Design And Implementation Of Tunnel Damage Data Monitoring System Based On FCN

As the development of rail transit,railway has become the most preferred mode of transport by Chinese people.With the improved subway network gradually spread across the country,the monitoring and management level for keeping the quality and safety of subway underground tunnels is continuously updated.Nowadays,although 3D laser scanning technology has been used to collect field data in tunnel monitoring,the late transmission,processing,sharing and application of data have not reached the level of systematic management.As a result,there are still fragmented data,information island and other conditions,resulting in a large amount of waste of manpower and resources.Furthermore,if there is any accident occurs,tunnels,as underground operations,will cause serious consequences.Therefore,how to efficiently manage the data of tunnel quality inspection and improve the safety and reliability of subway tunnels has become the desiderative tasks of us at present.On the basis of deep research on the design patterns and technical method of domestic and international rail transit tunnel monitoring data management system,the author analyzed the pros and cons of the current subway tunnel monitoring methods and information management models,constructed an information data management system based on Spring framework,and improved tunnel monitoring data sharing and management and tunnel quality maintenance.At present,the rail transit operation data management system is working well.The main work and innovation of this thesis are as follows:(1)Build data management system.Designed and implemented the data monitoring management system of rail transit operation period based on technologies such as Spring,MyBatis,Vue,etc.,proposed data standardization,successfully solved the problem of 3D laser scanning data information islands,and achieved standardized management of monitoring data,which greatly improved Productivity of relevant personnel.(2)Optimize the system data extraction process.On the basis of improving the accuracy by wavelet transform denoising,the gray scale image of the tunnel,generated according to thecloud data of tunnel monitoring points obtained by using 3D laser scanning technology,is used to extract the water leakage disease data for bring conveniences to the monitors for understanding the situation of the tunnel directly.(3)Achieve the intelligent discrimination of gray scale image disease.The image segmentation model of fully convolutional neural network is chosen to be used because of the characteristics of the gray scale image,which improved the accuracy of image segmentation,automatically marked the water leakage disease in the gray scale image and solved the disadvantage that the gray scale image displayed in the system needs manual marking.(4)Achieve the operation data visualization.Provide users with various data presentations full of tables and charts,so as to facilitate users’ more intuitionistic understanding of tunnel monitoring data such as tunnel ovality,tunnel leakage,tunnel wrong platform,etc..Also,the speeded-up loading of imagines will provide better user experience.