Hardware Design And Algorithm Realization On Recongnizing The Goal Of Ships Based On YHFT Multi-core DSP

With the development of the satellite monitoring,how to quickly recognize target become a key technology.In order to achieve the satellite inner parts localization and higher performance,the issue will use the high-performance DSP chip which developed by Institute of Microelectronics in National Defense Scientific and Technical University as the processing core and the FPGA control the system.We build the hardware platform for image acquisition and recognition.We will research feasible algorithms of recognizing ship goals on the self-designed platform.The main work of this paper are:(1)Designing the framework based on multi-core DSP + FPGA.The framework leverages the DSP’s digital signal processing capabilities and FPGA’s control interface capability,flexible structure,and meet small size,low power consumption,high processing speed of the design requirements.(2)Completing hardware circuit design of the whole platform.(3)Completing commissioning of each functional module.FPGA enables image acquisition and display,extract from the information collected in the effective image data and grayscale image processing and data transmission with DSP and communications.DSP can communicate with the host computer.(4)Using gray threshold algorithm and effective edge information algorithm to identify the target ship,then find out the relative position of the object in the image and the results are displayed on the display device.(5)According to FT multicore DSP’s features,we make multicore optimization to improve the efficiency of the program.The hardware platform enables image acquisition and processing,ship targets identified and the results in the general case on a display device.