| Under the current vigorous economic development trend,people’s living standards are gradually improving,which is accompanied by a continuous increase in the number of private cars.However,most communities currently lag behind in the management of parking spaces.In some communities,one car corresponds to one fixed parking space,which leads to inefficient use of parking spaces.Some parking spaces in the community can be used freely.Although this method improves the efficiency of parking spaces,the parking spaces in the community are not transparent,and the owner cannot quickly enter the community to find parking spaces.Obviously this is due to the backwardness of the community’s vehicle management technology,which makes the parking space information undisclosed.At the same time,it seriously affects the travel experience of the residents in the community.Therefore,the intelligent management of the community parking spaces has become a top priority.This paper studies the problem in depth and proposes a solution in combination with video recognition technology.It divides the community into several areas,monitors the vehicles at the intersection of each area,identifies the information of the vehicles entering and leaving the area,and updates the number of vehicles in each area in real time.Afterwards,this paper designs and implements a set of intelligent parking space management system based on video recognition for the community,which mainly includes the design of vehicle monitoring terminal and parking space management cloud platform.The monitoring terminal is used to collect and process video data,and upload the information to the cloud platform.The main function of the cloud platform is to disclose the parking space information of the community in real time to facilitate the query of the vehicle owner.Considering that vehicle detection may occasionally result in misjudgment,the community administrator can log in to the cloud platform to modify the number of vehicles in each area to correct the information.The monitoring terminal uses the Raspberry Pi 4B as the core processor,expands the CSI camera as a data acquisition device,and expands the 4G communication module Huawei ME909s-821 a to enable data upload outdoors.The parking space management cloud platform is based on the B/S architecture and is developed on the development tool Eclipse using Java combined with JSP technology.It mainly includes the preparation of the front and back ends of the Web,and the design of the database table structure.It mainly includes the preparation of the code including the front and back of the Web,and the design of the database table structure.At the same time,I purchased the Alibaba Cloud server and built the Tomcatserver environment and My SQL database,and deployed the cloud platform to the Alibaba Cloud server in order to connect the platform to the public network.The correct processing of the video data directly affects the accuracy of the number of parking spaces on the cloud platform,so the vehicle detection algorithm is very important for this system.This paper improves the Vi Be algorithm,proposes and implements a vehicle detection algorithm suitable for this system,solves the first frame ghost problem of the original Vi Be algorithm and eliminates shadows.This paper also analyzes the changing rule of the number of white pixels in the virtual detection zone,and proposes a vehicle count and movement direction discriminating algorithm,which has a relatively small amount of calculation and meets the real-time requirements of this system.Finally,these algorithms are combined to form the personalized vehicle detection algorithm required by this system,and transplanted to the monitoring terminal to run.In the end of this paper,the monitoring terminal is placed in the community for field testing.The results show that the system can basically meet the real-time disclosure of parking space information in various areas of the community,and realize the intelligent management of the parking space in the community. |
