Development And Application Of Cloud Computing System Of Road Tunnel Surrounding Rock Classification

In this paper,we use the combination of cloud computing,deep learning and image recognition technology,based on the road tunnel surface images,summed up and use of advanced computer and machine learning knowledge at home and abroad,established a set of wisdom for the concept of automatic rock classification system.The system uses tens of thousands of face images to establish the Convolutional Neural Network model,and identifies the distributed features such as joints,fissures,fragmentation,roughness,smoothness,debris mixing and gushing.And the depth of the structural surface is described by using the deep learning technique and the intelligent interpretation method of the rock mass fissure image.The basic address of the area is obtained by using the ip address locating function.The rock type is determined by the color model.The final results of the classification of the surrounding rock were obtained by converting the discriminant factors of the surrounding rock classification into BQ values.The BQ value will be collected to establish a large database,and rent the cloud server will be the above steps in the cloud distributed processing,the formation of the surrounding rock cloud classification system.Through the concrete study,the following results are obtained:(1)The technology of cloud computing and deep learning techniques have promoted the classification of traditional rocks to artificial intelligence.The cloud automatically recognizes the hierarchical results and reduces the reliance on user subjective awareness.(2)The cloud system without extra professional instrument for classification of surrounding rock parameters,constraints on images automatically generated classification result,constraints on it after photos to the computer terminal can fast automatic classification of surrounding rock.The upload process also updates the data and gradually realizes the decision and revision of the results.(3)The cloud system on the degree of hard rock,rock mass integrity index,development degree of groundwater development and structural plane into BQ value is described,the structural plane occurrence correction coefficient with structural plane by replacing the group number and spacing of the method is suitable for the rapid classification in the process of excavation,has higher classification accuracy.(4)The author has successfully developed the classification system for highway tunnel surrounding rock,and applied it to practical engineering.The comparison analysis and verification show that the system is suitable for the classification of the surrounding rock of highway tunnel.At the same time,the system can also be applied to the railway hydraulic tunnel.