Design And Implementation Of Zedboard-based Face Tracking System

With the constant improvement of the circuit integration and computational complexity, the traditional SOC, which is suit for single serial or parallel computing, cannot satisfy the need of solving complex problems. The development tend of FPGA is aiming to achieve high efficiency and low power consumption. System on programmable chip, which is developed on the basis of programmable logic devices, is a flexible and highly efficient embedded system design solution. The main characteristic is the built-in processors within the traditional FPGA, such as Xilinx company’s high performance ARM Cortex-A9 hardcore based solution, the NIOS II based on soft core solution of Altera company. These could realize the run-time configuration and system reconfigurability. The embedded system using these soft or hard IP core solution owns the characteristics of both serial and parallel processing, which makes it more flexible to deal with complex problems. It can avoid the repeated work, make complex modified the function of the basic components easy, greatly reducing the development cost, and shortening the time to market of the product. Therefore, with the advantages of SOPC, the defects of the current embedded system are analyzed in detail in this paper. On the basis of the hardware and software collaboration design methodology, this paper designs and achieves the two dimension(up, down, left and right) face intelligent recognition embedded monitoring system based on Xilinx company’s Zedboard development board.Study the development of the programmable logic device in the integrated circuit. As an example, the advantages of Xilinx company’s Zynq series SOPC and SOPC-based embedded system design are introduced in detail. Then introduce the hardware-software collaboration design methodology and design process, study the structure and framework of Xilinx company Zynq series and the embedded system development based on this series of Zedboard board. Finally, based on the hardware and software collaboration design methodology, the design of human face tracking system is realized. Main research results of this article are as follows:First of all, summarize SOC and SOPC, analyze their current situation of development,study their respective advantages and disadvantages, describe faults of the early embedded system, discuss advantages of the embedded system based on SOPC devices and significance of intelligent monitoring technology;Secondly, study the characteristics and the development process of hardware and software collaboration technology, the disadvantage of traditional pure software and hardware design is also studied, introduce the hardware and software collaborative development tools provided by Xilinx company, then take Xilinx company Zedboard as an example to study its structure and interface bus in detail;Thirdly, study the development of Zedboard embedded system, put forward the monitoring and control system with functions of face recognition and tracking, build Linux system on Zedboard, introduce and transplant the Qt and Open CV, design and develop user-defined IP core, study face detection algorithm, PCA recognition algorithm, and the PID control algorithm.