Study Of The Technology Of The Automatic Vehicle Identification

Now much attention has also been paid onto the study of the technology of the automatic vehicle identification. And it is one of the important study in the intelligence traffic system. With its advantages, it seems to be an extremely useful method to solve many questions of transportation. With more fast development of transportation and the use of higher-tech in the automatic monitor and control, the AVI seems more important.In this thesis, we present an automatic vehicle identification system based on still images. From the following parts: location of the vehicle, car type identification, extraction of the license explained respective the critical technologies used in the system. We use MATLAB as tool to carry out the pre-procession of the original images. And a simplified threshold value algorithm was put forward. The location algorithm of number plates' sub-image based on the line-procession was also advanced. A synthesis network model on number plate recognition combined with lattice input network and feature input network was constructed, and learning algorithm of BP network was improved, which remarkably enhanced the recognizing correctness rate and recognizing speed.