Learnable three-dimensional Gabor convolutional network with global affinity attention for hyperspectral image classification

Benefiting from the development of hyperspectral imaging technology, hyperspectral image (HSI) classification has become a valuable direction in remote sensing image processing. Recently, researchers have found a connection between convolutional neural networks (CNNs) and Gabor filters. Therefore, some Gabor-based CNN methods have been proposed for HSI classification. However, most Gabor-based CNN methods still manually generate Gabor filters whose parameters are empirically set and remain unchanged during the CNN learning process. Moreover, these methods require patch cubes as network inputs. Such patch cubes may contain interference pixels, which will negatively affect the classification results. To address these problems, in this paper, we propose a learnable three-dimensional (3D) Gabor convolutional network with global affinity attention for HSI classification. More precisely, the learnable 3D Gabor convolution kernel is constructed by the 3D Gabor filter, which can be learned and updated during the training process. Furthermore, spatial and spectral global affinity attention modules are introduced to capture more discriminative features between spatial locations and spectral bands in the patch cube, thus alleviating the interfering pixels problem. Experimental results on three well-known HSI datasets (including two natural crop scenarios and one urban scenario) have demonstrated that the proposed network can achieve powerful classification performance and outperforms widely used machine-learning-based and deep-learning-based methods.     

The relation between the radii and the densities of magnetic skyrmions

Compared with the traditional magnetic bubble, a skyrmion has a smaller size, and better stability and therefore is considered as a very promising candidate for future memory devices. When skyrmions are manipulated, erased and created, the density of skyrmions can be varied, however the relationship between the radii and the densities of skyrmions needs more exploration. In this paper, we study this problem both theoretically and by using the lattice simulation. The average radius of skyrmions as a function of material parameters, the strength of the external magnetic field and the density of skyrmions is obtained and verified. With this explicit function, the skyrmion radius can be easily predicted, which is helpful for the future study of skyrmion memory devices.     

上海光源储存环磁聚焦结构设计

上海光源是一台正在建设中的低发射度第三代同步辐射光源. 经过优化后, 储存环有两种直线节长度, 周长432m,在能量3.5GeV下束流发射度为3.9nm.rad, 直线节处的β函数和色散函数有足够的调节范围. 跟踪研究表明, 即使带上磁铁高阶场误差, 储存环仍有足够大的动力学孔径和能量接受度.     

Spin-correlation function of the fully frustrated Ising model and ± J Ising spin glass on the square lattice

In this work we study the correlation function of the ground state of two-dimensional fully frustrated Ising model as well as spin glass. The Pfaffian method is used to calculate free energy and entropy as well as correlation function. We estimate the exponent of spin correlation function for fully frustrated model and spin glass. In this paper an overview of the latest results on the spin correlation function is presented.     

高速光纤系统中光双二进制传输性能分析

研究了光双二进制调制格式的传输性能,分析了光双二进制调制格式的不同实现形式及其对应的时频特性,并进行了仿真。在此基础上,讨论了采用低通滤波器实现光双二进制调制格式时滤波器带宽对调制码型的影响。     

基于OTRA的通用双二阶滤波器设计

提出了把通用双二阶滤波器视为负反馈放大器的观点,根据反馈电路的基本方程,利用3个MOS电阻、2个电容、2个运算跨阻放大器(OTRA)构成了通用双二阶滤波器,该滤波器不仅可从同一端口输出信号,从不同端口输入信号,而且可获得低通、高通、带通、带阻和全通输出.该电路结构简单,便于电控调谐.     

Ta2O5/SiO2硬膜双腔滤光片激光损伤特性研究

研究了Ta2O5／SiO2硬膜双腔干涉滤光片带内、带边及带外的吸收和激光损伤特性。实验发现，对于作用激光，带通滤光片的驻波场分布、吸收率和损伤阈值在带内、带边和带外的响应特性对作用激光波长均呈现出明显的选择性。根据实验结果，结合滤光片的驻波场分析，给出了带通滤光片的损伤机理。     

基于振动陀螺的动态零位校准及其误差补偿

根据固态振动陀螺输出信号和噪声特点对其构成的低成本惯性测量单元(IMU)的原始传感信号进行了快速小波滤波和灰色理论建模处理。运用累加生成操作(AGO)方法得到有规则的单值对应非线性函数,并获得陀螺零位输出在三维空间中的单值映射模型。以时间和温度为输入,根据灰色神经网络建立陀螺的漂移模型,对累加生成方法生成的单值对应非线性函数进行逼近,从而提高了动态测量精度。同时采用活动阈值融合算法,优化陀螺和加速度计动态测量数据。实验证明,上述方法和算法有效提高了系统测量精度。     

地面目标跟踪算法研究

本文重点研究地面运动目标跟踪中利用道路约束提高单目标状态估计精度的方法。按照在Kalman滤波过程中作用的对象,将其归结为三类,即利用道路约束进行量测预处理、模型修正、状态估计处理。首先对这三类方法的原理进行了介绍,给出各自的应用条件。然后将各种方法进行可能的组合,从状态估计精度和计算时间复杂度方面做出了分析。仿真实验将这些方法进行比较,结果表明各种利用道路约束的方法都不同程度地改善了估计性能,且合理的组合可以更有效地利用道路约束,从不同的层面共同提高估计精度。对于各种算法时间复杂度的定量分析,也便于实际应用时对算法的选择。     

二阶自适应陷波器频率估计方法统计性能分析

为对如何提高自适应陷波器频率估计精度提供参考,通过评估自适应陷波器频率估计方法性能,对基于均方误差函数的自适应陷波器频率估计方法进行了统计性能分析。首先,根据误差函数的不同,将自适应陷波器划分为自适应FIR陷波器和自适应IIR陷波器。然后,将自适应FIR陷波器看作自适应IIR陷波器的特例,重点分析了自适应陷波器的误差函数及稳态下的频率估计统计性能,讨论了自适应陷波器参数对正弦信号频率估计精度和收敛速度的影响。最后,给出正弦信号的频率估计计算结果。结果表明,实际计算结果同理论计算结果一致,证明了统计性能分析的正确性。