Design And Implementation Of City Real-time Scenery Sharing And Management System

As a new information carrier,pictures are widely welcomed by people with their intuitive,concise and vivid visual experience.With the advent of the information age,the rapid development of mobile Internet and cloud storage technology,the quality of pictures is constantly improving.Simultaneously,the cost of picture acquisition and dissemination is getting lower and lower.People use smart phones and professional photography devices to record their personal lives and beautiful scenery around them,and share them on the platform,this way has gradually become a basic way of life for people,and social software for pictures has emerged.Based on the analysis of users' needs and the comparative analysis of social software of image sharing,this paper decides to provide users with a software that can enjoy the beautiful scenery of different places without leaving home,and also provided users with software that can enjoy the beauty of various places without leaving the home,and also provide a platform for sharing and communication for users who like photography,and realize the functions of taking and sharing.Users can use the fragmented time to use the software at their leisure.The software adapt to the modern people's lifestyle,and fit the development trend of mobile sharing social networks.The system is divided into seven modules,including personal center,message notification,real-time scenery dynamic,around the moment,real-time scenery activity,search management and observe clouds to identify the weather.The core function modules of the system are the real-time scenery dynamics,around the moment and realtime scenery activity.Users can view the dynamic which released by others,view the real-time scenery pictures around them,or participate in the real-time scenery activity and share the real-time scenery around them.Search management provides users with near real-time city scenery search and user search.Users can upload and share cloud pictures in the modules of observe the clouds to identify the weather,and the system can predict the weather by identifying the cloud shape.Based on the requirements of the project,determines to use the microservice architecture as the overall architecture of the system,uses Dubbo and Zookeeper distributed service framework,Nginx reverse proxy server and Spring Boot framework to carry out the background development of the system,uses Elastic Search search engine provides search service.For the model of observe clouds to identify the weather,using transfer learning to compare experiments with pre-trained convolutional neural network models on Image Net,the classification accuracy and running time of the Resnet50 model are better than other models,so it is selected as the classification model for cloud classification.In this paper,the demand analysis,outline design,detailed design and implementation of the urban real-time scenery sharing and management system are described in detail and the operation effect diagram of the software is given.After functional testing and performance testing,to ensure that the software achieves the user's various functional requirements,stability,scalability and response time have reached the development goals.At present,the project has been online,is in the user use stage,and is constantly updated and improved.The company combines the actual needs of users and cutting-edge technology to provide users with novel and user-friendly products.The performance of the system has been improved in many iterations.