Parking Space Query System Of Intelligent Parking Lot Based On Deep Learning

With the rapid development of China’s industrial technology,private cars have gradually become a necessity for every household.However,the contradiction between the increasing number of vehicles and the limited parking space becomes more and more obvious.In most major cities,parking has become the biggest problem for car owners.The main reason for this problem is not only the unbalanced proportion of parking spaces to the number of private cars,but also the management of parking lots.For the vast majority of parking lots at present,the conventional parking management model is still adopted.The parking information can not be conveyed to private car owners in advance.It not only wastes a lot of time in finding parking spaces,but also causes traffic congestion at the entrances and exits of parking lots in some busy commercial areas,thus affecting the traffic of the whole city.Therefore,parking problems will have a great impact on both individuals and society.In order to handle this problem,this paper will propose a method of parking space query in Intelligent Parking Lot System based on Deep Learning.In terms of the shortcomings of the current parking space management mode,the reform will be carried out so that the owners can view the parking space information of a parking lot in real-time through the Wechat public platform on his mobile phone,which reduces the time for finding parking spaces,improve the travel efficiency,and even temper the problem of traffic congestion in a certain extent.This subject is designed in three main contents as follow:(1)The core of this design is to use the built YOLO object detection network model on LINUX platform to real-time process the monitoring video picture of the parking lot,and then though the designed algorithm to conclude whether the parking space is available,at the same time,recognize the license plate of the vehicle on the parking space and calculate the time that the parking space is occupied by the vehicle.(2)Design a corresponding map of parking space according to the parking space position division in the parking lot.According to the parking space status,fill the corresponding color in the corresponding parking space in the map,and update the parking space information map in real time according to the processing results of monitoring video.(3)Connect the parking information with the designed Wechat mini program,the driverscan obtain the parking information in real-time by using the Wechat mini program.After several trials and tests,the design of this subject can normally complete the functions of parking space display,licence plate recognition,parking timing and the real-time display of parking information function on Wechat public platform.What’s more,the system is stable and reliable,basically realizing the functions of this design that is expected.