Design And Implementation Of High Speed Train Passenger Intelligent Service System

In recent years,Chinas high-speed railway has been booming and ranks among the best in the world.People have become accustomed to traveling by high-speed railway.In order to improve the quality of service on the high-speed train,so that people can enjoy a modern,high-tech journey,this topic will be based on deep learning based computer vision technology and high-speed train combined,is committed to provide better scientific and technological experience for the crew and passengers.This topic is devoted to the baggage service,which is one of the passenger services.It is devoted to solving the problems of taking luggage by mistake and missing taking luggage on the train.In order to realize the intelligent luggage service,the subject adopts the pedestrian and baggage target detection algorithm based on Yolov3,the baggage classification algorithm based on Resnet18 and the pedestrian and baggage reidentification algorithm based on Resnet50 and Bnneck,and uses the Milvus search engine to store and query the feature vectors.In order to provide better services for users,the system also provides an interface to facilitate users to view real-time surveillance images,query luggage and pedestrian records,and manage camera equipment.This system uses the form of Docker container for application deployment,and uses Kubernets for container arrangement and management.In addition,the system uses mirror warehouse and code warehouse for development,construction and deployment,so as to provide sustainable code update and system iteration.After the functional test,the system can identify and record the passengers and baggage on the train,and search for them.At the same time,it provides good expansion performance.To a certain extent,it can meet the intelligent demand for baggage service on high-speed trains,which has been recognized by customers.