Infrared image detail enhancement approach based on improved joint bilateral filter
Infrared image detail enhancement approach based on improved joint bilateral filter
复制标题
基于改进联合双边滤波器的红外图像细节增强方法
DOI：
10.1016/j.infrared.2016.06.017
复制
发表时间：
2016-07-01
影响因子：
3.3
通讯作者：
Chen, Xiaohong
中科院分区：
文献类型：
--
作者：
Liu, Ning;Chen, Xiaohong
提出了一种新的红外图像细节增强方法。该方法不仅可以达到增强数字图像细节的目的，而且可以使处理后的图像更接近于真实的情况。受联合双边滤波器的启发，利用相邻两幅图像计算核函数，以区分原始图像中的细节信息。我们还设计了一种新的核函数来改进联合双边滤波器，消除非线性滤波所带来的梯度反转伪影。新核函数基于自适应浮现系数实现细节层的确定。利用自适应浮现系数沿着和两个关键参数对图像细节信息进行修正，实现图像细节增强。最后，我们将处理后的细节层与基本层结合，将高动态图像重新排列成适合监视器的低动态范围，以获得更好的视觉效果。数值计算表明，该技术在细节增强方面具有较好的应用价值。文中给出了数据流程图。(C)2016爱思唯尔B.V.保留所有权利。
In this paper, we proposed a new infrared image detail enhancement approach. This approach could not only achieve the goal of enhancing the digital detail, but also make the processed image much closer to the real situation. Inspired by the joint-bilateral filter, two adjacent images were utilized to calculate the kernel functions in order to distinguish the detail information from the raw image. We also designed a new kernel function to modify the joint-bilateral filter and to eliminate the gradient reversal artifacts caused by the non-linear filtering. The new kernel is based on an adaptive emerge coefficient to realize the detail layer determination. The detail information was modified by the adaptive emerge coefficient along with two key parameters to realize the detail enhancement. Finally, we combined the processed detail layer with the base layer and rearrange the high dynamic image into monitor-suited low dynamic range to achieve better visual effect. Numerical calculation showed that this new technology has the best value compare to the previous research in detail enhancement. Figures and data flowcharts were demonstrated in the paper. (C) 2016 Elsevier B.V. All rights reserved.