Research Of Image Processing Algorithms And Hardware Design Based On FPGA

As the image size becoming larger and larger, image processing with software can't meet with the real-time demand of image processing system; with the great development of the programmable chip, such as FPGAs, image-processing with hardware becomes feasible. Currently, the image-processing based on FPGA hardware is a hot research field.This thesis built a reconfigurable architecture that consisted of image-preprocessing and communication software on PC, control automation, computation block, RAM and communication adapter for image-processing research based on FPGA hardware using verilogHDL language. The computation block, which was implemented independently for every algorithm, took charge of the computation of the specified algorithm. A portable interface in the reconfigurable architecture was designed for the computation block, with which any computation unit of any algorithm that was compatible with the interface could be added in or removed out of the architecture for debugging or running.Based on the architecture, several typical image processing algorithms, which included order filter, median filter, convolution and Gauss filter, morphological operator etc., were implemented in verilogHDL. Then the FPGA-based image processing methods and the optimized strategies for the hardware-design algorithm were discussed. With the performance analyzing, it was found that the time-consuming of hardware processing was greatly reduced than that of software processing; comparison between hardware and software processing showed that both of the results were identical. In the last section, techniques for large image processing and arbitrary-size window design were developed, and their optimization strategies were shown.The full designs of all modules were synthesized and simulated on xilinx ISE8.2 and ModelSim, and were implemented on the xilinx Spartan-3E XC3S500 hardware platform.This thesis investigated the design technologies of specific chip for image processing using FPGA and provided preliminary study for the real-time image processing system based on FPGA.