improved YOLO model is tested and discussed; (4) The detection performance of the improved YOLO model with other State-of-The-Art (SOTA)
The combinations for different vibration classes in the test set are picked to simulate six different kinds of mixed vibration events to visualize the results of multitarget detection of YOLO-A30 in specific scenarios

Summary

Introduction

Phase-sensitive optical time-domain reflectometer (Φ-OTDR) is a simple and effective method to measure single-mode fiber vibration [1,2], which has the advantages of distributed detection capability up to 100 km at a single station [3], as well as high resolution, anti-electromagnetic interference, corrosion resistance, low energy loss, flame and explosion resistance [4], and can be combined with Raman pumping amplification to achieve long-range distributed abnormal vibration location inspection [5], which makes Φ-OTDR extensively employed in perimeter security and oil pipeline monitoring. The long-range detection and high sensitivity of Φ-OTDR will inevitably result in high nuisance-alarm rates (NARs). The interference signals are caused by the airflow in the detection environment. Based on the hardware realization of long-distance vibration detection, the subsequent signal processing and vibration classification problems become the essential factors for Φ-OTDR to be able to identify abnormal vibration signals precisely, and the precision of the identification significantly affects the eventual false alarm probability of the system.

Objectives

Results

Conclusion