Catálogo de publicaciones - libros

Past few years, a lot of research on moment functions have been explored in pattern recognition. Several new techniques have been investigated to improve conventional regular moment by proposing the scaling factor of geometrical function. In this paper, integrated scaling formulations of Aspect Invariant Moment and Higher Order Scaling Invariant with United Moment Invariant are presented in Writer Identification to seek the invarianceness of authorship or individuality of handwriting perseverance. Mathematical proving and results of computer simulations are included to verify the validity of the proposed technique in identifying eccentricity of the author in Writer Identification.

In this paper we present a user independent real-time capable automatic method for recognition of facial expressions related to basic emotions from stereo image sequences. The method automatically detects faces in unconstraint pose based on depth and color information. In order to overcome difficulties caused by increasing change in pose, lighting transitions, or complicated background, we introduce a face normalization algorithm based on an Iterative Closest Point algorithm. In normalized face images we defined a set of physiologically motivated face regions related to a subset of facial muscles which are apt to automatically detect the six well-known basis emotions. Visual facial expression analysis takes place by an optical flow based feature extraction and a nearest neighbor classification, which uses a distance measure, i.e. the current flow vector pattern is matched against empirically determined ground truth data.

This paper is part of a larger study dedicated by the authors to the description of printed Romanian language as an information source. Here, the statistical investigation attempts to get an answer concerning the mathematical model of the language with orthography and punctuation marks included into the alphabet. To come out to an accurate result, the authors processed the information obtained out of multiple data sets sampled from a corpus linguistics, by using the following statistical inferences: probability estimation with multiple confidence intervals, test of the hypothesis that the probability belongs to an interval, and test of the equality between two probabilities. The second type statistical error probability involved in the tests was considered. The experimental results, which are new for printed Romanian, refer to the letter, digram and trigram statistical structure in a corpus linguistics of 93 books (about 50 millions characters).

Natural language interfaces to databases are considered one of the best alternatives for final users who wish to make complex, uncommon and frequent queries, which is a very common need in organizations. The use of this type of interfaces has been very limited, due to their limited publicizing and the complexity to adapt them to users’ needs, and because their precision varies widely. We propose as a solution to the problem of customizing this type of interfaces, the use of an ontology as a knowledge base whose design is simple and flexible enough to make the use and acceptance of these interfaces more accessible. This paper describes the design of the ontology, as well as a series of comparative evaluations of this approach versus the customization process of a commercial interface. This evaluation aims at assessing the acceptance of this approach by of those that will potentially customize the interface to a database, in contrast to the precision tests that are commonly applied to this type of interfaces. In spite of the difficulties found to carry out the evaluations, the results show that the use of our approach is preferred as a natural language interface customization means to the process of the most popular commercial interface. The estimations indicate that the potential people on charge of the process of customization of this type of interfaces considers that using the ontology as interface knowledge base would allow to answer a wider diversity of types of queries than those that would allow to answer a commercial interface.

During the last years, they have been published many studies carrying out the improving and optimization of the coherent optical systems by "computer experiments". Based on some computational models known in the literature, this paper proposes to present the main author’s results obtained by numerical simulation using a Runge - Kutta type method. The used computational method refers to the nonlinear transport coupled equations in the case of the fiber amplifier and to the rate equations for the laser systems. Some new feature of the computer modeled systems have been put into evidence, for designers utility in different applications.

This paper aims to present a numerical simulation of soliton propagation, based on Korteweg-de Vries equation, using a powerful PC program (Maple10) who permits to perform numerical calculations, plot or animate functions and manage analytical expressions. We discuss a model of one soliton propagation in a nonlinear medium and the interaction between two solitons, taking account both of fiber dispersion and nonlinearity. Numerical simulations show how soliton propagate in optical fibers and how two solitons interact, passing one through another, with only a phase change. These simulations are thought to be useful for both the designers working in digital data transmission and students performing numerical simulation on soliton propagation as computer experiments in Optoelectronics laboratory.

This article give the definition of assignability measue of generalized decentralized control system and the related theorem and the proof as well, which is a natural generalization of conclusion. This measure quantifies how close a mode is to being a finite decentralized fixed mode.If a mode is close to being a FDFM, large controller gains are required to shift it .This result is useful for the analysis of systems with small modeling errors and regularly perturbed systems.

In this article, a technique based on the acoustic emission (AE) signal fractal and wavelet analysis are proposed for tool condition monitoring. it is difficult to obtain an effective result by these raw acoustic emission data. The local characterize of frequency band, which contains the main energy of AE signals, is depicted by the wavelet multi-resolution analysis, fractal dimension can describe the complexity of time series. It is found that the fault signal can effectively be extracted by wavelet transform and fractal dimension. Experimental results prove that this method is effectively.

This paper presents uniformly ultimate boundedness (UUB) control design for switched linear systems with parametric uncertainties. Only the possible bound of the uncertainty is needed. Under arbitrary switching laws, a continuous state feedback control scheme is proposed in order to guarantee uniformly ultimate boundedness of every system response within an arbitrary small neighborhood of the zero state. The design techniques are based on common Lyapunov functions and Lyapunov minimax approach.

The momentless state equations of an hyperboloid shell which is obtained as a surface of revolution are considered under a localized transversal load and localized initial values of stresses. An approximate analytical solution is proposed by using a family of wavelets.