基于改进高斯核度量和KPCA的数据聚类新方法

大多数超椭球聚类（hyper-ellipsoidal clustering,HEC）算法都使用马氏距离作为距离度量,已经证明在该条件下划分聚类的代价函数是常量,导致HEC无法实现椭球聚类.本文说明了使用改进高斯核的HEC算法可以解释为寻找体积和密度都紧凑的椭球分簇,并提出了一种实用HEC算法-K-HEC,该算法能够有效地处理椭球形、不同大小和不同密度的分簇.为实现更复杂形状数据集的聚类,使用定义在核特征空间的椭球来改进K-HEC算法的能力,提出了EK-HEC算法.仿真实验证明所提出算法在聚类结果和性能上均优于K-means算法、模糊C-means算法、GMM-EM算法和基于最小体积椭球（minimum-volume ellipsoids,MVE）的马氏HEC算法,从而证明了本文算法的可行性和有效性.     

自主探测系统的Bayes递推估计策略

探测系统被动地接爱外界的测量信息是不够的,因此方式获得的数据是不完备的,由此揄得出的结论在一定程度上具有不确定性,是不能接受的,为获得可靠的信息,常采用多传感器的数据融合或装备自主传感器的方法,文中将Bayes递推估计策略应用于自主探测系统,使其能最逼真的再再所关心的探测对象。     

New approach to metamerism measurement on halftone color images

There is often a phenomenon depending on dominant tones which results in the end users dissatisfaction with halftone color reproduction. This phenomenon is known as “metamerism” and it is the subject of research for many scientists.The metamerism on halftone color images with different dominant tones (blue, red, green and yellow) was researched in this paper, at the light sources D₅₀, D₆₅ and daylight. The results of the objective and visual evaluation were presented on the basis of colorimetric measurement of the grey balance patch, by the means of MacAdam ellipsoids in three dimension CIE L^*a^*b^* system. Both the metamerism index MI and the ellipsoid volume have also been calculated.This paper confirms that the metamerism of the halftone color images can be estimated by measuring CIE L^*a^*b^* values of the grey balance patch and by calculating MacAdam ellipsoids.     

Direct numerical simulation of moderate-Reynolds-number flow past arrays of ellipsoids

Through particle-resolved direct numerical simulations of flow past arrays of ellipsoids, the hydrodynamic force on ellipsoids depends on the particle orientation, aspect ratio, particle Reynolds number, and solid volume fraction is revealed at moderate Reynolds numbers. The results show that the mean drag force on arrays of prolate/oblate ellipsoids decreases/increases as the Hermans orientation factor increases when flows are in the reference direction defined by the average symmetric axis of particles. The individual drag force on a prolate/oblate ellipsoid increases/decreases with the increase of incidence angle, and it is also affected by the orientation of surrounding particles. The individual lift force is also significant when the aspect ratio is away from unity at large particle Reynolds numbers. Based on simulation results, correlations for the hydrodynamic force on ellipsoids at arbitrary particle Reynolds numbers, solid volume fractions, Hermans orientation factors, incidence angles, and aspect ratios are formulated.     

Data Clustering Method Using a Modified Gaussian Kernel Metric and Kernel PCA

Most hyper‐ellipsoidal clustering (HEC) approaches use the Mahalanobis distance as a distance metric. It has been proven that HEC, under this condition, cannot be realized since the cost function of partitional clustering is a constant. We demonstrate that HEC with a modified Gaussian kernel metric can be interpreted as a problem of finding condensed ellipsoidal clusters (with respect to the volumes and densities of the clusters) and propose a practical HEC algorithm that is able to efficiently handle clusters that are ellipsoidal in shape and that are of different size and density. We then try to refine the HEC algorithm by utilizing ellipsoids defined on the kernel feature space to deal with more complex‐shaped clusters. The proposed methods lead to a significant improvement in the clustering results over K‐means algorithm, fuzzy C‐means algorithm, GMM‐EM algorithm, and HEC algorithm based on minimum‐volume ellipsoids using Mahalanobis distance.     

Bubble dynamics in bubbling fluidized beds of ellipsoidal particles

This article presents a CFD-DEM study on the effect of particle shape on bubble dynamics in bubbling fluidized beds. The particles used are ellipsoids, covering from disk-type to cylinder-type. The phenomena such as bubble coalescence and splitting are successfully generated, and the results are compared with literature, showing a good agreement. The results demonstrate that the bubble forming/rising regions and patterns are influenced significantly by particle shape. Ellipsoidal particles have asymmetrical bubble patterns with two or more circulation vortices while the bubbles for spherical particles form at the bed centerline and rise through the center of the bed. Hence, the vertical mass flux at the bed centerline for spheres is always positive, and ellipsoids have negative or positive vertical mass fluxes. The solid mixing estimated based on the dispersion coefficient revealed poor mixing for ellipsoids. Spherical particles have a larger bubble size and higher bubble rising velocity than ellipsoids.     

凸规划的一种变椭球半径的内点算法

针对凸规划的拉格朗日对偶问题,用拉格朗日乘子法求解受椭球约束的子问题,在迭代公式中,通过改变变椭球半径,使对偶迭代序列在可行域内产生,简化了计算过程,给出了一种新的收敛更快的算法。     

自适应椭球包围盒改进织物碰撞检测方法

为了快速处理织物与模型的碰撞以获得仿真结果,提出一种基于自适应椭球包围盒的碰撞检测方法.该方法以法向距离和径向距离的加权平均值作为表面误差,数值化表示椭球与模型之间的拟合度,将优化K均值聚类方法用于自适应划分模型生成一系列逼近模型的最小体积包围椭球;根据表面误差的大小对椭球分类处理,误差小的椭球直接取代模型区域与织物碰撞,将三角网格之间的碰撞问题转化为质点相对于椭球的穿透测试,误差大的椭球与织物的碰撞检测仅用于排除不可能发生碰撞的三角形,再继续进行三角形之间的精确碰撞检测.实验结果表明,文中方法能快速、自适应地划分任意模型生成高度拟合的椭球,可大幅度地减少碰撞检测计算量,在提高计算速度的同时保证了仿真的真实感.     

Shape factors for the analysis of diffusion in air drying of grains

The shape factors (φ) of different varieties of wheat, corn (maize), rice, sorghum, sunflower seeds and soybean have been experimentally determined. Shape factor values ranged from 0.63 to 0.99. These values are useful for correcting literature moisture diffusion coefficients in grains of different shapes, obtained from the analysis of drying curves; this allows direct comparison of diffusivity in different grains.     

Stereological estimation of the volume-weighted mean volume of arbitrary particles observed on random sections*

A stereological estimator of the weighted mean volume of particles of arbitrary shape is described. This unbiased estimator is based on simple point-sampling of linear intercept lengths. The complete absence of shape assumptions effectively breaks the long-standing ‘convexity-barrier’: the only requirement here is that individual particles can be unambiguously identified by their profiles on random sections. Practical details of the simple estimation procedure and an example with very irregular particles are reported. Finally, an estimator of the variance of the weighted distribution of particle volume is discussed. This estimator is also valid for particles of arbitrary shape. For any mixture of ellipsoids (spheres, oblates, prolates and triaxial ellipsoids) the estimator is reduced to a simple function of measurements of diameters in the section plane.