基于计算机视觉的桥梁健康监测数据异常检测  

Anomaly Detection of Bridge Health Monitoring Data Based on Computer Vision

作　　者：杨志 钟其仁 YANG Zhi;ZHONG Qiren(Huoshan County Transportation Bureau,Huoshan 237200,China;Anhui Civil Engineering Research Center for Disaster Prevention and Mitigation,Hefei 230009,China;School of Civil and Hydraulic Engineering,Hefei University of Technology,Hefei 230009,China)

机构地区：[1]霍山县交通运输局,安徽霍山237200 [2]土木工程防灾减灾安徽省工程技术研究中心,安徽合肥230009 [3]合肥工业大学土木与水利工程学院,安徽合肥230009

出　　处：《安徽建筑大学学报》2024年第6期58-65,共8页Journal of Anhui Jianzhu University

基　　金：国家自然科学基金资助项目(51878234)。

摘　　要：针对桥梁健康监测系统中因设备故障、环境影响等导致部分监测数据产生异常,将时频分析和计算机视觉相结合,用于监测数据的异常检测。首先根据监测数据时域图像对其进行类型划分和标记,采用时频分析法实现监测数据样本可视化,制备图像数据库用于构建和训练深度神经网络模型。然后利用深度学习框架搭建ResNet18神经网络模型,通过反向传播机制和Adam优化算法优化模型权重参数,使用批标准化、数据增强等方法提高模型准确率和泛化能力。最后使用完成训练的模型对桥梁监测数据样本进行异常检测,验证模型性能。结果表明,所提方法对监测数据异常检测准确率为95.62%,对其他桥梁的监测数据样本检测准确率为95.28%,具有良好的稳定性和识别性能。In order to identify the abnormal monitoring data caused by equipment failures and environmental influences in the bridge health monitoring system,a combined approach of time-frequency analysis and computer vision is proposed for anomaly detection.The monitoring data are categorized based on its time-domain images and are visualized by the time-frequency analysis method,then an image database was built for the deep neural network model training.Subsequently,a ResNet18 neural network model is built using a deep learning framework.The model's weight parameters are optimized through the backpropagation mechanism and the Adam optimization algorithm,and the accuracy as well as generalization are improved using methods such as batch normalization and data augmentation.Finally,the trained model is applied to detect anomalies in other bridge monitoring data samples,further validating its performance.The results show that the proposed method achieves an accuracy of 95.62%in detecting abnormal monitoring data and 95.28%in detecting other bridge monitoring data samples,demonstrating good stability and recognition performance.

关 键 词：数据异常 时频分析 计算机视觉 桥梁健康监测 

分 类 号：U446.3[建筑科学—桥梁与隧道工程]