Biomimetic (topological) pattern recognition - A new model of pattern recognition theory and its application

A new theoretical model of Pattern Recognition principles was proposed, which is based on "matter cognition" instead of "matter classification" in traditional statistical Pattern Recognition. This new model is closer to the function of human being, rather than traditional statistical Pattern Recognition using "optimal separating" as its main principle. So the new model of Pattern Recognition is called the Biomimetic Pattern Recognition (BPR)(1). Its mathematical basis is placed on topological analysis of the sample set in the high dimensional feature space. Therefore, it is also called the Topological Pattern Recognition (TPR). The fundamental idea of this model is based on the fact of the continuity in the feature space of any one of the certain kinds of samples. We experimented with the Biomimetic Pattern Recognition (BPR) by using artificial neural networks, which act through covering the high dimensional geometrical distribution of the sample set in the feature space. Onmidirectionally cognitive tests were done on various kinds of animal and vehicle models of rather similar shapes. For the total 8800 tests, the correct recognition rate is 99.87%. The rejection rate is 0.13% and on the condition of zero error rates, the correct rate of BPR was much better than that of RBF-SVM.

Theoretical study on analytical potential function and spectroscopic parameters for CaF molecule

The equilibrium properties and potential energy curves of the ground electronic state of CaF have been calculated using the Brueckner Doubles calculation with a triples contribution added [BD(T)] and the gradient-corrected density functional theory with three-parameter exact exchange mixing (B3LY-P) method, with 6-311 + G*,6-311 + G(2df,2pd) and 6-311 + G(3df,3pd) basis sets. All the computational PECs are fitted to analytical potential energy functions using Murrell-Sorbie, Huxley and Tang-Toennies potentials. Based on this, the spectroscopic parameters are calculated, and then compared with some other theoretical and experimental data. (C) 2004 Elsevier B.V. All rights reserved.

Application of Dempster-Shafer evidence theory to unsupervised classification in multisource remote sensing

The aim of this paper is to show that Dempster-Shafer evidence theory may be successfully applied to unsupervised classification in multisource remote sensing. Dempster-Shafer formulation allows for consideration of unions of classes, and to represent both imprecision and uncertainty, through the definition of belief and plausibility functions. These two functions, derived from mass function, are generally chosen in a supervised way. In this paper, the authors describe an unsupervised method, based on the comparison of monosource classification results, to select the classes necessary for Dempster-Shafer evidence combination and to define their mass functions. Data fusion is then performed, discarding invalid clusters (e.g. corresponding to conflicting information) thank to an iterative process. Unsupervised multisource classification algorithm is applied to MAC-Europe'91 multisensor airborne campaign data collected over the Orgeval French site. Classification results using different combinations of sensors (TMS and AirSAR) or wavelengths (L- and C-bands) are compared. Performance of data fusion is evaluated in terms of identification of land cover types. The best results are obtained when all three data sets are used. Furthermore, some other combinations of data are tried, and their ability to discriminate between the different land cover types is quantified

Calculating the equation of state parameters and predicting the spinodal curve of isotactic polypropylene/poly(ethylene-co-octene) blend by molecular dynamics simulations combined with Sanchez-Lacombe lattice fluid theory

Molecular dynamics simulations are adopted to calculate the equation of state characteristic parameters P*, rho*, and T* of isotactic polypropylene (iPP) and poly(ethylene-co-octene) (PEOC), which can be further used in the Sanchez-Lacombe lattice fluid theory (SLLFT) to describe the respective physical properties. The calculated T* is a function of the temperature, which was also found in the literature. To solve this problem, we propose a Boltzmann fitting of the data and obtain T* at the high-temperature limit. With these characteristic parameters, the pressure-volume-temperature (PVT) data of iPP and PEOC are predicted by the SLLFT equation of state. To justify the correctness of our results, we also obtain the PVT data for iPP and PEOC by experiments. Good agreement is found between the two sets of data. By integrating the Euler-Lagrange equation and the Cahn-Hilliard relation, we predict the density profiles and the surface tensions for iPP and PEOC, respectively. Furthermore, a recursive method is proposed to obtain the characteristic interaction energy parameter between iPP and PEOC. This method, which does not require fitting to the experimental phase equilibrium data, suggests an alternative way to predict the phase diagrams that are not easily obtained in experiments.

Electronic structures of 5d transition metal monoxides by density functional theory

Bond distances, vibrational frequencies, electron affinities, ionization potentials, and dissociation energies of the diatomic 5d transition metal (except La) monoxides and their positively and negatively charged ions were studied by use of density functional methods B3LYP, BLYP, B3PW91, BPW91, B3P86, BP86, MPW1PW91, PBE1PBE, and SVWN. Our calculation shows that for each individual species, the calculated properties are quite sensitive to the method used. Compared with hybrid density functional method B3PW91 (B3P86), pure density functional method BPW91 (BP86) gives longer bond distance (lower vibrational frequency) from HfO to PtO for neutral species, HfO+ to IrO+ for cationic species, and HfO- to AuO- for anionic species. While for B3LYP and BLYP, the trend was observed for cationic species from HfO+ to IrO+ and anionic species from HfO- to AuO- (except TaO-), but not for neutrals. Pure density function methods BLYP, BPW91, and BP86 give larger dissociation energy compared with hybrid density functional methods B3LYP, B3PW91, and B3P86. SVWN in most cases gives the smallest bond distance, while BLYP gives the largest value. MPW1PW91 and PBE1PBE show the same performance in predicting the spectroscopic constants.

Pressure effects on the thermodynamics of trans-decahydronaphthalene/polystyrene polymer solutions: Application of the Sanchez-Lacombe lattice fluid theory

The cloud-point temperatures (T-cl's) of trans-decahydronaphthalene(TD)/polystyrene (PS, (M) over bar (w) = 270 000) solutions were determined by light scattering measurements over a range of temperatures (1-16degreesC), pressures (100-900 bar), and compositions (4.2-21.6 vol.-% polymer). The system phase separates upon cooling and T-cl was found to increase with rising pressure for constant composition. In the absence of special effects, this finding indicates positive excess volume for the mixing. Special attention was paid to the demixing temperatures as a function of pressure for different polymer solutions and the plots in the T-phi plane (where phi signifies volume fractions). The cloud-point curves of polymer solutions under different pressures were observed for different compositions, which demonstrated that pressure has a greater effect on the TD/PS solutions when far from the critical point as opposed to near the critical point. The Sanchez-Lacombe lattice fluid theory (SLLFT) was used to calculate the spinodals, the binodals, the Flory-Huggins (FH) interaction parameter, the enthalpy of mixing, and the volume changes of mixing. The calculated results show that modified PS scaling parameters can describe the thermodynamics of the TD/PS system well. Moreover the SLLFT describes the experimental results well.

Phase-separation behavior of the system PES/phenoxy: An application of the Sanchez-Lacombe lattice fluid theory

Cloud-point curves reported for the system polyethersulfone (PES)/phenoxy were calculated by means of the Sanchez-Lacombe (SL) lattice fluid theory. The one adjustable parameter epsilon(12)*/k (quantifying the interaction energy between mers of the different components) can be evaluated by comparison of the theoretical and experimental phase diagrams. The Flory-Huggins (FH) interaction parameters are computed based on the evaluated epsilon(12)*/k and are approximately a linear function of volume fraction and of inverse temperature. The calculated enthalpies of mixing of PES/phenoxy blends for different compositions are consistent with the experimental values obtained previously by Singh and Walsh [1].

波前重建法射线追踪及走时表压缩

As a fast and effective method for approximate calculation of seismic numerical simulation, ray tracing method, which has important theory and practical application value, in terms of seismic theory and seismic simulation, inversion, migration, imaging, simplified from seismic theory according to geometric seismic, means that the main energy of seismic wave field propagates along ray paths in condition of high-frequency asymptotic approximation. Calculation of ray paths and traveltimes is one of key steps in seismic simulation, inversion, migration, and imaging. Integrated triangular grids layout on wavefront with wavefront reconstruction ray tracing method, the thesis puts forward wavefront reconstruction ray tracing method based on triangular grids layout on wavefront, achieves accurate and fast calculation of ray paths and traveltimes. This method has stable and reasonable ray distribution, and overcomes problems caused by shadows in conventional ray tracing methods. The application of triangular grids layout on wavefront, keeps all the triangular grids stable, and makes the division of grids and interpolation of a new ray convenient. This technology reduces grids and memory, and then improves calculation efficiency. It enhances calculation accuracy by accurate and effective description and division on wavefront. Ray tracing traveltime table, which shares the character of 2-D or 3-D scatter data, has great amount of data points in process of seismic simulation, inversion, migration, and imaging. Therefore the traveltime table file will be frequently read, and the calculation efficiency is very low. Due to these reasons, reasonable traveltime table compression will be very necessary. This thesis proposes surface fitting and scattered data compression with B-spline function method, applies to 2-D and 3-D traveltime table compression. In order to compress 2-D (3-D) traveltime table, first we need construct a smallest rectangular (cuboidal) region with regular grids to cover all the traveltime data points, through the coordinate range of them in 2-D surface (3-D space). Then the value of finite regular grids, which are stored in memory, can be calculated using least square method. The traveltime table can be decompressed when necessary, according to liner interpolation method of 2-D (3-D) B-spline function. In the above calculation, the coefficient matrix is stored using sparse method and the liner system equations are solved using LU decomposition based on the multi-frontal method according to the sparse character of the least square method matrix. This method is practiced successfully in several models, and the cubic B-spline function can be the best basal function for surface fitting. It make the construction surface smooth, has stable and effective compression with high approximate accuracy using regular grids. In this way, through constructing reasonable regular grids to insure the calculation efficiency and accuracy of compression and surface fitting, we achieved the aim of traveltime table compression. This greatly improves calculation efficiency in process of seismic simulation, inversion, migration, and imaging.

Tensor density matrix theory for determining population and alignment of symmetric top molecule using rotationally unresolved fluorescence

General expressions used for transforming raw laser-induced fluorescence (LIF) intensity into the population and alignment parameters of a symmetric top molecule are derived by employing the density matrix approach. The molecular population and alignment are described by molecular state multipoles. The results are presented for a general excitation-detection geometry and then applied to some special geometries. In general cases, the LIF intensity is a complex function of the initial molecular state multipoles, the dynamic factors and the excitation-detection geometrical factors. It contains a population and 14 alignment multipoles. How to extract all initial state multipoles from the rotationally unresolved emission LIF intensity is discussed in detail.

Theory for determining alignment parameters of symmetric top molecule using (n+1) LIF

Expressions used for extracting the population and alignment parameters of a symmetric top molecule from (n + 1) laser-induced fluorescence (LIF) are derived by employing the tensor density matrix method. The molecular population and alignment are described by molecular state multipoles. The LIF intensity is a complex function of the initial molecular state multipoles, the dynamic factors, and the excitation-detection geometrical factors. The problem of how to extract the initial molecular state multipoles from (2 + 1) LIF, as an example, is discussed in detail. (C) 2000 American Institute of Physics. [S0021-9606(00)30744-9].