Subspace distance analysis with application to adaptive Bayesian algorithm for face recognition

We propose subspace distance measures to analyze the similarity between intrapersonal face subspaces, which characterize the variations between face images of the same individual. We call the conventional intrapersonal subspace average intrapersonal subspace (AIS) because the image differences often come from a large number of persons. An intrapersonal subspace is referred to as specific intrapersonal subspace (SIS) if the image differences are from just one person. We demonstrate that SIS varies significantly from person to person, and most SISs are not similar to AIS. Based on these observations, we introduce the maximum a posteriori (MAP) adaptation to the problem of SIS estimation, and apply it to the Bayesian face recognition algorithm. Experimental results show that the adaptive Bayesian algorithm outperforms the non-adaptive Bayesian algorithm as well as Eigenface and Fisherface methods if a small number of adaptation images are available.     

Averaging of nonsmooth systems using dither

It was shown by Zames and Shneydor and later by Mossaheb that a high-frequency dither signal of a quite arbitrary shape can be used to narrow the effective nonlinear sector of Lipschitz continuous feedback systems. In this paper, it is shown that also discontinuous nonlinearities of feedback systems can be narrowed using dither, as long as the amplitude distribution function of the dither is absolutely continuous and has bounded derivative. The averaged system is proven to approximate the dithered system with an error of the order of dither period.     

An empirical investigation of competency factors affecting e-business success in European SMEs

In the last decade there was growing interest in strategic management literature about factors that influence a company's ability to use IT. There is general consensus that knowledge and competency are necessary in developing an IT capability, but there is very little understanding of what the necessary competencies are, and how they influence IS usage in different contexts. The small and medium-sized enterprise context is particularly interesting for two reasons: it constitutes a major part of the economy and it has been relatively unsuccessful in exploiting e-business.     

Parameter identification for nonlinear systems: Guaranteed confidence regions through LSCR

In this paper we consider the problem of constructing confidence regions for the parameters of nonlinear dynamical systems. The proposed method uses higher order statistics and extends the LSCR (leave-out sign-dominant correlation regions) algorithm for linear systems introduced in Campi and Weyer [2005, Guaranteed non-asymptotic confidence regions in system identification. Automatica 41(10), 1751-1764. Extended version available at 〈http://www.ing.unibs.it/∼campi〉]. The confidence regions contain the true parameter value with a guaranteed probability for any finite number of data points. Moreover, the confidence regions shrink around the true parameter value as the number of data points increases. The usefulness of the proposed approach is illustrated on some simple examples.     

How to protect exchanged secrets in the fair exchange protocol with off-line TTP

The most efficient approach to a fair exchange is to use an off-line trusted third party (TTP for short) who gets involved only when a dispute between two parties occurs. However, exchanged secrets that are not protected properly may be exposed to the TTP when one of these two parties asks the TTP to mediate the dispute. The first work to protect exchanged secrets from TTP’s misuse was proposed by Franklin and Reiter in 1997. They provided an on-line semi-trusted third party instead of an off-line third party for a fair exchange. Their schemes forced the TTP to take part in the protocol for all the cases, and thus, deemed to be impractical.The present paper, introduces several models, including single and multiple TTPs, to focus the attention not only on security properties, but also on reliability functions. In the single TTP mode, our new scheme can protect the exchanged data through an efficient exchange protocol, whereas in the multiple TTP mode, we proposed an approach to balance the security level with the reliability of the system.     

An efficient controller structure with minimum roundoff noise gain

Based on a polynomial operator approach, a new sparse controller structure is derived, which is actually an improved version of the recently proposed structure [Li, G. (2004). A polynomial-operator-based DFIIt structure for IIR filters. IEEE Transactions on Circuits and Systems II, 51, 147-151]. The performance of the proposed structure is analyzed by deriving the corresponding expression of roundoff noise gain and the problem of finding optimized sparse structures is solved efficiently with a genetic algorithm (GA). A numerical example is given, which shows that the newly developed structure can achieve much better performance than some well-known structures and particularly outperforms the traditional optimal fully parametrized realization greatly in terms of reducing roundoff noise and implementation complexity.     

Physical design of the VCI wrappers for the on-chip packet-switched network named SPIN

This paper presents the physical design methodology of the VCI/SPIN wrappers. The challenge was the validation of the wrappers’ specification with Alliance CAD Tools.A wrapper is a standard gateway used by a subscriber to access an interconnect. Because there are as many wrappers as subscribers, it is relevant to have an estimation of the wrapper’s area, its frequency and then compare these results to the router’s.The interest to perform this study with academic tools is to show that commercial tools are not the unique way for experimentations.     

A rule-based approach for robust clump splitting

This paper presents a robust rule-based approach for the splitting of binary clumps that are formed by objects of diverse shapes and sizes. First, the deepest boundary pixels, i.e., the concavity pixels in a clump, are detected using a fast and accurate scheme. Next, concavity-based rules are applied to generate the candidate split lines that join pairs of concavity pixels. A figure of merit is used to determine the best split line from the set of candidate lines. Experimental results show that the proposed approach is robust and accurate.