基于自适应分数阶微分的红外目标增强算法

Linear Enhancement Algorithm of Infrared Target Based on Adaptive Fractional Differentiation

针对红外图像存在灰度范围窄、图像细节不清晰、目标边缘模糊的问题,提出了一种基于自适应分数阶微分的红外目标增强方法。该方法首先利用图像的梯度、信息熵进行有效融合,并且自适应调整分数阶微分以增强图像中的目标边缘;然后采用图像像素灰度的标准差和均值进行融合去确定目标的分割阈值,以区分出图像中的背景和目标部分;通过对图像中的目标区域进行线性增强,以进一步突显目标。经过实验验证:本文提出的方法能够有效地区分红外图像中的目标和背景,局部目标背景比(Target-to-Background Ratio,TBR)平均提高了0.5,视觉效果比较理想。

To solve the problems associated with infrared images,such as narrow gray range,unclear image details and fuzzy target edge,an infrared target enhancement method based on adaptive fractional differentiation is proposed.In this method,first,the gradient and information entropy of image are used for effective fusion,and the fractional differentiation is adaptively adjusted to enhance the edge of the target in the image.Subsequently,the standard deviation and mean value of the image pixel gray are fused to determine the segmentation threshold of the target,to distinguish the background and target in the image.The target area of the image is linearly enhanced to better highlight the target.Experimental results show that the proposed method can effectively distinguish the target and background in the infrared image.The average local target-to-background ratio(TBR)increased by 0.5,and the visual effect was ideal.