Lecca Michela
J Opt Soc Am A Opt Image Sci Vis. 2020 May 1;37(5):849-858. doi: 10.1364/JOSAA.384197.
Milano Retinexes are spatial color algorithms grounded on the Retinex theory and widely applied to enhance the visual content of real-world color images. In this framework, they process the color channels of the input image independently and re-scale channel by channel the intensity of each pixel by the so-called local reference white, i.e., a strictly positive value, computed by reworking a set of features sampled around . The neighborhood of to be sampled, its sampling, the features to be processed, as well as the mathematical model for the computation of the local reference white vary from algorithm to algorithm, determining different levels of enhancement. Based on the analysis of a group of Milano Retinexes, this work proves that the Milano Retinex local reference whites can be expressed by a generalized equation whose parameters model specific aspects of the Milano Retinex spatial color processing. In particular, tuning these parameters leads to different Milano Retinex implementations. This study contributes to a better understanding of the similarities and differences among the members of the Milano Retinex family, and provides new taxonomic schemes of them based on their own mathematical properties.
米兰视网膜算法是基于视网膜理论的空间颜色算法，广泛应用于增强真实世界彩色图像的视觉内容。在此框架下，它们独立处理输入图像的颜色通道，并通过所谓的局部参考白色逐通道重新缩放每个像素的强度，即通过对在周围采样的一组特征进行处理来计算出的一个严格为正的值。要采样的邻域、其采样方式、要处理的特征以及用于计算局部参考白色的数学模型因算法而异，从而决定了不同的增强级别。基于对一组米兰视网膜算法的分析，这项工作证明了米兰视网膜局部参考白色可以用一个广义方程来表示，其参数对米兰视网膜空间颜色处理的特定方面进行建模。特别是，调整这些参数会导致不同的米兰视网膜算法实现。这项研究有助于更好地理解米兰视网膜算法家族成员之间的异同，并基于它们自身的数学特性为它们提供新的分类方案。
