Extensions of LDA by PCA mixture model and class-wise features

Linear discriminant analysis (LDA) is a data discrimination technique that seeks transformation to maximize the ratio of the between-class scatter and the within-class scatter. While it has been successfully applied to several applications, it has two limitations, both concerning the underfitting problem. First, it fails to discriminate data with complex distributions since all data in each class are assumed to be distributed in the Gaussian manner. Second, it can lose class-wise information, since it produces only one transformation over the entire range of classes. We propose three extensions of LDA to overcome the above problems. The first extension overcomes the first problem by modelling the within-class scatter using a PCA mixture model that can represent more complex distribution. The second extension overcomes the second problem by taking different transformation for each class in order to provide class-wise features. The third extension combines these two modifications by representing each class in terms of the PCA mixture model and taking different transformation for each mixture component. It is shown that all our proposed extensions of LDA outperform LDA concerning classification errors for synthetic data classification, hand-written digit recognition, and alphabet recognition.     

A tolerant approach to reconstruct topology from unorganized trimmed surfaces

Presented in this paper is a topology reconstruction algorithm from a set of unorganized trimmed surfaces. Error-prone small geometric elements are handled to give proper topological information. It gives complete topology to topologically complete models, and it is also tolerant to incomplete models. The proposed algorithm is vertex-based in that clues for topological information are searched from the set of vertices first, not from that of edges.     

Preparation of Compound Ultrafine CeO2 by Wet-Solid-Phase Mechanochemical Modification

To satisfy practical requirements from industrial applications, an alternate route for synthesis compound ultrafine CeO2 powders by wet-solid-phase mechanochemical modification using industrial grade hydrated cerium carbonate as raw material was proposed.The effect of modifier reaction percentage, reaction time, calcining temperature and modifier amount on particle size, density, suspensibility, and hardness of compound CeO2 powder was investigated.The phase evolutions of preparation process were characterized by XRD.SEM micrograph of the final product shows that compound CeO2 powders obtained are well-dispersed, spherically-shaped, uniformly-sized and submicron-sized particles.The method is readily available in raw material, low in cost, simple in process, and has great potential for industrialization.The compound CeO2 powders of different physical properties can be synthesized by controlling the above-mentioned influence factors in preparation process.     

Measured equation of invariance with linear tensor relation for three-dimensional scattering problems

The postulate of a linear tensor relation in the measured equation of invariance (MEI) is proposed for three-dimensional (3D) problems. As a result, all three components of the field vector to be solved are coupled in the MEI. Combined with the finite difference method, the present method is applied to the analysis of scattering by 3D conducting objects and results in a significant improvement in the accuracy of numerical results as compared to those obtained with an uncoupled linear relation.     

多粒度全光网中的多层联合路由

多粒度交换技术利用波带级路由以及光纤级路由,可有效降低光交叉连接器的复杂度.但多粒度交换增加了光网络的逻辑层次结构,使得网络中的路由与资源分配问题更为复杂,因此,在多粒度全光网中实现多层联合路由是提高网络性能的关键.本文对多粒度交叉连接结构进行了分析,并对多粒度全光网中的多层联合路由问题进行了研究.     

Effect of thermal treatment on ZnO substrate for epitaxial growth

ZnO is a highly efficient photon emitter, and has optical and piezoelectric properties that are attractive for a variety of applications in sensors and potentially optoelectronic devices such as emitters. Due to its identical stacking order and close lattice match to GaN, it is also being developed as a substrate material for GaN epitaxy. However, the surface finish of the ZnO is such that much of the damage induced by sawing and follow up mechanical polishing remains. We developed a thermal treatment method to eliminate surface damage on the 0 face of ZnO (0 0 0 1) to prepare it for epitaxial growth. Atomic force microscopy images of ZnO (0 0 0 1) annealed at 1050 °C for 3 h etc. show that residual scratches from mechanical polishing are removed and atomically flat, terrace-like surfaces are attained. In addition, low-temperature photoluminescence and high-resolution X-ray diffraction measurements have been employed to investigate the effect of annealing on ZnO substrates.     

高阻值Cd1-xZnxTe晶体的生长及其性能测试

通过适当的工艺措施，采用Bridgman法生长了直径为30mm的X射线及γ射线探测器级的Cd.9Zn0．1Te晶锭．测试结果表明：该晶锭结晶质量良好，位错密度低，成分均匀，杂质含量低，红外透过率和电阻率都十分接近本征Cd.9Zn0．1Te的值．并从晶体的生长特性、缺陷和杂质的角度，分析了生长高性能晶体的条件，研究了生长Cdl-xZnxTe晶体的x值与缺陷和杂质浓度之间的关系．     

杭-宁输气管道工程定向钻穿越设计浅析

杭州-宁波天然气输气管道工程（简称杭-宁输气管道工程）是浙江省天然气长输管网主干线之一，根据工程河流穿越众多、渠塘穿越频繁的特点，对杭-宁输气管道工程水域穿越的特点和穿越方式作了总结；重点分析了水域定向钻穿越设计及需要注意的问题，为今后浙江省天然气管网建设的水域穿越设计作了探讨。     

转动玻璃圆筒激光器

本文介绍了转动玻璃圆筒激光器的设计及实验结果,得到了脉冲能量为13J,重复频率为15Hz的激光输出。电光转换效率为2.2％,斜效率为3.4％。     

Data fitting with a spline using a real-coded genetic algorithm

To obtain a good approximation for data fitting with a spline, frequently we have to deal with knots as variables. The problem to be solved then becomes a continuous nonlinear and multivariate optimization problem with many local optima. Therefore, it is difficult to obtain the global optimum. In this paper, we propose a method for solving this problem by using a real-coded genetic algorithm. Our method can treat not only data with a smooth underlying function, but also data with an underlying function having discontinuous points and/or cusps. We search for the best model among candidate models by using the Bayes Information Criterion (BIC). With this, we can appropriately determine the number and locations of knots automatically and simultaneously. Five examples of data fitting are given to show the performance of our method.