Phase-OTDR-event-detection
收藏资源简介:
该数据集由伊兹密尔理工学院团队构建,基于Phase-OTDR系统采集的光纤振动事件数据,包含15,612条样本,涵盖背景、挖掘、敲击等6类事件。数据以12×10,000的强度矩阵形式存储,通过Gramian角度场和递归图等数学方法转化为多通道RGB图像。其创新性在于将一维时序数据转换为视觉可解释的二维表征,显著提升了深度学习模型对复杂光纤传感数据的解析能力,主要应用于管道、铁路等基础设施的实时安全监测领域。
This dataset was developed by a research team from Izmir Institute of Technology. It is built upon fiber optic vibration event data collected using the Phase-OTDR system, containing a total of 15,612 samples covering six categories of events including background, excavation, tapping, and others. The raw data is stored as a 12×10,000 intensity matrix, and is converted into multi-channel RGB images via mathematical approaches such as Gramian Angular Field (GAF) and Recurrence Plot. The core innovation of this dataset lies in converting one-dimensional time-series data into visually interpretable two-dimensional representations, which significantly enhances the analytical capability of deep learning models for complex fiber optic sensing data. It is primarily applied in the field of real-time safety monitoring for infrastructure such as pipelines and railways.
Phase-OTDR-event-detection 数据集概述
数据集来源
- 该数据集与题为“Phase-OTDR Event Detection Using Image-Based Data Transformation and Deep Learning”的论文相关联。
- 作者包括:Muhammet Cagri Yeke, Samil Sirin, Kivilcim Yuksel Aldogan, Abdurrahman Gumus。
数据集状态
- 代码和数据集将在论文发表后提供。
- 当前访问地址为:https://github.com/miralab-ai/Phase-OTDR-event-detection
数据集用途
- 用于基于图像数据转换和深度学习的相位光时域反射计(Phase-OTDR)事件检测研究。

- 1Phase-OTDR Event Detection Using Image-Based Data Transformation and Deep Learning伊兹密尔理工学院, 伊斯帕塔应用科学大学 · 2025年



