Research And Implementation Of Infrared And Visible Image Fusion Algorithm

In this paper,the fusion of infrared light and visible light is studied,mainly from the aspects of related theories,algorithms that can be used and FPGA-based image fusion system design.Based on the overall architecture of image fusion system,this paper describes the imaging principle of infrared and visible light detectors,determines their imaging characteristics,and according to the different image imaging rules of the two,performs pre-processing before organic combination.There are also differences.Before the organic fusion of visible light and infrared light,the pre-processing process used includes image enhancement,image calibration,noise elimination,and registration.Both the subjective and the objective evaluation and estimation of the processed image are carried out respectively.The focus of the evaluation is mainly on the objective aspect.In the process of objective evaluation,the main methods involved include spatial frequency,average gradient,and edge retention.And information entropy.According to the defect that the current infrared and visible light image fusion is not clear in the target information,this paper proposes an infrared and visible light image fusion algorithm based on infrared target extraction.After improving the LOG operator for edge detection,combined with the Contourlet transform The method provides a special rule for the entire organic binding process.According to the results of the relevant experiments,it can be seen that the images formed by the organic combination method are subjected to the operations of complementing each other.Therefore,the obtained images are better than those obtained by the previous methods,the objects are clearer,and the scene information is more comprehensive.Easy to observe by the human eye.The rate of image fusion processing has been significantly improved and it is easy to implement on FPGAs.In the process of designing an FPGA,a preliminary design is performed in a modular manner,and then software simulation and hardware debugging are performed for each module.The first step is to simulate each module separately to determine its basic functions.The second step is to debug each module unit through hardware.The third part is to consider all the modules as a whole,and to make them overall.The data obtained after the experiment was analyzed and evaluated.In summary,the research program of this paper is feasible to complete the design requirements,after further experiments and optimized engineering applications,will help the development of image fusion technology.