Based On Image Resource Digest Representation And Distributed Ensemble Method Research

Technology service improves independent innovation ability for city group,it has great significance to cultivate emerging industries,collaborating with the city group technology service the cloud computing platform puts forward urgent requirements for efficient service resource collaboration scheduling mechanism and ensemble method.As an important technology service in the automobile production industry,vehicle model image is an important part of business process.Vehicle image representation method and vehicle automatic identification,supporting efficient integration of technology services and collaborative scheduling mechanisms become the focus of this paper.To this end,this paper proposes a vehicle retrieval method for based on image resource digest representation and distributed ensemble method research.Firstly,this paper studies a method for digest representation of vehicle image resources,extracts important features of image resources,and integrates retrieval directories with small data.In order to test the effectiveness of the digest representation method in vehicle image retrieval,and this article conducts relevant experiments for vehicle retrieval cases.The experimental results show that the average accuracy rate of vehicle model retrieval reaches 90.78% at a recall rate of 90%,which proves that the image digest representation can be used for retrieval services after resource integration.Secondly,in order to improve the real-time transmission and reduce network traffic,this paper designs a distributed digest hierarchical network structure based on edge computing.Data is transmitted from the distributed storage space to the upper layer of the tree structure,ensure the correctness of resource transmission between multi-core and multi-tier cities and improve storage efficiency.Then,we established a system analysis model based on Petri-net,using different parameters,observe the ability requirements for storage,calculation,and transmission for the image resourcedigest representation and distributed ensemble method,it can be used to find system bottlenecks early and achieve multi-party balance.The simulation results show that the method proposed in this paper saves 2250 times the resource storage space and shortens the resource integration delay by 9.15 times compared with the single vehicle target image.Finally,through the introduction of the prototype system,the image resource digest library and retrieval process are described.Based on image resource digest representation and distributed ensemble method proposed in this paper provides new ideas and technical support for the ensemble and management of vehicle image resources.