License plate recognition system based on image recognition

As people’s living standards rise, efficient and intelligent traffic management system has become the demand of the society. Although there are many researches on vehicle license plate recognition technology in China, there are still many shortcomings in practical application. To solve the above problems, this paper uses OpenCV machine vision library based on Python development platform and convolutional neural network to study the process of license plate recognition, and proposes a vehicle license plate recognition system based on machine vision technology and pattern recognition method. System using digital image correlation algorithm to detect and locate license plate, based on the analysis to extract the edge of the information, character segmentation and character quantity calculation, at last, we use Convolutional Neural network (CNN) for character recognition and classification in order to obtain accurate and clear plates in formation. At last, we use convolution neural network (CNN) for character recognition and classification in order to obtain accurate and clear plates information, its accuracy is as high as 98%, has the good recognition effect.