Infrared And Visible Image Fusion Technology Research

The aim of image fusion is to integrate complementary and redundant information from multiple images to create a composite that contains a 'better' description of the scene than any of the individual source images. Image fusion plays important roles in many different fields such as remote sensing, defense system, computer vision, robot and biomedical imaging, etc. This thesis mostly focuses on the research of algorithms and applications of the infrared and visible image fusion.Affine transformation is used for real-time pixel-level image registration and can be programmed in DSP in the style of look up table. Besides, an adaptive temporal recursive filtering algorithm is used to solve the problems about noise and visual range in low light level imaging. The experiment shows that this algorithm can not only maintain a clear image but also effectively eliminate the noise and extend the visual range of low-light imaging system.Based on structural similarity index for image fusion quality measure, a multimodal fused image metric based on complex wavelet structural similarity is presented. The advantages of our measure are that it does not require a reference image and fits with human visual perception.A fusion algorithm for infrared and visible images based on region segmentation and the dual-tree complex wavelet transform. Segmentation is firstly performed on the infrared image and visible image respectively with top-bottom-hat filtering and the threshold method. After image segmentation, the DT-CWT coefficients from different regions are merged separately. Finally the fused image is obtained by performing inverse DT-CWT. The evaluation results show the validity of the presented algorithm.In our color fusion method, segmentation is firstly performed on the infrared image and visible image. A chance of region by region color transferring is given to ensure the fusion image similar to natural images. Experimental results evidence this method which could provide better sense of hierarchy than the global color fusion method in terms of subjective evaluation.