| Since the first Geographic Information System(GIS)appeared in 1960s,it has been widely used in various fields of society.With the development of Web technology,geographic information System has gradually changed from "system" to "Service".However,Web technology is a universal standard,which is difficult to solve the problems in the field of surveying and mapping.Therefore,the Open Geospatial Information Alliance(OGC)and other organizations have formulated a series of standards and specifications in the surveying and mapping(GIS)industry,which are used to realize the standardization of data sharing and interoperating Web services among different GIS systems,among which the Web Processing Service(WPS)is aimed at processing surveying and mapping geospatial information.However,some geospatial data may have the characteristics of large amount of data and complex processing operations,such as processing INSAR data.Among INSAR technologies,D-INSAR and PS-INSAR technologies involve the matching of multiple SAR images,and also involve a wide range of matrix operations(such as Fourier transform or convolution,etc.),which leads to excessive data transmission of spatial data from client to server,especially when performing service composition,data may be transmitted in distributed servers,making service difficult to complete;The large amount of calculation of spatial data leads to great calculation pressure on the server;Data is easy to be leaked,and it is difficult to ensure data security.Aiming at the deficiency of OGC geographic information service,this paper attempts to apply containerization technology to WPS service.Firstly,based on Docker technology,the program is packaged into Docker image,and spatial data migration is replaced by image migration.The theoretical model and implementation technology of WPS service based on image migration are proposed.Based on this model,a method of service composition based on micro-service is further studied.Finally,based on this model,a single service of 17 INSAR data registration and a service combination of slope and aspect extraction and fusion are realized.As a result,the transmission of SAR data on the network is avoided,and the calling efficiency of INSAR matching WPS service and the feasibility of service combination of slope and aspect extraction and fusion are significantly improved.The details are as follows:(1)the principle,architecture,key implementation technologies of Web services and OGC WPS service standards(coding of requests and responses of main interfaces)are studied.On the basis of Web service mode,this paper analyzes the defects of existing OGC specifications,and points out the solutions.(2)The Docker container technology is studied,including image layered construction,network definition and use,remote API specification and implementation technology,data volume and so on.On this basis,the concept and theoretical model of WPS service based on image migration are put forward to solve the shortage of WPS service.On the basis of the original standard,the standard interface is legally extended to be compatible with the original standard.Finally,based on the characteristics of the model,the idea of micro-service is proposed to solve the problem of service composition deployment and implementation.(3)Based on the theoretical model of WPS service of mirror migration,the overall technical framework of user layer,WPS core business layer and processing function layer is put forward,and the implementation technology and functions of each component of the mirror migration WPS service model are designed.Based on Docker API technology,the image migration WPS service is user-friendly,easy to implement and manage,and compatible with the original specification.At the same time,the implementation technology of micro-service architecture for this service composition is studied.(4)Based on the theoretical model and implementation technology of image migration WPS service proposed in this paper,the service combination experiments of INSAR data registration monomer service based on phase correlation method and slope,aspect extraction and fusion experiment are realized respectively.By analyzing the efficiency of spatial data transmission,it is concluded that the theory and technology proposed in this paper have obvious advantages in processing large amounts of data.Finally,a scheme to solve the defects of this model and technology when the amount of data is small is also proposed. |
PDF Full Text Request