非局部均值去噪(Non-local means,NLM)算法利用图像的自相似性,取得了很好的去噪效果.然而,NLM算法对图像中不相似的邻域块分配了过大的权重,此外算法的搜索窗大小和滤波参数等通常是固定的且无法根据图像内容的变化做出自适应的调整....非局部均值去噪(Non-local means,NLM)算法利用图像的自相似性,取得了很好的去噪效果.然而,NLM算法对图像中不相似的邻域块分配了过大的权重,此外算法的搜索窗大小和滤波参数等通常是固定的且无法根据图像内容的变化做出自适应的调整.针对上述问题,本文提出一种无监督多重非局部融合(Unsupervised multi-non-local fusion,UMNLF)的图像去噪方法,即变换搜索窗等组合参数得到多个去噪结果,并利用SURE(Stein's unbiased risk estimator)对这些结果进行无监督的随机线性组合以获得最终结果.首先,为了滤除不相似或者相似度较低的邻域块,本文引入一种基于可微分硬阈值函数的非局部均值(Non-local means with a differential hard threshold function,NLM-DT)算法,并结合快速傅里叶变换(Fast Fourier transformation,FFT),初步提升算法的去噪效果和速度;其次,针对不同的组合参数,利用快速NLM-DT算法串联生成多个去噪结果;然后,采用蒙特卡洛随机采样的思想对上述多个去噪结果进行随机的线性组合,并利用基于SURE特征加权的移动平均滤波算法来抑制多个去噪结果组合引起的抖动噪声;最后,利用噪声图像和移动平均滤波后图像的SURE进行梯度的反向传递来优化随机线性组合的系数.在公开数据集上的实验结果表明:UMNLF算法去噪结果的峰值信噪比(Peak signal to noise ratio,PSNR)超过了NLM及其大部分改进算法,以及在部分图像上超过了BM3D算法.同时,UM-NLF相比于BM3D算法在视觉上产生更少的振铃伪影,改善了图像的视觉质量.展开更多
Alzheimer’s disease(AD)is the most common neurodegenerative disease characterized by progressive cognitive impairment.The pathogenesis of AD is complex,and its susceptibility and development process are affected by a...Alzheimer’s disease(AD)is the most common neurodegenerative disease characterized by progressive cognitive impairment.The pathogenesis of AD is complex,and its susceptibility and development process are affected by age,genetic and epigenetic factors.Recent studies confirmed that gut microbiota(GM)might contribute to AD through a variety of pathways including hypothalamic pituitary adrenal axis and inflammatory and immune processes.CM formula,herbs,and monomer enjoy unique advantages to treat and prevent AD.Hence,the purpose of this review is to outline the roles of GM and its core metabolites in the pathogenesis of AD.Research progress of CMs regarding the mechanisms of how they regulate GM to improve cognitive impairment of AD is also reviewed.The authors tried to explore new therapeutic strategies to AD based on the regulation of GM using CM.展开更多
文摘非局部均值去噪(Non-local means,NLM)算法利用图像的自相似性,取得了很好的去噪效果.然而,NLM算法对图像中不相似的邻域块分配了过大的权重,此外算法的搜索窗大小和滤波参数等通常是固定的且无法根据图像内容的变化做出自适应的调整.针对上述问题,本文提出一种无监督多重非局部融合(Unsupervised multi-non-local fusion,UMNLF)的图像去噪方法,即变换搜索窗等组合参数得到多个去噪结果,并利用SURE(Stein's unbiased risk estimator)对这些结果进行无监督的随机线性组合以获得最终结果.首先,为了滤除不相似或者相似度较低的邻域块,本文引入一种基于可微分硬阈值函数的非局部均值(Non-local means with a differential hard threshold function,NLM-DT)算法,并结合快速傅里叶变换(Fast Fourier transformation,FFT),初步提升算法的去噪效果和速度;其次,针对不同的组合参数,利用快速NLM-DT算法串联生成多个去噪结果;然后,采用蒙特卡洛随机采样的思想对上述多个去噪结果进行随机的线性组合,并利用基于SURE特征加权的移动平均滤波算法来抑制多个去噪结果组合引起的抖动噪声;最后,利用噪声图像和移动平均滤波后图像的SURE进行梯度的反向传递来优化随机线性组合的系数.在公开数据集上的实验结果表明:UMNLF算法去噪结果的峰值信噪比(Peak signal to noise ratio,PSNR)超过了NLM及其大部分改进算法,以及在部分图像上超过了BM3D算法.同时,UM-NLF相比于BM3D算法在视觉上产生更少的振铃伪影,改善了图像的视觉质量.
基金Supported by National Natural Science Foundation of China(No.81704004)Scientific and Technological Research Program of Tianjin Municipal Education Commission(No.2018KJ031)Research and Innovation Project of Graduate(No.YJSKC-20191003,No.ZXYCXLX201804)。
文摘Alzheimer’s disease(AD)is the most common neurodegenerative disease characterized by progressive cognitive impairment.The pathogenesis of AD is complex,and its susceptibility and development process are affected by age,genetic and epigenetic factors.Recent studies confirmed that gut microbiota(GM)might contribute to AD through a variety of pathways including hypothalamic pituitary adrenal axis and inflammatory and immune processes.CM formula,herbs,and monomer enjoy unique advantages to treat and prevent AD.Hence,the purpose of this review is to outline the roles of GM and its core metabolites in the pathogenesis of AD.Research progress of CMs regarding the mechanisms of how they regulate GM to improve cognitive impairment of AD is also reviewed.The authors tried to explore new therapeutic strategies to AD based on the regulation of GM using CM.
