Research On Fusion System Of Infrared And Visible Light Image Based On FPGA

In the fields of military,security and maritime rescue,the most important function is notthe higher imaging quality,but the ability to work in the equipment of all-weather.Obviously the single imaging mechanism of equipment is unable to meet the requirements.By configuring different sensors,such as different imaging mechanism of infrared image and visible light image,using the complem entarity of the details and the method of image fusion,we can makes equipment has the ability to work in the equipment of all-weather.By combining the infrared image and visible light image,the fusion image can not only realize the particular scene or track targets accurately described,but also do further image processing.The precision of registration in the whole pretreatment procedure has a great influence on the fusion effect.Based on the basic theory of image registration and image fusion,this article focuses on advances the infrared and visible light image fusion system which based on FPGA.By using the information from infrared and visible sensors,and considering the ability of FPGA,the constraint factors,this system improve the registration and fusion algorithm,achieve real-time processing to the image realize the real-time processing on FPGA system,meet the needs of the practical application of engineering.In this paper,feature points detection and matching are realized based on corner detection.First the two sensors using the Harris corner detection to extract the image feature pointsat the same time.Second the FPGA build the relationship between the two figure matrix,optimized the relationship between the infrared image and the visible light image.Then constructing the relation matrix of two graphs and optimizing the regional mapping relationship between infrared and visible images.Next according to the mapping relationship we can construct the initial point set for matching.Finally,using random sample consensus algorithm to obtain the optimal parameters and determine the error evaluation parameters.At the same time,combined with the high speed data throughput and parallel computing capacity of FPGA,the algorithm was improved to realize data pipeline data processing.Experimental results show that the system design based on the relevant data of two way sensor and system resource,combining with the image registration,with the method of weighted average,realize the real time and fusion of infrared and visible light output,in principle on the steps of image registration and fusion,has obtained certain achievement.