Incremental Linear Discriminant Analysis Using Sufficient Spanning Sets and Its Applications

This paper presents an incremental learning solution for Linear Discriminant Analysis (LDA) and its applications to object recognition problems. We apply the sufficient spanning set approximation in three steps i.e. update for the total scatter matrix, between-class scatter matrix and the projected data matrix, which leads an online solution which closely agrees with the batch solution in accuracy while significantly reducing the computational complexity. The algorithm yields an efficient solution to incremental LDA even when the number of classes as well as the set size is large. The incremental LDA method has been also shown useful for semi-supervised online learning. Label propagation is done by integrating the incremental LDA into an EM framework. The method has been demonstrated in the task of merging large datasets which were collected during MPEG standardization for face image retrieval, face authentication using the BANCA dataset, and object categorisation using the Caltech101 dataset. © 2010 Springer Science+Business Media, LLC.

A lattice dynamical treatment for the total potential energy of single-walled carbon nanotubes and its applications: relaxed equilibrium structure, elastic properties, and vibrational modes of ultra-narrow tubes

In this paper, we propose a lattice dynamic treatment for the total potential energy of single-walled carbon nanotubes (SWCNTs) which is, apart from a parameter for the nonlinear effects, extracted from the vibrational energy of the planar graphene sheet. The energetics, elasticity and lattice dynamics are treated in terms of the same set of force constants, independently of the tube structures. Based upon this proposal, we have investigated systematically the relaxed lattice configuration for narrow SWCNTs, the strain energy, the Young's modulus and Poisson ratio, and the lattice vibrational properties with respect to the relaxed equilibrium tubule structure. Our calculated results for various physical quantities are nicely in consistency with existing experimental measurements. In particular, we verified that the relaxation effect makes the bond length longer and the frequencies of various optical vibrational modes softer. Our calculation provides evidence that the Young's modulus of an armchair tube exceeds that of the planar graphene sheet, and that the large diameter limits of the Young's modulus and Poisson ratio are in agreement with the experimental values of graphite; the calculated radial breathing modes for ultra-narrow tubes with diameters ranging between 2 and 5 angstrom coincide with the experimental results and the existing ab initio calculations with satisfaction. For narrow tubes with a diameter of 20 angstrom, the calculated frequencies of optical modes in the tubule's tangential plane, as well as those of radial breathing modes, are also in good agreement with the experimental measurements. In addition, our calculation shows that various physical quantities of relaxed SWCNTs can actually be expanded in terms of the chiral angle defined for the corresponding ideal SWCNTs.

Effect of Anode Floating Voltage and its Applications in Characterizing Silicon Drift Detectors

Anode floating voltage is predicted and investigated for silicon drift detectors (SDDs) with an active area of 5 mm(2) fabricated by a double-side parallel technology. It is demonstrated that the anode floating voltage increases with the increasing inner ring voltage, and is almost unchanged with the external ring voltage. The anode floating voltage will not be affected by the back electrode biased voltage until it reaches the full-depleted voltage (-50 V) of the SDD. Theoretical analysis and experimental results show that the anode floating voltage is equal to the sum of the inner ring voltage and the built-in potential between the p(+) inner ring and the n(+) anode. A fast checking method before detector encapsulation is proposed by employing the anode floating voltage along with checking the leakage current, potential distribution and drift properties.

Biomimetic pattern recognition theory and its applications

Biomimetic pattern recogntion (BPR), which is based on "cognition" instead of "classification", is much closer to the function of human being. The basis of BPR is the Principle of homology-continuity (PHC), which means the difference between two samples of the same class must be gradually changed. The aim of BPR is to find an optimal covering in the feature space, which emphasizes the "similarity" among homologous group members, rather than "division" in traditional pattern recognition. Some applications of BPR are surveyed, in which the results of BPR are much better than the results of Support Vector Machine. A novel neuron model, Hyper sausage neuron (HSN), is shown as a kind of covering units in BPR. The mathematical description of HSN is given and the 2-dimensional discriminant boundary of HSN is shown. In two special cases, in which samples are distributed in a line segment and a circle, both the HSN networks and RBF networks are used for covering. The results show that HSN networks act better than RBF networks in generalization, especially for small sample set, which are consonant with the results of the applications of BPR. And a brief explanation of the HSN networks' advantages in covering general distributed samples is also given.

High-Dimensional Space Geometrical Informatics and its applications to image restoration

With a view to solve the problems in modern information science, we put forward a new subject named High-Dimensional Space Geometrical Informatics (HDSGI). It builds a bridge between information science and point distribution analysis in high-dimensional space. A good many experimental results certified the correctness and availability of the theory of HDSGI. The proposed method for image restoration is an instance of its application in signal processing. Using an iterative "further blurring-debluring-further blurring" algorithm, the deblured image could be obtained.