Parameters Inversion of Fluid-Saturated Porous Media Based on Neural Networks

The multi-layers feedforward neural network is used for inversion of material constants of fluid-saturated porous media. The direct analysis of fluid-saturated porous media is carried out with the boundary element method. The dynamic displacement responses obtained from direct analysis for prescribed material parameters constitute the sample sets training neural network. By virtue of the effective L-M training algorithm and the Tikhonov regularization method as well as the GCV method for an appropriate selection of regularization parameter, the inverse mapping from dynamic displacement responses to material constants is performed. Numerical examples demonstrate the validity of the neural network method.

A novel FOCAL technique based on BP-ANN

In this paper, the feed-forward back-propagation artificial neural network (BP-ANN) algorithm is introduced in the traditional Focus Calibration using Alignment procedure (FOCAL) technique, and a novel FOCAL technique based on BP-ANN is proposed. The effects of the parameters, such as the number of neurons on the hidden-layer and the number of training epochs, on the measurement accuracy are analyzed in detail. It is proved that the novel FOCAL technique based on BP-ANN is more reliable and it is a better choice for measurement of the image quality parameters. (c) 2005 Elsevier GmbH. All rights reserved.

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.

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.

Research on continuous speech recognition with small vocabulary based on biomimetic pattern recognition

Based on biomimetic pattern recognition theory, we proposed a novel speaker-independent continuous speech keyword-spotting algorithm. Without endpoint detection and division, we can get the minimum distance curve between continuous speech samples and every keyword-training net through the dynamic searching to the feature-extracted continuous speech. Then we can count the number of the keywords by investigating the vale-value and the numbers of the vales in the curve. Experiments of small vocabulary continuous speech with various speaking rate have got good recognition results and proved the validity of the algorithm.

Cancer subtypes recognition and its gene expression profiles analysis based on geometrical learning

The accurate recognition of cancer subtypes is very significant in clinic. Especially, the DNA microarray gene expression technology is applied to diagnosing and recognizing cancer types. This paper proposed a method of that recognized cancer subtypes based on geometrical learning. Firstly, the cancer genes expression profiles data was pretreated and selected feature genes by conventional method; then the expression data of feature genes in the training samples was construed each convex hull in the high-dimensional space using training algorithm of geometrical learning, while the independent test set was tested by the recognition algorithm of geometrical learning. The method was applied to the human acute leukemia gene expression data. The accuracy rate reached to 100%. The experiments have proved its efficiency and feasibility.

A novel face recognition method based on hypersausage neural networks

In this paper, we firstly give the nature of 'hypersausages', study its structure and training of the network, then discuss the nature of it by way of experimenting with ORL face database, and finally, verify its unsurpassable advantages compared with other means.

Keyword spotting system based on biomimetic pattern recognition

The mandarin keyword spotting system was investigated, and a new approach was proposed based on the principle of homology continuity and point location analysis in high-dimensional space geometry theory which are both parts of biomimetic pattern recognition theory. This approach constructed a hyper-polyhedron with sample points in the training set and calculated the distance between each test point and the hyper-polyhedron. The classification resulted from the value of those distances. The approach was tested by a speech database which was created by ourselves. The performance was compared with the classic HMM approach and the results show that the new approach is much better than HMM approach when the training data is not sufficient.

Continuous speech recognition based on ICA and 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 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. 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 that the method based on ICA and Geometrical Learning outperforms HMM in a different number of training samples.

Structure approximate based on double weighted neural network

Double weighted neural network; is a kind of new general used neural network, which, compared with BP and RBF network, may approximate the training samples with a move complicated geometric figure and possesses a even greater approximation. capability. we study structure approximate based on double weighted neural network and prove its rationality.

Object-recognition with oblique observation directions based on biomimetic pattern recognition

In this paper, we propose a new scheme for omnidirectional object-recognition in free space. The proposed scheme divides above problem into several onmidirectional object-recognition with different depression angles. An onmidirectional object-recognition system with oblique observation directions based on a new recognition theory-Biomimetic Pattern Recognition (BPR) is discussed in detail. Based on it, we can get the size of training samples in the onmidirectional object-recognition system in free space. Omnidirection ally cognitive tests were done on various kinds of animal models of rather similar shapes. For the total 8400 tests, the correct recognition rate is 99.89%. The rejection rate is 0.11% and on the condition of zero error rates. Experimental results are presented to show that the proposed approach outperforms three types of SVMs with either a three degree polynomial kernel or a radial basis function kernel.

Continuous speech recognition based on ICA and 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 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. 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 that the method based on ICA and Geometrical Learning outperforms HMM in a different number of training samples.

Fast Discrimination of Commerical Corn Varieties Based on Near Infrared Spectra

The existing methods for the discrimination of varieties of commodity corn seed are unable to process batch data and speed up identification, and very time consuming and costly. The present paper developed a new approach to the fast discrimination of varieties of commodity corn by means of near infrared spectral data. Firstly, the experiment obtained spectral data of 37 varieties of commodity corn seed with the Fourier transform near infrared spectrometer in the wavenurnber range from 4 000 to 12 000 cm (1). Secondly, the original data were pretreated using statistics method of normalization in order to eliminate noise and improve the efficiency of models. Thirdly, a new way based on sample standard deviation was used to select the characteristic spectral regions, and it can search very different wavenumbers among all wavenumbers and reduce the amount of data in part. Fourthly, principal component analysis (PCA) was used to compress spectral data into several variables, and the cumulate reliabilities of the first ten components were more than 99.98%. Finally, according to the first ten components, recognition models were established based on BPR. For every 25 samples in each variety, 15 samples were randomly selected as the training set. The remaining 10 samples of the same variety were used as the first testing set, and all the 900 samples of the other varieties were used as the second testing set. Calculation results showed that the average correctness recognition rate of the 37 varieties of corn seed was 94.3%. Testing results indicate that the discrimination method had higher precision than the discrimination of various kinds of commodity corn seed. In short, it is feasible to discriminate various varieties of commodity corn seed based on near infrared spectroscopy and BPR.

A New Discrimination Method of Maize Seed Varieties Based on Near-Infrared Spectroscopy

A new discrimination method for the maize seed varieties based on the near-infrared spectroscopy was proposed. The reflectance spectra of maize seeds were obtained by a FT-NIR spectrometer (12 000-4 000 cm(-1)). The original spectra data were preprocessed by first derivative method. Then the principal component analysis (PCA) was used to compress the spectra data. The principal components with the cumulate reliabilities more than 80% were used to build the discrimination models. The model was established by Psi-3 neuron based on biomimetic pattern recognition (BPR). Especially, the parameter of the covering index was proposed to assist to discriminating the variety of a seed sample. The authors tested the discrimination capability of the model through four groups of experiments. There were 10, 18, 26 and 34 varieties training the discrimination models in these experiments, respectively. Additionally, another seven maize varieties and nine wheat varieties were used to test the capability of the models to reject the varieties not participating in training the models. Each group of the experiment was repeated three times by selecting different training samples at random. The correct classification rates of the models in the four-group experiments were above 91. 8%. The correct rejection rates for the varieties not participating in training the models all attained above 95%. Furthermore, the performance of the discrimination models did not change obviously when using the different training samples. The results showed that this discrimination method can not only effectively recognize the maize seed varieties, but also reject the varieties not participating in training the model. It may be practical in the discrimination of maize seed varieties.

scenario-based interactive intention understanding in pen-based user interfaces

Interactive intention understanding is important for Pen-based User Interface (PUI). Many works on this topic are reported, and focus on handwriting or sketching recognition algorithms at the lexical layer. But these algorithms cannot totally solve the problem of intention understanding and can not provide the pen-based software with high usability. Hence, a scenario-based interactive intention understanding framework is presented in this paper, and is used to simulate human cognitive mechanisms and cognitive habits. By providing the understanding environment supporting the framework, we can apply the framework to the practical PUI system. The evaluation of the Scientific Training Management System for the Chinese National Diving Team shows that the framework is effective in improving the usability and enhancing the intention understanding capacity of this system.