Embedded Video Monitoring System Of The Fpga Image Processing Subsystem Design

With the continuous development of image processing technology,image processing technology have been widely used in the national economy and all aspects of social life.At the same time,the requirements of the image processing is increasing.Traditional digital image processing devices are mainly ASIC(Application Specific Integrated Circuit) and DSP (Digital Signal processors).Since the beginning of the 20th century,accompanied by the development of semiconductor technology,with its application of flexible,highly integrated, powerful,short design cycle,and low development costs,FPGA(Filed Programmable Gate Array)are increasingly being used in the field of image processing.More and more practice has proved that,FPGA's parallel processing capability and pipeline operations can dramatically improve the speed of image processing,the FPGA-based image processing system has broad prospects for development.In this paper,we researched a image pre-processing subsystem in an embedded video monitoring system.First of all,we realized a hardware platform for common re-configurable image processing algorithm which could complete image acquisition,receiving,processing, storage,output and other functions.FPGA itself entirely is a result of re-configurable,so the structure of the hardware platform can be easily upgraded and repeatly configurated.Second, in the platform,this article used the Verilog HDL hardware language to achieve a variety of image pre-processing algorithmin the FPGA chip.In the realization of the process,in order to give full play to the FPGA's strong power in parallel processing,we do a certain amount of improvement to make it possible to use parallel processing methods.The experimental results show that the image pre-processing system is able to complete a variety of high-level image algorithms in milliseconds,which completely meet the 50/s output requirements of the video surveillance system.Finally,according to a variety of types of noise arising from the practical application of video surveillance system,we designed a image processing control module based on feedback theory.The module can analyse the peak signal to noise ratio(PSNR) of the images handled,then control FPGA to using a more targeted approach to process the next image.