Research On Infrared And Visual Image Fusion Method

The infrared image can distinguish the target from the background according to the level of thermal radiation.It is not affected by lighting conditions,and its imaging effect is very good under all weather conditions.We can clearly see the hot target from the infrared image.But it has the problems of low resolution,low contrast,and blurred image details.The visible image is consistent with the human visual system,providing texture details with high spatial resolution and clarity.However,it is limited by lighting and requires good lighting conditions to have a good display effect.Therefore,by fusing these two types of images,the resulting image contains both the thermal radiation information in the infrared image and the detailed texture information in the visible image.This thesis proposes a new infrared and visible image fusion method based on total variation.In terms of gray-scale image fusion,this method has a better fusion effect than other methods.And a better method of color image fusion was found through the conversion of color space.The main research contents are as follows: The thesis begins with an introduction to the background,significance,and domestic and foreign research on image fusion.Secondly,it introduces the relevant knowledge of image fusion,including the characteristics of infrared and visible images,the characteristics and classification of image fusion,the subjective and objective evaluation criteria for fusion images,and different color spaces.Then,some commonly used image fusion methods at the pixel level are introduced.And compare them with the image fusion method based on global variation proposed in this thesis,and compare their fusion images with gray image fusion method.Finally,in different color spaces,the infrared and color visible images are fused,and the effect of color image fusion is compared.At the end,a color image quality improvement method is proposed to improve the quality of color fusion images.