Digits speech recognition based on geometrical learning

We investigate the use of independent component analysis (ICA) for speech feature extraction in digits speech recognition systems.We observe that this may be true for a recognition tasks based on geometrical learning with little training data. In contrast to image processing, phase information is not essential for digits speech recognition. We therefore propose a new scheme that shows how the phase sensitivity can be removed by using an analytical description of the ICA-adapted basis functions via the Hilbert transform. Furthermore, since the basis functions are not shift invariant, we extend the method to include a frequency-based ICA stage that removes redundant time shift information. The digits speech recognition results show promising accuracy, Experiments show method based on ICA and geometrical learning outperforms HMM in different number of train samples.

Enhancement of the far-field output power and the properties of the very-small-aperture lasers

We report on a VSAL structure fabricated by a 650 nm edge emitting laser diode with an Au-coated facet and an aperture size of 250 x 500 nm. The far field output power can maintain at 1 mW and the power density is 7.5 mW/mu m(2). Some properties of the VSAL including the threshold current change, the red-shift of the spectral position, and the strong relative-intensity-noise are presented. The physical mechanisms responsible for these phenomena are also discussed, which may contribute to the understanding and application of the potential device for near-field optics.

Iris recognition algorithm based on point covering of high-dimensional space and neural network

In this paper, we constructed a Iris recognition algorithm based on point covering of high-dimensional space and Multi-weighted neuron of point covering of high-dimensional space, and proposed a new method for iris recognition based on point covering theory of high-dimensional space. In this method, irises are trained as "cognition" one class by one class, and it doesn't influence the original recognition knowledge for samples of the new added class. The results of experiments show the rejection rate is 98.9%, the correct cognition rate and the error rate are 95.71% and 3.5% respectively. The experimental results demonstrate that the rejection rate of test samples excluded in the training samples class is very high. It proves the proposed method for iris recognition is effective.

Application of multi-weighted neuron for iris recognition

In this paper, from the cognition science point of view, we constructed a neuron of multi-weighted neural network, and proposed a new method for iris recognition based on multi-weighted neuron. In this method, irises are trained as "cognition" one class by one class, and it doesn't influence the original recognition knowledge for samples of the new added class. The results of experiments show the correct rejection rate is 98.9%, the correct cognition rate and the error recognition rate are 95.71% and 3.5% respectively. The experimental results demonstrate that the correct rejection rate of the test samples excluded in the classes of training samples is very high. It proves the proposed method for iris recognition is effective.

Performance analysis of very-small-aperture lasers

The fabrication of very-small-aperture lasers is demonstrated, and their performance is analyzed. Because of strong optical feedback caused by a gold film on the front facet of the laser, its behavior changes: The threshold current decreases, the density of light inside the laser diode and the redshift effect of the spectra are enhanced, and the laser diode's lifetime is shorter than that of common laser diodes with large driving current. (c) 2005 Optical Society of America

Effect of fabrication conditions on I-V properties for ZnO varistor with high concentration additives by sol-gel technique

Polycrystalline nano-grain-boundary multi-doping ZnO-based nonlinear varistors with higher concentration additives have been fabricated by sol-gel and standard solid-state reaction method, of which the best sample has a very high threshold voltage of E-b = 3300 V/mm. The effect of sintering processes, sintering temperature and sintering time, and that of additive concentration of Bi2O3 on E-b of the samples are systematically investigated. The results show that the great merit of sol-gel method is its high threshold voltage obtained by a lower sintering temperature than the solid-state reaction method. The present work also shows that five phases including solid-state sintering, rich Bi liquid phase formation and ZnO as well as other additive dissolution, ZnO grain growth, the secondary phase sufficient formation and evolution have been experienced at different sintering temperatures. The hole type defect and nonhomogeneity of the microstructure will lead to the decrease of threshold voltage, i.e., the grain size and the homogeneity of the material will be important factors and directly affect the characteristic of the varistor. The sintering characteristic and the influence of Bi2O3 content on the threshold voltage are also discussed. (c) 2004 Elsevier B.V. All rights reserved.

Manifold covering theory in biomimetic pattern recognition

We describe a new model which is based on the concept of cognizing theory. The method identifies subsets of the data which are embedded in arbitrary oriented lower dimensional space. We definite manifold covering in biomimetic pattern recognition, and study its property. Furthermore, we propose this manifold covering algorithm based on Biomimetic Pattern Recognition. At last, the experimental results for face recognition demonstrates that the correct rejection rate of the test samples excluded in the classes of training samples is very high and effective.

Improved performance in organic light emitting diodes with a mixed electron donor-acceptor film involved in hole injection

Organic light emitting diodes using a mixed layer of electron acceptor 3, 4, 9, 10 perylenetetracarboxylic dianhydride and electron donor copper phthalocyanine (PTCDA:CuPc) on indium tin oxide (ITO) anodes were fabricated. The device properties were found to be strongly dependent on the thickness of the PTCDA:CuPc film: both the power efficiency and the driving voltage of the device were optimized with a thickness of PTCDA:CuPc ranging from 10 to 20 nm. As compared to the conventional ITO/CuPc hole injection structure, the ITO/PTCDA:CuPc hole injection structure could remarkably enhance both the luminance and the power efficiencies of devices. A mechanism of static-induced, very efficient hole-electron pairs generation in mixed PTCDA:CuPc films was proposed to explain the experimental phenomena. The structural and optical properties of PTCDA:CuPc film were examined as well. (c) 2007 American Institute of Physics.

Organic light emitting diodes with an organic acceptor/donor interface involved in hole injection

Organic light emitting diodes with an interface of organic acceptor 3-, 4-, 9-, 10-perylenetetracarboxylic dianhydride (PTCDA) and donor copper phthalocyanine (CuPc) involved in hole injection are fabricated. As compared to the conventional device using a 5 nm CuPc hole injection layer, the device using an interface of 10 nm PTCDA and 5 rim CuPc layers shows much lower operating voltage with an increase of about 46% in the maximum power efficiency. The enhanced device performance is attributed to the efficient hole generation at the PTCDA/CuPc interface. This study provides a new way of designing hole injection.

Photoluminescence from Er-doped Si-in-SiNx thin films

Photoluminescence from Er3+-implanted Si-in-SiN, films emitting efficiently visible light were investigated. A Stark structure in the Er3+ photoluminescence spectrum was observed at room temperature, which reveals more than one site symmetry for the Er3+-centers in the Si-in-SiN, matrix. The correlation between the visible photoluminescence from the silicon nanoparticles and the 1.54 mu m emission from the Er3+-centers was discussed. (c) 2006 Elsevier B.V. All rights reserved.

Blind image restoration based on the theory of high-dimensional space geometry

In practical situations, the causes of image blurring are often undiscovered or difficult to get known. However, traditional methods usually assume the knowledge of the blur has been known prior to the restoring process, which are not practicable for blind image restoration. A new method proposed in this paper aims exactly at blind image restoration. The restoration process is transformed into a problem of point distribution analysis in high-dimensional space. Experiments have proved that the restoration could be achieved using this method without re-knowledge of the image blur. In addition, the algorithm guarantees to be convergent and has simple computation.

Design and characteristics of quantum cascade laser-based CO detection system

National High Technology Research and Development Program of China 2007AA03Z112；Program of Ministry of Education of China 20060183030；Program of Jilin Provincial Science and Technology Department of China 20070709；Program of Bureau of Science and Technology of Changchun City 2007107

Information geometry theory of high-dimension space and application for speaker independent continuous digit speech recognition

In the light of descriptive geometry and notions in set theory, this paper re-defines the basic elements in space such as curve and surface and so on, presents some fundamental notions with respect to the point cover based on the High-dimension space (HDS) point covering theory, finally takes points from mapping part of speech signals to HDS, so as to analyze distribution information of these speech points in HDS, and various geometric covering objects for speech points and their relationship. Besides, this paper also proposes a new algorithm for speaker independent continuous digit speech recognition based on the HDS point dynamic searching theory without end-points detection and segmentation. First from the different digit syllables in real continuous digit speech, we establish the covering area in feature space for continuous speech. During recognition, we make use of the point covering dynamic searching theory in HDS to do recognition, and then get the satisfying recognized results. At last, compared to HMM (Hidden Markov models)-based method, from the development trend of the comparing results, as sample amount increasing, the difference of recognition rate between two methods will decrease slowly, while sample amount approaching to be very large, two recognition rates all close to 100% little by little. As seen from the results, the recognition rate of HDS point covering method is higher than that of in HMM (Hidden Markov models) based method, because, the point covering describes the morphological distribution for speech in HDS, whereas HMM-based method is only a probability distribution, whose accuracy is certainly inferior to point covering.

Mandarin speech emotion recognition based on high dimensional geometry theory

In this paper, a novel approach for mandarin speech emotion recognition, that is mandarin speech emotion recognition based on high dimensional geometry theory, is proposed. The human emotions are classified into 6 archetypal classes: fear, anger, happiness, sadness, surprise and disgust. According to the characteristics of these emotional speech signals, the amplitude, pitch frequency and formant are used as the feature parameters for speech emotion recognition. The new method called high dimensional geometry theory is applied for recognition. Compared with traditional GSVM model, the new method has some advantages. It is noted that this method has significant values for researches and applications henceforth.

Viruses' complete genomes classification based on geometrical learning

In this paper, we proposed a method of classification for viruses' complete genomes based on graph geometrical theory in order to viruses classification. Firstly, a model of triangular geometrical graph was put forward, and then constructed feature-space-samples-graphs for classes of viruses' complete genomes in feature space after feature extraction and normalization. Finally, we studied an algorithm for classification of viruses' complete genomes based on feature-space-samples-graphs. Compared with the BLAST algorithm, experiments prove its efficiency.