Research And Implementation Of Embedded Platform Of License Plate Recognition System

With the rapid development of China’s economy, more and more cars on the city’s road, which gives human travel great convenience, but also made more challenging. Intelligent transportation systems as a means of transport to ease this pressure, it becomes more important to develop their technical research. License Plate Recognition System is an important aspect of intelligent transportation system, has great significance and wide application background. License plate recognition system based on DSP, FPGA, PC machine is relatively common, but its power consumption, large size, high cost, and low stability determine that we need to design a more convenient, more flexible license plate recognition system. This article is designed under the embedded platform, eventually complete this automatic license plate recognition system by acquiring image, processing image, recognizing image.This thesis began as the research of domestic and international background of the license plate recognition system and instructions of the significance of the embedded platform license plate recognition system. Software aspects of this system is mainly using Linux as a development and runtime systems, cutting through the Linux kernel, root file system build, Qt interface library transplant, Opencv image processing library transplant for constructing a set of embedded image processing development and operation of the system application environment, and by using Qt Creator interface development tools to develop and debug, eventual complete the application software development of the license plate recognition system. Hardware aspect of this system, the main choice is embedded hardware development board S3C6410 as the core processor, start the program by loading Bootloader, set up NFS start, remote start building the Linux root file system, after the adoption of cross compiler License plate recognition system application, open USB camera development board, get the license plate image through the camera, and on this embedded platform to identify the license plate information.The system algorithms for image processing applications in ARM such as: image preprocessing, license plate location, character segmentation, character recognition and other parts content. But the image preprocessing algorithm needs further improvements, such as adding tilt correction, but has met the basic functions of License Plate Recognition, small size and high stability of system requirements. Finally, this system is also in need of improvement to make additions and prospects.