一种基于双边滤波和图像融合的增强方法

摘要

在系统研究可见光和红外图像融合技术的基础上，提出一种图像融合算法用以检测可见光图像中的隐含目标。为了在检测隐含目标的同时平滑图像中目标的边缘信息，采用双边滤波来去除噪声并在此基础上进行图像融合以获取图像完整信息。首先通过对图像添加噪声得到降质退化图像，然后采取双边滤波算法对降质图像进行平滑去噪，此后，再对滤波去噪后图像进行融合。结合双边滤波和图像融合方法，设计四种算法来验证双边滤波和图像融合在多源图像应用中的强大功能。研究测试了算法对可见光图像中因光线原因隐藏的人体目标检测性能。实验结果表明，该方法可以有效提高在背景图像中进行目标检测或识别的概率。%Thispaperproposesanalgorithmofimagefusionfordetectingthehiddentargetsinthe visible light images.In order to smooth the edge of the image target while detecting the hidden image,the bilateral filer is used to remove the noise and the image fusion is performed accordingly to get its integrated information.The edge of the image is degraded by the noise,and the typical bilateral filtering method is used to save the image and remove the noise,then bilateral filtering fusion is performed for the noise removed image.Combined with the two kinds of processing technology,four algorithms are used to verify the strong functions of the bilateral filtering and image fusion in the application of multi -source images.The algorithms,for detection performance for the body target hidden by the different factors,are tested.The experiments show that the algorithm can effectively improve the probability of detection or identification of the target from images background.