Event identification based on sample feature correction algorithm for Φ-OTDR

Abstract

To address the problem of decreased recognition accuracy of event samples in practical Φ-OTDR monitoring scenarios due to external environmental interference, this paper proposes a feature correction algorithm based on sample feature weighting method. By establishing a correlation evaluation method and a weight allocation scheme based on sample feature correlation, combined with the BP algorithm, an average recognition rate of 99.50% for four types of events (climbing, strong wind, koncking and background, 6000 samples) in strong wind environments was achieved, which is 3% higher than the algorithm using BP classifier. The results demonstrate that the proposed algorithm can effectively enhance the performance of Φ-OTDR in complex environments.

Enhancing thickness determination of nanoscale dielectric films in phase diffraction-based optical characterization systems with radial basis function neural networks

“…The neural network approaches in the photonics area have received a lot of interest in the last decade by eliminating the need for human intervention, increasing accuracy, and speeding up the process. It has a wide range of application areas in the open literature, including optical distance measurement [27], defect detection [28], event identification [29], strain sensing [30][31][32], acoustic sensing [33,34], temperature sensing [35], optical fiber bending measurement [36], optical communication networks [37][38][39], optical reflectivity measurements [40,41], tactile sensing [42], leakage detection [43] etc. They have also been used in the thickness characterization of dielectric films due to their importance in fiber optic sensor technologies [44,45].…”

Defect Recognition in Ballastless Track Structures Based on Distributed Acoustic Sensors

“…DASs not only have the advantages of anti-EMI, non-corrosiveness and no required power supply but can also detect and locate weak vibration signals along optical fibers. They have been widely applied in fields [7][8][9][10][11][12] such as oil pipeline security monitoring, perimeter security and rail transit. Collecting the vibration signals of trains and events along a line by using DAS technology combined with signal processing, deep learning and other ways to identify track defects has provided a reliable helper method that has an important application value for the safe and high-quality development of high-speed railways.…”

Vibration Event Recognition Using SST-Based Φ-OTDR System

“…The images were labeled, tagged, and then recognized by You Only Live Once (YOLO) as calm state, rigid collision with the ground, impact protection network, vibration protection network, and cutting protection network of five states and localization; a recognition rate of 96.14% was attained. In 2023 [8], Du X et al performed wavelet transform and feature extraction on the original signal, and then optimized the feature vector using the sample feature correction algorithm. Support Vector Machines (SVM)was then used to recognize watering, climbing, knocking, pressure, as well as a spurious perturbation event.…”