This paper proposes a novel algorithm for image partition using irregular region in fractal image compression, which greatly increases the compression ratios achieved over traditional block-based partition. Also, due ...
详细信息
ISBN:
(纸本)0780378652
This paper proposes a novel algorithm for image partition using irregular region in fractal image compression, which greatly increases the compression ratios achieved over traditional block-based partition. Also, due to the large search space involved (transformations and match), Similar Extension algorithms which are described are used to construct the irregular region transformations, and results for Similar Extension algorithms are shown. The results show that the algorithm of irregular region achieves almost double the compression ratio of simple block-based system at a similar decompressed image quality.
A support vector machine (SVM) was originally developed to solve two-class non-fuzzy problems. An SVM can act as a linear learningmachine when handling data in a high dimensional feature space for non-linear separabl...
详细信息
A support vector machine (SVM) was originally developed to solve two-class non-fuzzy problems. An SVM can act as a linear learningmachine when handling data in a high dimensional feature space for non-linear separable and non-separable problems. A few methods have been proposed to solve two-class and multi-class classification problems by including fuzzy concepts. In this paper, we propose a new fuzzy support vector machine which improves the traditional SVM by adding fuzzy memberships to each training sample to indicate degree of membership of this sample to different classes. This fuzzy SVM is more complete and meaningful, and could generalize the traditional non-fuzzy SVM to a fuzzy one, i.e., the traditional non-fuzzy SVM is an extreme case of our fuzzy SVM when the degrees of membership of a sample to two different classes are the same.
Most of the present classifying methods are based on VSM (the vector space model), of which the widely used method is kNN (the k-nearest neighbors). But most of them are highly complicated on computation, and cannot b...
详细信息
ISBN:
(纸本)0780378652
Most of the present classifying methods are based on VSM (the vector space model), of which the widely used method is kNN (the k-nearest neighbors). But most of them are highly complicated on computation, and cannot be used on the occasion of classifying a large number of specimen and the classifier must be rebuilt to increment the training corpora in order to have tough scalability. Two new notions, mutual dependence and equivalent radius, are put forward in this paper. And then a new classifying algorithm based on the two notions, SECTILE is offered in this paper. Later SECTILE is applied to classifying Chinese documents and compared to kNN and CCC methods. The experimental results suggests that SECTILE outperforms kNN and CCC methods, and can be used online to classify a large number of specimen and has good scalability, while the precision and recall of classification are kept.
After approximating the nonlinear of the system used in the analysis of the voltage instability problem, the components models in the system are built. And based these models, it is pointed out that voltage instabilit...
详细信息
ISBN:
(纸本)0780378652
After approximating the nonlinear of the system used in the analysis of the voltage instability problem, the components models in the system are built. And based these models, it is pointed out that voltage instability problem incorporate not only nonlinear dynamics but also discrete events and some of the manipulated variables are discrete, too. Therefore, the characteristics of the problem are hybrid in the nature. Voltage instability problem can de described effectively by hybrid system model. And it is proposed to analyze the problem by the hybrid system of differential, switched algebraic, and state-reset (DSAR) model.
A neural-network-based iterative controller is presented focusing on the speed control of ultrasonic motors. Suitable ranges of the adaptive learning rates are presented through the theoretical analysis on the propose...
详细信息
A neural-network-based iterative controller is presented focusing on the speed control of ultrasonic motors. Suitable ranges of the adaptive learning rates are presented through the theoretical analysis on the proposed model, which could guarantee the fastest convergence of the neural network controller. Numerical results show that the neural-network-based controller is effective for various kinds of reference speeds of ultrasonic motors. Comparisons with the existing method show that the precision of control could be increased using the proposed method. Simulations also show that the proposed scheme is fairly robust against random disturbance to the control variables.
Online signature verification is a new and active topic in the research and application fields of biometrics. This paper proposes a low cost online signature verification method based on the matching of curves about x...
详细信息
Online signature verification is a new and active topic in the research and application fields of biometrics. This paper proposes a low cost online signature verification method based on the matching of curves about x- and y-axis attaching some dynamic features. Just as human being's behavior, different local weight and unfixed threshold are introduced to improve the performance of signature verification system. Finally, this paper presents some applications of online signature verification on ATM, onsite credit card verification and Internet e-commerce.
In the distributed interactive simulation based on HLA/RTI, a key problem is how to establish information exchange and delivery mechanism among federates effectively to meet the requirements of system scalability. Thi...
详细信息
ISBN:
(纸本)0780378652
In the distributed interactive simulation based on HLA/RTI, a key problem is how to establish information exchange and delivery mechanism among federates effectively to meet the requirements of system scalability. This paper discusses the basic data filtering mechanism of RTI1.3, then a new data distribution strategy is proposed: this algorithm computes region matching using grid-based method firstly, then it allocates multicast groups based on publisher dynamically. The simulation result show that the strategy proposed could reduce complexity of region matching and make use of multicast groups effectively.
To implement a XML data warehouse system, a key issue is how to efficiently maintain materialized views, to preserve their consistency with data sources when base data changes. Self-maintenance of views has been shown...
详细信息
ISBN:
(纸本)0780378652
To implement a XML data warehouse system, a key issue is how to efficiently maintain materialized views, to preserve their consistency with data sources when base data changes. Self-maintenance of views has been shown to dramatically improve performance for view maintenance, and studied extensively in the context of the relational data model. But for XML database, there are still lots of work left to be done. In this paper, we give a preliminary result on self-maintainability of deletions of views over XML data. We give a necessary and sufficient condition of self-maintainability of deletions, and an algorithm to implement self-maintenance of deletions of views for XML data.
In this paper, we propose a new method to extract boundary based on the combination of stack filter, Hopfield neural network and self-organization neural network in order to get all the advantages of the three methods...
详细信息
ISBN:
(纸本)0780378652
In this paper, we propose a new method to extract boundary based on the combination of stack filter, Hopfield neural network and self-organization neural network in order to get all the advantages of the three methods. Compared to the boundary extraction method based on Hopfield neural network, the new method has a stronger ability to resist mixed-distributed noises and the result from boundary test is much better. The speed of optimal training on the stack filter is improved greatly and memory needed is decreased dramatically compared to that based on stack filter.
This paper investigates the problem of how to carry out 3D scene reconstruction from multiple views. It improves the algorithms of projective reconstruction based on the homography induced by the infinite plane which ...
详细信息
ISBN:
(纸本)0780378652
This paper investigates the problem of how to carry out 3D scene reconstruction from multiple views. It improves the algorithms of projective reconstruction based on the homography induced by the infinite plane which should have 4 points on a reference plane visible in all views given by Hartley and Rother et al., and proposes a new linear algorithm based on 3 points on a reference plane visible in all views. It avoids the difficult task of determining whether 4 object points are coplanar or not, because 3 points which are not collinear just determine a plane.
暂无评论