Design And Implementation Of Large Truck Recognition System Based On Artificial Intelligence

With the rapid development of social economy,more and more convenient urban traffic environment has improved our travel.However,accompanied by the problem of traffic safety can not be ignored.It is not difficult to find that the hidden danger index of large trucks is very high.If the intelligent identification of large trucks can be realized,the intelligent transportation system will serve people more quickly and conveniently.This paper mainly studies the machine learning theory and deep learning theory of artificial intelligence,designs and implements a large-scale truck recognition system based on artificial intelligence.The system applies machine learning method and deep learning method to large-scale truck recognition,and realizes the innovation of intelligent transportation application.The system adopts HOG+SVM method in machine learning and VGGNet-16 model in deep learning,respectively.The two methods are used to identify large trucks.At the same time,the two algorithms are optimized.Tests show that the recognition efficiency of VGGNet-16 using deep learning model is higher.OpenCV+Python+TensorFlow tool is used in the development process.The workflow of the system is as follows: firstly,the region selection is carried out,and the monitoring video of each intersection in the selected area is converted into image in real time through OpenCV tool.Secondly,two recognition models with recognition ability are established to identify them separately.If the traffic of large trucks is identified during the prohibited period in the urban area,the alarm will be triggered.MySQL is used in the database of the system.It stores the information of past vehicles and counts the number of large trucks appearing in each period.This information helps users to analyze the rules of truck travel and prepare for the later system upgrade.