Design Of Large Area Smart Fusion Camera Based On ZYNQ

With the application of remote sensing technology in all fields of people's life more and more widely,people pay more and more attention to the development of remote sensing technology.As the most important part of remote sensing technology,the acquisition and processing of remote sensing image has always been the focus of people's research.Aiming at remote sensing imaging technology,the thesis design a large array imaging system for remote sensing imaging.Besides,to solve the problem that high resolution remote sensing image has great difficulty in storage and processing,the thesis study the fusion algorithm based on compressed sensing,and complete the Algorithm transplantation in the camera system,the main work is listed as follows:Design of remote sensing imaging system,the paper chooses the large area array sensor CMV20000 as the image acquisition source.The resolution of this sensor is5120*3840,and the frame rate is up to 30 frames per second.It is highly application valuable in the field of video surveillance and satellite remote sensing.The large array can satisfy the requirement of remote sensing image fusion;and XC7Z020 of Xilinx Zynq-7000 series is chosen to migrate algorithm and design the sensor driver.This article uses the FPGA programmable logic part to achieve the front-end CMV20000 driver design,and adopt ARM to achieve algorithm.The hardware design successfully implemented the front-end driver of the CMV20000 to 20 frames per second,and complete the acquisition,correction,caching,and output of images.Finally the image is displayed through the HDMI interface with a resolution of 1280*720.Design of remote sensing image fusion algorithm based on compressed sensing.Based on the fusion of high-resolution panchromatic images(PAN)and low-resolution multispectral images(MS),the paper proposed a algorithm : a novel remote sensing image fusion scheme based on NSCT and Compressed Sensing.Compared with the traditional Compressed Sensing algorithm,this algorithm can achieve a more effective fusion with these evaluation parameters:AG?IE?MI?AB/F and CC.Finally,this paper achieve the algorithms on the self-built hardware platform,and performs algorithm successfully on the platform,and display the result via HDMI.