基于光照模型的低照度图像增强  

A low-light image enhancement method based on lighting model

作　　者：张学乾 靳伍银[1] ZHANG Xueqian;JIN Wuyin(School of mechanical and electrical engineering,Lanzhou university of technology,Lanzhou 730050,China)

机构地区：[1]兰州理工大学机电工程学院,兰州730050

出　　处：《激光杂志》2023年第8期65-73,共9页Laser Journal

基　　金：国家自然科学基金项目(No.12062009);甘肃省高等学校产业支撑计划项目(No.2022CYZC-24)。

摘　　要：为了增强低照度图像,降低不均匀光照的影响,根据光照反射模型和图像融合理论,提出了一种基于物理模型和非线性函数的图像增强方法。首先,将图像从RGB转换为HSI颜色空间,利用多角度Gabor函数提取I分量的反射分量,然后根据韦伯-费希纳定律和对数缩放函数构造亮度增强模型增强反射分量。最后,采用一种图像融合策略,融合不同参数增强后的反射分量。实验结果表明,本方法产生的图像具有更高的可见性和更好的视觉质量,在方差、信息熵、平均梯度几个客观评价指标方面也优于目前的低照度图像增强方法。In order to enhance low-illumination images and reduce the influence of uneven illumination,an image enhancement method based on physical model and nonlinear function is proposed according to the light reflection model and image fusion theory.Firstly,the image is first converted from RGB to HSI color space,the reflection component of the I component is extracted by using the multi-angle Gabor function,and then the brightness enhancement model is constructed according to the Weber-Fichner law and the logarithmic scaling function to enhance the reflection compo-nent.Finally,an image fusion strategy is adopted to fuse the reflected components enhanced by different parameters.The experimental results show that the images generated by the proposed method have higher visibility and better visual quality,and are superior to the current low-illumination image enhancement methods in terms of objective evaluation indicators such as variance,information entropy and average gradient.

关 键 词：图像增强 GABOR函数 图像融合 低照度 

分 类 号：TN209[电子电信—物理电子学]