Design Of Image Processing Hardware Platform Based On DSP And FPGA

With the development of technology, a variety of image processing hardware platforms have been developed.Among them, the most commonly way is combining DSP with FPGA for image processing. The digital processing hardware platform based on DSP and FPGA has the flexible architecture and highly versatility. It is very suitable for modular design and improves data processing speed. In the hardware platform, the FPGA processor is used for processing the relatively simple underlying algorithms as well as system management and logic control. The DSP processor, with the feature of fast computing speed, flexible addressing modes and the powerful communication mechanism, is used for processing the more complex core algorithms. The two parts can complement each other.In this thesis, the framework of DSP and FPGA is applied to design the digital image processing hardware platform. TI’s 8-core DSP TMS320C6678 and ALTERA’s Stratix IV GX FPGA EP4SGX230KF40 are introduced in the platform. The platform is composed of image acquisition, processing, storage, and display functions etc. FPGA is responsible for image acquisition and logic control, while DSP is responsible for complex image algorithms. Communication between DSP and FPGA is implemented through shared RAM and Serial Rapid I/O interface.The thesis firstly introduces the background and development of image processing hardware platforms, significance of the topic and the main work. Secondly, it describes the overall system design as well as functions and features of each module’ hardware circuits in detail. Thirdly, it discusses the system hardware debugging, system initialization, peripherals testing, multicore image processing software design and then gives the test results. Finally, the thesis presents a summary of the author’s work and future prospects.