Image Acquisition Arm9-based Container And Container Number Recognition System Design,

With the rapid development of information technology, control technology, semiconductor technology and network technology, networked embedded system with high performance, real time and low power consumption is becoming a hot research field. Because of its high performance and versatility, embedded system has developed quickly. For example, the Internet of Things system which is considered to be the future development direction has strong links with the development of embedded system.Generally, the study and application of the embedded system both in now or future will have important value and practical significance.The intention of this system's design is to improve a company's image acquisition system structure of the container and realize the container image collection by introducing embedded image acquisition technology based on ARM9. The first work of this paper is put forward the solutions and prove that the embedded system owns better performance than the original one. Then, complete the construction of the developmental environment such as hardware and software, the communication, the transplantation of the bootloader and the transplantation of the linux operating system. After that, this paper analyzing the driver design method of the embedded system and realize the USB camera driving by the GSPCA driver. Then, it describes the development and design of the embedded application based on the linux operating system.The third work of this paper is develop the embedded image acquisition system of the container and develop the image network transmission system of the container by the socket programming technology.The recognization of the container number is the other critical part of the whole system. This paper applies the projection of the edge detection method to discuss the preliminary positioning of the container character and the accurate positioning combined with the prior knowledge such as the gap between the container character and the arrangement characteristics. After realize the image segmentation, double BP neural network is adopted to realize the number and character recognition of the container. Finally, this paper proves the feasibility of the embedded system used in the image acquisition and done a useful research work of the number recognition system of the container.