A shell for signal interpretation expert systems

System monitoring and diagnosis through signal interpretation are important tasks in many engineering fields. Although monitoring or diagnosis expert systems help human experts a lot, it is not so easy to build a signal interpretation expert system. We have developed two expert systems for speech signal processing. The close analysis of the two systems revealed common functions and modules applicable to wide range of signal interpretation problems. On the basis of the experience, we have developed a shell for signal interpretation expert systems named SSI. In order to evaluate SSI, we also developed two expert systems by using it and obtained satisfactory results. This paper describes the design philosophy and architecture of SSI together with some examples.     

Sintering of fibrous barium titanate powder compacts with preferred orientation

The sintering of fibrous BaTiO₃ powder particles was investigated. Special emphasis was given to the role of particle orientation in the compact on densification and microstructure development. Compacts were made by dry-pressing. During the initial stage of sintering, the fibrous particles rearranged and bundles of particles were formed. The volume of pores between bundles of particles decreased on further heating. Grain growth started when the sintered density reached ca. 56% of the theoretical density. Higher temperatures of sintering increased the degree of the crystal axis orientation. Thus, highly orientated sintered bodies with high densities were prepared by heating at 1500 °C.     

Average-case analysis of the Modified Harmonic algorithm

In this paper we analyze the average-case performance of the Modified Harmonic algorithm for on-line bin packing. We first analyze the average-case performance for arbitrary distribution of item sizes over (0,1]. This analysis is based on the following result. Letf ₁ andf ₂ be two linear combinations of random variables {N _i } _i=1 ^k where theN _i 's have a joint multinomial distribution for eachn= _i=1 ^k ,N _i . LetE(f ₁) O andE(f ₂) 0. Then lim_n E(max(_{f 1},f ₂ ))/n = lim _n max(E(f ₁),E(f ₂))/n. We then consider the special case when the item sizes are uniformly distributed over (0,1]. For specific values of the parameters, the Modified Harmonic algorithm turns out to be better than the other two linear-time on-line algorithms—Next Fit and Harmonic—in both the worst case as well as the average case. We also obtain optimal values for the parameters of the algorithm from the average-case standpoint. For these values of the parameters, the average-case performance ratio is less than 1.19. This compares well with the performance ratios 1.333. and 1.2865. of the Next Fit algorithm and the Harmonic algorithm, respectively.     

Average-case analysis of the Modified Harmonic algorithm

In this paper we analyze the average-case performance of the Modified Harmonic algorithm for on-line bin packing. We first analyze the average-case performance for arbitrary distribution of item sizes over (0,1]. This analysis is based on the following result. Letf ₁ andf ₂ be two linear combinations of random variables {N _i } _i=1 ^k where theN _i 's have a joint multinomial distribution for eachn=σ _i=1 ^k ,N _i . LetE(f ₁) ≠ O andE(f ₂)≠ 0. Then lim_{n →}∞E(max(_{f 1},f ₂ ))/n = lim _n →∞ max(E(f ₁),E(f ₂))/n. We then consider the special case when the item sizes are uniformly distributed over (0,1]. For specific values of the parameters, the Modified Harmonic algorithm turns out to be better than the other two linear-time on-line algorithms—Next Fit and Harmonic—in both the worst case as well as the average case. We also obtain optimal values for the parameters of the algorithm from the average-case standpoint. For these values of the parameters, the average-case performance ratio is less than 1.19. This compares well with the performance ratios 1.333. and 1.2865. of the Next Fit algorithm and the Harmonic algorithm, respectively.     

Numerical solutions of the singular integral equations in the crack analysis using the body force method

In this paper, numerical solutions of the singular integral equations of the body force method in the crack problems are discussed. The stress fields induced by two kinds of displacement discontinuity are used as fundamental solutions. Then, the problem is formulated as a hypersingular integral equation with the singularity of the form r ². In the numerical calculation, two kinds of unknown functions are approximated by the products of the fundamental density function and the Chebyshev polynomials. As examples, the stress intensity factors of the oblique edge crack, kinked crack, branched crack and zig-zag crack are analyzed. The calculation shows that the present method gives accurate results even for the extremely oblique edge crack and kinked crack with extremely short bend which has been difficult to analyze by the previous method using the approximation by the products of the fundamental density function and the stepped functions etc.     

The Effect of Small Additions of Fe and Heavy Deformation on the Precipitation in an Al–1.1Mg–0.5Cu–0.3Si At. Pct Alloy

Metallurgical and Materials Transactions A - The effect of 0.03 and 0.08 at. pct Fe additions on the formation of secondary phases in an Al–1.1Mg–0.5Cu–0.3Si at. pct alloy was...     

Gas permeation property of polyaniline films

We examined the influence of polyaniline (PAn)'s unit sequence and doping with low molecular weight dopants or polymer dopants on permeation property. It was found that CO₂ permeability was increased by the formation of a quinonediimine unit in PAn with the oxidation. CO₂ sorption amount of PAn was decreased by oxidation. The increase of CO₂ permeability with oxidation, therefore, resulted from the increase of diffusivity, which was attributable to morphological variation by the increase of a quinonediimine unit. The permselectivity of PAn films was found to be remarkably improved by doping. In particular, the selectivity value of the PAn film doped with polyvinyl sulphonic acid as a polymer dopant went up to over 2,000. This remarkable increase of selectivity was found to result in the increase of selectivity, depending on diffusivity. It was also found that the permselectivity of the PAn film doped by polymer dopants was surpassed, as compared with that doped by low molecular dopants. © 1995 John Wiley & Sons, Inc.     

HMM-based state classification of a user with a walking support system using visual PCA features

The improvement of safety and dependability in systems that physically interact with humans requires investigation with respect to the possible states of the user’s motion and an attempt to recognize these states. In this study, we propose a method for real-time visual state classification of a user with a walking support system. The visual features are extracted using principal component analysis and classification is performed by hidden Markov models, both for real-time fall detection (one-class classification) and real-time state recognition (multi-class classification). The algorithms are used in experiments with a passive-type walker robot called “RT Walker” equipped with servo brakes and a depth sensor (Microsoft Kinect). The experiments are performed with 10 subjects, including an experienced physiotherapist who can imitate the walking pattern of the elderly and people with disabilities. The results of the state classification can be used to improve fall-prevention control algorithms for walking support systems. The proposed method can also be used for other vision-based classification applications, which require real-time abnormality detection or state recognition.     

Face relighting using discriminative 2D spherical spaces for face recognition

As part of the face recognition task in a robust security system, we propose a novel approach for the illumination recovery of faces with cast shadows and specularities. Given a single 2D face image, we relight the face object by extracting the nine spherical harmonic bases and the face spherical illumination coefficients by using the face spherical spaces properties. First, an illumination training database is generated by computing the properties of the spherical spaces out of face albedo and normal values estimated from 2D training images. The training database is then discriminately divided into two directions in terms of the illumination quality and light direction of each image. Based on the generated multi-level illumination discriminative training space, we analyze the target face pixels and compare them with the appropriate training subspace using pre-generated tiles. When designing the framework, practical real-time processing speed and small image size were considered. In contrast to other approaches, our technique requires neither 3D face models nor restricted illumination conditions for the training process. Furthermore, the proposed approach uses one single face image to estimate the face albedo and face spherical spaces. In this work, we also provide the results of a series of experiments performed on publicly available databases to show the significant improvements in the face recognition rates.     

Studies on relationship between activity and electron density on carbonyl oxygen in sex pheromone mimics of the American cockroach,part XI

From the relationship between structures of the 2-substituents of verbanyl analogs and their sex pheromone activities to the American cockroach, electron density of the carbonyl oxygen atom in the substituent, as estimated by the [¹⁷O]NMR chemical shift, was estimated to be an important factor which influenced the activity, in addition to length of the substituent and the position of the carbonyl group. (+)-Verbanyl methylcarbonate (XX), possessing the highest electron density on the carbonyl oxygen atom, showed the strongest activity among the analogs.For Part X, see Manabe et al., 1983. J. Chem. Ecol. 9:533–549.