Research On Remote Sensing Model Integration Framework Based On Microservices

With the continuous development of remote sensing and geography towards multi-disciplinary and multi-field research,a larger number of remote sensing models and geoscience models have been developed by remote sensing scientists and geographers at home and abroad for different geographical problems.These models are highly condensed knowledge of experts and an effective way to solve complex geographical phenomena.They are crucial in future prediction,phenomenon understanding,abnormal diagnosis,comprehensive analysis,and management decision-making.Remote sensing model integration is an important way to synthesize multi-disciplinary models and integrate a variety of remote sensing and geoscience resources to systematically study environmental issues as a whole.However,due to the heterogeneousness and complexity of the model,there are still many problems in interoperability and sharing in the traditional model integration.Aiming to promote the sharing and reuse of models in the field of remote sensing,this paper analyzes the realistic needs of model integration and the general characteristics of remote sensing models,then proposes a remote sensing models integration framework based on microservices.This integration framework realizes the dynamic integration of remote sensing models,and provides an effective way for researchers to call and shared models.The basic ideas studied in this paper are:Firstly,this paper analyses the characteristics of remote sensing models and extracts the necessary attributes required for the description of them.And according to the model description attributes,this paper designs XML-based remote sensing model description language,RSMDL,to express the remote sensing model uniformly.Secondly,through the analysis and research on the operating environment requirements of the remote sensing models,this paper formulates a unified encapsulation specification of them,which makes the file content in the model have a clearer organization structure,so as to improve the efficiency of model integration.Then,this paper introduces the hierarchical structure of the remote sensing model integration framework based on microservices,and analyzes the content and function of the framework module level and interface in detail.The overall business process of the framework is emphasized from the user’s point of view.Finally,this paper verifies the feasibility and operability of the framework proposed in this paper by developing and integrating a classical model in soil conservation model,the revised universal soil loss equation,RUSLE.