Research On Image-Fusion Algorithm

The image-fusion becomes more and more popular in the field of the image-disposing.By the image-fusion,more informations can be integrated and we can improve the observation on the image.So this thesis mainly researches algorithms of image fusion from four aspects.In order to appraise the effect of the image-infusion objectively and efficiently,researing the methods which can be applied in the condition of having the standard image and in the condition of not having the standard image.when the standard image can be gained, adopting the methods based on structural similarity,when the standard image can not be gained,the thesis adopts the method based on the correlative coefficient of the high frequency heft by the Curvelet transform.Aiming at the problems existing in image fusion methods based on wavelet transform, an algorithm of image fusion based on Curvelet transform is proposed. This transform was developed from the wavelet transform and it has overcome some inherent limitations of wavelet in representing directions of edges in image.When the Curvelet transform is introduced to image fusion,the characteristics of original images are taken better and more information for fusion is obtained.In disposing the high frequency coefficients by Curvelet transform,researching the local variance and contrast based on Curvelet transform, and integrating the two algorithms,then the contrast algorithm of local variance is gained,this algorithm can lead to the better effect.Aiming at the characteristics of infrared and visible light images fusion, this paper adopts an image fusion method based on image-segmentation. By segmenting the infrared image to obtain object,the method fuses the infrared object with visible image to make full use of the source images. the Otsu for the image-segmentation and the recursion segmentation based on Otsu have been researched .The experiments in this thesis prove that the algorithms and methods mentioned in this paper can effectively improve the effect of image fusion,and can be valuable in the engineering application.