Research on Vehicle Type Recognition based on Datasurveying

In the field of transportation, many of the vehicle amount of raw data value which can be accompanied by a large number of unknown information, and these information can provide people with more help. This paper aimed at the master data, and by using the method of survey of data modeling and prediction. First this paper introduces the data survey methods and the characteristics of the model, and then applied to the vehicles recognition judgment in their ability to recognize and to compare the analysis. The conclusion by principal component analysis and discriminant analysis method bus and van in four categories, the two on the training set and testing set of identification is quite high (high purity), so the four classified into two categories, the key to deal with opel (has been) and class and Po (saab) class. After the processing of raw data, and finally compares the model results.