Urban Rail Train Number Recognition System Based on Image Processing

Abstract

Aiming at solving problems such as easy to fall off damage, loss of vehicle number risk for RFID number recognition system, a type of number recognition system that integrates image acquisition and image processing is proposed. Firstly, the industrial camera is used to take pictures of the car number on the side of the urban rail train, then the techniques such as accelerated robust feature algorithm and neural network are used to locate, correct and segment the captured pictures, and finally the VGG-16 network model is used to recognize the segmented car number characteristics. The test results show that the system has good robustness, high recognition accuracy and fast speed, and can meet the requirements of urban rail train number acquisition.