A constrained Kalman filtering algorithm based on auditory masking threshold is proposed for enhancing speech degraded by colored noise. The auditory masking threshold is used as a constraint to ob
A constrained Kalman filtering algorithm based on auditory masking threshold is proposed for enhancing speech degraded by colored noise. The auditory masking threshold is used as a constraint to ob
A dual Unscented Kalman Filter (UKF) is applied to enhance speech signals degraded by colored noise. The speech model is a nonlinear one and is modeled by a neural network. The state vector and the
A dual Unscented Kalman Filter (UKF) is applied to enhance speech signals degraded by colored noise. The speech model is a nonlinear one and is modeled by a neural network. The state vector and the
In this paper, we demonstrate through a chemistry experiment that conformational addressing can be achieved using the hairpin structure of a DNA molecule. The hairpin structure made by single-strand DNA (ssDNA) self-h...
详细信息
The MPEG-4 AAC introduces a new tool called Perceptual Noise Substitution (PNS) whose function is to use locally generated noise to replace the noise inside the coded music. The key issue to the PNS tool relies on the...
详细信息
作者：
刘刚敬忠良孙韶媛李建勋李振华Henry LeungInstitute of Aerospace Information and Control
School of Electronics and Information Technology Shanghai Jiaotong University Shanghai 200030 ICT 451
Department of Electrical and Computer Engineering University of Calgary 2500 University Drive N. W. Calgary Alberta Canada T2N 1N4n this paper a novel image fusion method based on the expectation maximization (EM) algorithm and steerable pyramid is proposed. The registered images are first decomposed by using steerable pyramid. The EM algorithm is used to fuse the image components in the low frequency band. The selection method involving the informative importance measure is applied to those in the high frequency band. The final fused image is then computed by taking the inverse transform on the composite coefficient representations. Experimental results show that the proposed method outperforms conventional image fusion methods.
In this paper, a novel image fusion method based on the expectation maximization (EM) algorithm and steerable pyramid is proposed. The registered images are first decomposed by using steerable pyramid. The EM algorith...
详细信息
In this paper, a novel image fusion method based on the expectation maximization (EM) algorithm and steerable pyramid is proposed. The registered images are first decomposed by using steerable pyramid. The EM algorithm is used to fuse the image components in the low frequency band. The selection method involving the informative importance measure is applied to those in the high frequency band. The final fused image is then computed by taking the inverse transform on the composite coefficient representations. Experimental results show that the proposed method outperforms conventional image fusion methods.
作者：
阚海斌沈鸿Department of Computer Science and Engineering
Fudan UniversityShanghai 200433China Graduate School and Information ScienceJapan Advanced Institute of Science and Technology1-1AsahidaiTatsunokuchiIshikawa 923-1292Japan Graduate School and Information Science
Japan Advanced Institute of Science and Technology1-1AsahidaiTatsunokuchiIshikawa923-1292Japan
In this paper, we firstly generalize the relations among the basis vectors of LLL reduced basis to semi k-reduced basis. Then we analyze the complexities of the nearest plane algorithm and round-off algorithm on semi ...
详细信息
In this paper, we firstly generalize the relations among the basis vectors of LLL reduced basis to semi k-reduced basis. Then we analyze the complexities of the nearest plane algorithm and round-off algorithm on semi k-reduced basis, which, compared with L. Babai's results on LLL reduced basis, have better approximate ratios and contain almost the same time complexities.
作者：
Iida, FumiyaPfeifer, RolfSteels, LucKuniyoshi, YasuoUniversity of Zurich
Artificial Intelligence Laboratory Department of Informatics Andreasstr. 15 Zurich Switzerland Vrije Universiteit Brussel
Artificial Intelligence Laboratory and Sony Computer Science Laboratory Department of Informatics Pleinlaan 2 Brussels Belgium University of Tokyo
School of Information Science and Technology Dept. of Mechano-Informatics Laboratory for Intelligent Systems and Informatics Engineering Bldg. 8 7-3-1 Hongo Bunkyo-kuTokyo Japan
Deriving the class structure of object-oriented software has been studied intensively. We have proposed a methodology to divide the conceptual model used in the object-oriented analysis into basic elements, such as cl...
详细信息
ISBN:
(纸本)9729881618
Deriving the class structure of object-oriented software has been studied intensively. We have proposed a methodology to divide the conceptual model used in the object-oriented analysis into basic elements, such as classes, attributes, methods, relations, and to define constraint characteristics and constructing operations on each element. In the methodology, we have applied the field theory in the quantum physics to software and proposed the software field concepts (Ohki and Kambayashi, 2002a). Our thesis is that software is a kind of fields in which software elements, such as methods and attributes, interact each other to produce certain behavioral patterns. The methodology explains well the characteristics of class libraries (Ohki and Kambayashi, 2002b). Once the software elements are extracted from the software field, the methodology allows constructing design patterns from the characteristics of the elements (Ohki and Kambayashi, 2002a). Although we defined the extract operations to elicit the software elements, we failed to show that those operations have reasons and are correct (Ohki and Kambayashi, 2002a). In order to overcome this problem, in this paper, we introduce the distribution functions to represent the software elements, and to formulate the interactions of the functions. Using the distribution functions and the interactions between them, we have succeeded to suggest how to extract the software elements from the software field, and how to derive the design patterns by using the characteristics of the extract elements. This paper first describes the basic concepts of the software field, and then introduces the distribution functions to represent the software elements. In the latter part of this paper describes that it is applicable to derive typical design patterns.
This paper presents a novel predictive information criterion (PIC) for hidden Markov model (HMM) selection. The PIC criterion is exploited to select the best HMMs, which provide the largest prediction information for ...
详细信息
This paper presents a novel predictive information criterion (PIC) for hidden Markov model (HMM) selection. The PIC criterion is exploited to select the best HMMs, which provide the largest prediction information for generalization of future data. When the randomness of HMM parameters is expressed by a product of conjugate prior densities, the prediction information is derived without integral approximation. In particular, a multivariate t distribution is attained to characterize the prediction information corresponding to HMM mean vector and precision matrix. When performing HMM selection in tree structure HMMs, we develop a top-down prior/posterior propagation algorithm for estimation of structural hyperparameters. The prediction information is accordingly determined so as to choose the best HMM tree model. The parameters of chosen HMMs can be rapidly computed via maximum a posteriori (MAP) estimation. In the evaluation of continuous speech recognition using decision tree HMMs, the PIC model selection criterion performs better than conventional maximum likelihood and minimum description length criteria in building a compact tree structure with moderate tree size and higher recognition rate.
暂无评论