Data Mining Analysis Of Traffic Congestion And Vehicle Behavior Based On Cars Of Internet

With the rapid development of society and the improvement of people’s living standards, cars are becoming more and more people’s transportation means, which have brought a great deal of convenience to people. But on the other hand, the increasing number of cars have put some pressure on transportation. Traffic congestion, traffic accidents and other traffic phenomena will not only affect people’s daily lives, but also hinder the development of transportation. Therefore, using information technology to help solve these problems is of great importance.Cars of internet is a branch of things of internet, and is an important part of intelligent transportation system. Cars of internet is a huge network consisting of vehicle position, drive route, vehicle speed and so on. Cars of internet based on software and achieved by computer image processing technology is becoming more and more popular. The implementation procedures of this kind of cars of internet include capturing video from camera on road, the recognition of license plate number and saving traffic information to the database.Vast amounts of traffic data saving by cars of internet contains great potential information, which has great practical significance. Using data mining technology to analysis massive traffic data scientifically, extract the knowledge behind the data, and applicate these knowledge to the cars of internet in actual life, will help make people, vehicle and road more harmonious.The main work is as follows:(1) Research the methods of license plate location from an image, license plate character segmentation and character recognition. Implement these processes about extracting license plate area from an image, segmenting the license plate character after the plate has corrected tilted and recognizing character based on multiple classifiers.(2) Study the traffic data from the cars of internet about Yan’an road north to south road in Hangzhou city, and implement traffic congestion forecast by using C4.5 decision tree algorithm.(3) Segment the trajectory in the mutant sample point, define the structure representation of trajectory data and the similarity of trajectory segmentation, and realize the clustering analysis of vehicle trajectory segmentation data by using DBSCAN algorithm.