摘要
为了实现夜视场景增强,获得更适合人类视觉感知的图像,提出一种基于对比度增强和两尺度分解的红外与可见光图像融合方法。首先利用引导滤波与动态范围压缩算法,对可见光图像进行自适应增强;为提高算法的运行效率,采用图像两尺度分解,将红外图像与增强后的可见光图像分解为基本层和细节层,同时引入视觉显著性检测以生成权重图,对细节层图像进行融合,图像基本层融合采用加权平均融合方法;最后,对各层融合图像进行重构,以获得最终的融合图像。实验结果表明,本方法不仅能有效突出红外目标信息,同时保留图像的纹理细节信息,有效提升图像的可视性,在视觉质量和客观评价方面优于其他目前常用的图像融合方法,同时大幅降低了处理时间,提高了算法的运行效率。
In order to enhance night vision scene and obtain images more suitable for human visual perception,a fusion method of infrared and visible images based on contrast enhancement and two-scale decomposition is proposed.Firstly,guided filtering and dynamic range compression algorithm are used to enhance the visible image adaptively.In order to improve the efficiency of the algorithm,the two-scale image decomposition is used to decompose the infrared image and enhanced visible image into basic layer and detail layer,and visual saliency detection is introduced to generate weight map,and the detail layer image is fused,while the weighted average fusion method is used in the basic layer image fusion.Finally,the fusion image of each layer is reconstructed to obtain the final fusion image.The experimental results indicate that the proposed method can not only effectively highlight the infrared target information,but also retain the texture details of the image,effectively improve the visibility of the image.It is superior to other commonly used image fusion methods in terms of visual quality and objective evaluation,and at the same time,it greatly reduces the processing time and improves the operating efficiency of the algorithm.
作者
罗谨哲
荣传振
贾永兴
杨宇
LUO Jin-zhe;RONG Chuan-zhen;JIA Yong-xing;YANG Yu(College of Communications Engineering,Army Engineering University of PLA,Nanjing Jiangsu 210007,China)
出处
《通信技术》
2019年第8期1871-1877,共7页
Communications Technology
关键词
可见光图像增强
两尺度分解
视觉显著性检测
图像融合
visible image enhancement
two-scale decomposition
visual saliency detection
image fusion
