组合原则回溯

从组合原则产生的背景和关键入手，分析和叙述了里兹的组合原则在光谱学上的重要意义，同时，也阐述了它与玻尔氢原子理论之间的关系。     

Thec logc contribution to the electrolyte conductance and Onsager's reciprocal relation

The differential equations and the boundary conditions for the nonequilibrium binary distribution function of an unsymmetrical binary electrolyte are derived from the Ebeling-Falkenhagen continuity equation. The connection between the Onsager reciprocal relation and the binary distribution functions is shown. Further, Feistel's result for thec logc contribution to the conductance is extended to unsymmetrical binary electrolytes. The reason for the difference between Feistel's and Chen'sc logc term is explained, and the significance of Onsager's reciprocal relation for the calculation ofc logc and higher-concentration contributions of the conductance is discussed.     

Characterization of Titania/Silica Gel by Means of Low-Pressure Nitrogen Adsorption

Adsorbents synthesized by grafting of titania onto mesoporous silica gel surfaces at different temperatures were studied by means of nitrogen adsorption–desorption and water desorption. The pore size distribution f(R_p) of titania/silica gel depends on the titania concentration (C_TiO2) and the temperature of titania synthesis. Nonuniformity of TiO₂ phase is maximal at a low C_TiO2 value (3.2 wt.% anatase deposited at 473 K), and two peaks of the fractal dimension distribution f(D) are observed at such a concentration of titania, but at larger C_TiO2 values, only one f(D) peak is seen. More ordered filling of pores and adsorption sites by nitrogen, reflecting in the shape of adsorption energy distributions f(E) at different pressures of adsorbate, is observed for adsorbent with titania (rutile+anatase) grafted on silica gel at a higher temperature (673 K).     

Parameter choice by discrepancy principles for ill-posed problems leading to optimal convergence rates

Schock (Ref. 1) considered a general a posteriori parameter choice strategy for the Tikhonov regularization of the ill-posed operator equationTx=y which provides nearly the optimal rate of convergence if the minimal-norm least-squares solution belongs to the range of the operator (T ^* T) ^v , o<v1. Recently, Nair (Ref. 2) improved the result of Schock and also provided the optimal rate ifv=1. In this note, we further improve the result and show in particular that the optimal rate can be achieved for 1/2v1.The final version of this work was written while M. T. Nair was a Visiting Fellow at the Centre for Mathematics and Its Applications, Australian National University, Canberra, Australia. The work of S. George was supported by a Senior Research Fellowship from CSIR, India.     

Prox-regular functions in variational analysis

The class of prox-regular functions covers all l.s.c., proper, convex functions, lower- functions and strongly amenable functions, hence a large core of functions of interest in variational analysis and optimization. The subgradient mappings associated with prox-regular functions have unusually rich properties, which are brought to light here through the study of the associated Moreau envelope functions and proximal mappings. Connections are made between second-order epi-derivatives of the functions and proto-derivatives of their subdifferentials. Conditions are identified under which the Moreau envelope functions are convex or strongly convex, even if the given functions are not.

     

反馈调节的局部正则化鲁棒光流估计

提出了一种引入反馈调节机制的新局部鲁棒光流场计算思想.该思想结合局部法和全局法的特点,首先划分整场为有公共边界的子块,定义距离来判断每个点的子块归属,然后在每个子块内进行光流约束和平滑约束正则化的优化计算.将光流约束中的光强残差作为控制参量,衡量光流准确度的峰值信噪比作为输出,经反馈处理,调节控制参量使光流估计准确度趋向最优.实验结果表明,这种新的结合反馈调节机制局部光流场计算法能使光流准确度获得提升.     

融合注意力线性特征多样化的DR分级模型

糖尿病视网膜病变(diabetic retinopathy, DR)是目前人类的主要致盲疾病之一。针对DR数据集中样本类间差异小和类分布不均衡等制约分级性能提高的问题,本文提出一种融合注意力线性特征多样化(fusion of attention linear feature diversification, FALFD)的分级算法。该算法首先用改进的Res2Net残差网络作为模型骨干来增大感受野,进一步提高网络捕捉特征信息的能力;其次引入自适应特征多样化模块(adaptive feature diversification module, AFDM)对眼底图像可分辨的微小病理特征进行识别,获得具有高语义信息的局部特征,避免单一特征区域的限制,进而提高分级准确度;再后利用双线性注意力融合模块(bilinear attention fusion module, BAFM)增加可判别区域特征的网络权重占比;最后采用正则化焦点损失(focal loss, FL)进一步提升算法的分类性能。在IDRID数据集上,灵敏度和特异性分别为94.20%和97.05%,二次加权系数为87.83%;在APTO...     

交替使用小波去噪和全变差正则化的盲图像恢复算法

盲图像恢复就是在点扩散函数未知情况下从降质观测图像恢复出原图像.该文提出了一种交替使用小波去噪和全变差正则化的盲图像恢复算法.观测模型首先被分解成两个相互关联的子模型,这种分解转化盲恢复问题成为图像去噪和图像恢复两个问题,可以交替采用图像去噪和图像恢复算法求解.模糊辨识阶段,使用全变差正则化算法估计点扩散函数;图像恢复阶段,使用小波去噪和全变差正则化相结合的算法恢复图像.实验结果和与其它方法的比较表明该文算法能够获得更好的恢复效果.