基于多尺度加权引导滤波的红外图像增强方法

现有的红外图像存在细节模糊、边缘和纹理不清晰的问题。针对上述问题，本文提出一种基于加权引导滤波的红外图像增强方法。首先，将图像通过带转向核的多尺度加权引导滤波进行分层处理，得到多幅含有细节信息的细节层图像和基础层图像；接着，对细节层采用基于Markov-Possion的最大后验概率算法和Gamma校正算法对细节层进行增强；然后，对基础层采用限制对比度的自适应直方图均衡算法进行对比度拉伸，最后，进行线性融合得到增强后的图像。综合主、客观实验结果，得出本文方法具有良好的细节增强效果，处理后的图像边缘和纹理信息比较突出，且算法在信息熵（IE），熵增强（EME）和平均梯度（AG）3个指标都有较优的计算结果。基本满足红外图像细节得到增强，边缘纹理清晰的需求。

Infrared Image Enhancement Method Based on Multiscale Weighted Guided Filtering

Existing infrared images have problems on fuzzy details, unclear edges, and texture. This paper proposed an infrared image enhancement method based on weighted guided filtering to solve these problems. First, multiscale weighted guided filtering with steering kernel layered the input images. It obtained detailed layer images and a base layer image. Subsequently, maximum posterior probability algorithm based on Markov-Poisson and Gamma correction algorithms enhanced the detailed layer images. Adaptive histogram equalization algorithm with limited contrast stretched the contrast of the base layer image. Finally, enhanced images were obtained through linear fusion. The subjective and objective experimental results show that the proposed method had good detail enhancement effects, and the edges and texture information of the processed images were relatively prominent. The proposed method had better calculation results for information entropy(IE), entropy enhancement(EME), and mean gradient (AG). It satisfies the requirements for enhanced infrared images and clear edge textures.