ABSTRACT
III
ABSTRACT
As one of the main source of the information of human, image processing
technology played an import role in people's life. Currently, the image processing
technology has been widely used in military, industrial, and people's daily life. The main
platform of image processing are large-scale image processing workstation,
general-purpose computers and embedded system. Embedded image processing system
is compact, portable, flexible, and widely used in the occasion of the real-time image
processing. The core of embedded image processing system mainly contain ASIC,
FPGA and DSP. ASIC was high-speed and stable but poor flexibility, only suitable for
the application of the special purpose. So, studying the image processing technology
based on FPGA and DSP is very important.
The works in this paper are focus on the realization of image processing technology
based on FPGA and DSP , according to their own characteristics and existing
knowledge. The principal tasks are as follow:
(1) A IUT656 format video capture and VGA display system was designed. This
system take advantage of the parallel and pipelined characteristic of FPGA.
(2) A K-nearest neighbor classifier used for video object detection based on
FPGA and its Avalon bus interface were designed .It can be mounted on the Avalon bus
and as one part of the entire SOPC system.
(3) The structure of parallel SAR image denoising system based on TS201 was
determined. Data reception and transmission method according to the communication
protocol was designed. SAR image denoising algorithm based on the similarity measure
of statistics of the image block was transplanted and optimized on this system.
This work was supported by the National Natural Science Foundation of
China(No.60971128);Huwei innovation research project(No.IRP-2011-03-04).
Keywords: Image processing FPGA DSP
