针对大破损彩色纹理图像的修复问题,将TV-L1模型推广到非局部CTV-L1模型。该模型不仅包含非局部算子,同时还引入了CTV(color total variation)规则项,前者可以修复大破损纹理图像,后者充分考虑了彩色图像层与层之间的耦合关系,在处理彩...针对大破损彩色纹理图像的修复问题,将TV-L1模型推广到非局部CTV-L1模型。该模型不仅包含非局部算子,同时还引入了CTV(color total variation)规则项,前者可以修复大破损纹理图像,后者充分考虑了彩色图像层与层之间的耦合关系,在处理彩色图像时可以有效地保持边缘。为提高模型的运算效率,通过引入辅助变量和Lagrange乘子为其设计了相应的增广Lagrangian算法。数值实验结果证实所提出的模型在处理彩色图像时可以有效地保持边缘,同时去除图像中异常的不规则点。该研究可以推广到彩色纹理图像椒盐噪声去除及彩色纹理图像分割中。展开更多
Efficient large-scale nondestructive quality assessment of graphene on Pt is essential to the in-depth growth research and practical applications of graphene.Here,we present a very simple method for directly observing...Efficient large-scale nondestructive quality assessment of graphene on Pt is essential to the in-depth growth research and practical applications of graphene.Here,we present a very simple method for directly observing the domains and defects in graphene on Pt using an ordinary optical microscope.This was achieved by modifying graphene on Pt using methylene blue(MB).Because the chemical activities of graphene and Pt surface differ significantly,the adsorption and reaction of MB on graphene and platinum surface differ.We can determine the distribution of graphene crystal domains and defects by comparing the colors in the optical images.In addition,this characterization method causes no obvious damage to the Pt substrate and graphene.Moreover,it does not affect the recycling of the substrate or the subsequent characterization or application of graphene.Our study provides a nondestructive method for measuring the quality of graphene on Pt on a large scale,as well as a reference for the characterization and doping of other two-dimensional materials.展开更多
文摘针对大破损彩色纹理图像的修复问题,将TV-L1模型推广到非局部CTV-L1模型。该模型不仅包含非局部算子,同时还引入了CTV(color total variation)规则项,前者可以修复大破损纹理图像,后者充分考虑了彩色图像层与层之间的耦合关系,在处理彩色图像时可以有效地保持边缘。为提高模型的运算效率,通过引入辅助变量和Lagrange乘子为其设计了相应的增广Lagrangian算法。数值实验结果证实所提出的模型在处理彩色图像时可以有效地保持边缘,同时去除图像中异常的不规则点。该研究可以推广到彩色纹理图像椒盐噪声去除及彩色纹理图像分割中。
基金financially supported by the National Defense Technology Innovation Special Zone Project, the National Natural Science Foundation of China (51402342)the Science and Technology Commission of Shanghai Municipality (20501130200)。
文摘Efficient large-scale nondestructive quality assessment of graphene on Pt is essential to the in-depth growth research and practical applications of graphene.Here,we present a very simple method for directly observing the domains and defects in graphene on Pt using an ordinary optical microscope.This was achieved by modifying graphene on Pt using methylene blue(MB).Because the chemical activities of graphene and Pt surface differ significantly,the adsorption and reaction of MB on graphene and platinum surface differ.We can determine the distribution of graphene crystal domains and defects by comparing the colors in the optical images.In addition,this characterization method causes no obvious damage to the Pt substrate and graphene.Moreover,it does not affect the recycling of the substrate or the subsequent characterization or application of graphene.Our study provides a nondestructive method for measuring the quality of graphene on Pt on a large scale,as well as a reference for the characterization and doping of other two-dimensional materials.