Research Of The Modeling Method For Monitoring Tunnel Deformation Based On Active Stereo Omni-directional Vision Sensor

As tunnel is really an important part of the transport facility,it is necessary to inspect and keep the tunnel system in good condition periodically.The main rock mass structure appears varying degrees in different parts of deformation and significantly lower than design life due to the technology limitations and force majeure during the tunnel construction and its operation.And then the tunnel health problems impact on the path transportation security.How to detect the tunnel deformation quickly becomes the urgent problem to solve.Most of the existing non-machine vision detection methods have the disadvantages of large measurement errors,long time,less-automated and susceptibility to the surrounding environment.There are some shortcomings of machine vision detection method,such as high cost,large number of images acquisition,complicated data processing and poor real-time performance.In view of the above problems and existing methods,this thesis designs and implements the modeling method for monitoring tunnel deformation based on active stereo omni-directional vision sensor.Through the hardware device and software system of tunnel deformation detection,the panoramic laser images of the tunnel lining are collected and then processed image.After the precision analysis and the error analysis of the system,the deformation information of the tunnel lining is finally obtained.The main work of this thesis is as follows:1.Research on the hardware device of tunnel deformation detection system.The hardware part of the tunnel deformation detection device is mainly composed of tunnel detection robot,omni-directional vision sensor(ODVS)and circular laser light source.ODVS and circular laser light source are coaxially connected to make up the active stereo omni-directional vision sensor(ASODVS)for obtaining the data source-panoramic laser image.2.Research on software design of tunnel deformation detection system.The module of the system can be divided into image acquisition module and image processing module in accordance with the design principle of the software system and the system function analysis,the former runs on the tunnel detection robot,including the robot motion control module and the video image acquisition module.the latter runs on a specific processor,including the image enhancement module,the subpixel centerlines extracted of laser stripes module,a 3D coordinate acquisition module,and a deformation analysis module.3.Research on accuracy analysis and error analysis.Through imaging model of ASODVS and its parameter table,the precision and error of the system are analyzed combined with the actual measurement environment and experimental data.And then the causes of error and influence precision are analyzed from the system hardware and system software.Finally,the effective solution to improve the system measurement accuracy is discussed.