Research Of Image Data Fusion Technology Based On Raspberry Pi Platform

Image data fusion technology has been widely used in civil and military field,it can reduce the uncertainty of single sensor imaging.In recent years,multi focus image,infrared and visible images fusion became research hotspot,source image is processed by a certain algorithm,we can get clear fusion image with rich information.This paper mainly research image data fusion system,including software algorithm and hardware design.Hardware design consists of image data fusion system which based on Raspberry Pi and USB2.0-FPGA image data communication high speed transmission display system.The former mainly contain image acquisition and processing,image acquisition is completed by system camera module.The collected images are stored in Raspberry Pi SD card.The fusion algorithm is compiled in Raspberry Pi Linux environment,image after fusion process shows in PC and Yeelink platform.USB2.0-FPGA image data transmission system can show Raspberry Pi source image and fusion image in other display.This paper makes use of three improved image fusion algorithms aim at multi focus image,infrared and visible images.The first is based on traditional wavelet transform,the second is Contourlet transform which improved on the basis of wavelet,the third algorithm is based on NSCT and PCNN.The experiment result indicates the image which processed by third algorithm has better clarity and objective evaluate index.Finally,the whole fusion system has been debugged,verifying the system feasibility,images processed by the improved wavelet algorithm which using Opencv language compile,objective evaluate index,clarity and compute speed satisfy expected goal.Summarizing completed work,pointing out better algorithm and hardware can improve system performance.