Blood vessel segmentation is an important problem for quantitative structure analysis of retinal images, and many diseases are related to the structure changes. Manual segmentation is time consuming and computer aided...
详细信息
Blood vessel segmentation is an important problem for quantitative structure analysis of retinal images, and many diseases are related to the structure changes. Manual segmentation is time consuming and computer aided segmentation is required to deal with large amount images. This paper presents a new supervised method for segmentation of blood vessels in retinal photographs. Multiple kernel learning (MKL) is introduced to deal with the problem, utilizing features from Hessian matrix based vesselness measure, response of multiscale Gabor filter, and multiple scale line strength features. The method is evaluated on the publicly available DRIVE and STARE databases. The performance of the MKL method is evaluated and experimental results show the high accuracy of the proposed method.
作者：
Liang, HailiSu, HoushengWang, XiaofanChen, Michael Z.Q.[a] Department of Automation
Shanghai Jiao Tong University Key Laboratory of System Control and Information Processing Ministry of Education of China Shanghai 200240 China[b] School of Automation Key Laboratory of Ministry of Education for Image Processing and Intelligent Control Huazhong University of Science and Technology Luoyu Road 1037 Wuhan 430074 China[c] Department of Mechanical Engineering The University of Hong Kong Hong Kong
This paper investigates the problem of swarm aggregations of heterogeneous multi-agent systems. Comparing with the existing studies on swarm aggregations of homogeneous multi-agent systems, this paper is much more res...
详细信息
This paper investigates the problem of swarm aggregations of heterogeneous multi-agent systems. Comparing with the existing studies on swarm aggregations of homogeneous multi-agent systems, this paper is much more resembling the practical situations, where the agents have different dynamics. We show that the heterogeneous agents will gather with a certain error under some assumptions and conditions. The stability properties have been proven by theoretical analysis and verified via numerical simulation. The stability of the heterogeneous multi-agent systems has been achieved based on matrix theory and the Lyapunov stability theorem. Numerical simulation is given to demonstrate the effectiveness of the theoretical result. [PUBLICATION ABSTRACT]
This paper presents a new method to tune the parameters of the adaptation PI controller of full-order flux observer. The method employs an Immune Genetic Algorithm (IGA) based optimization routine that can be implemen...
详细信息
A novel statistical method using path integral Monte Carlo simulation based on quantum mechanics to detect edges of interested objects was proposed in this paper. Our method was inspired by essential characteristics o...
详细信息
This paper is considered with the H ∞ observer design problem for a class of nonlinear systems with the one-sided Lipschitz condition. The systems under consideration include the well-studied Lipschitz system as a sp...
详细信息
This paper is considered with the H ∞ observer design problem for a class of nonlinear systems with the one-sided Lipschitz condition. The systems under consideration include the well-studied Lipschitz system as a special case and possess inherent advantages with respect to conservativeness. For such systems in the presence of noises, we develop a Linear Matrix Inequality (LMI) based approach to design a nonlinear H ∞ observer by carefully dealing with the one-sided Lipschitz condition together with the quadratic inner-bounded condition. The resulting nonlinear H ∞ observer guarantees asymptotic stability of the estimation error dynamics with a prescribed H ∞ performance. Moreover, for the design purpose, the existence condition of the proposed nonlinear H ∞ observer is formulated in terms of LMIs by using a matrix generalized inverse technique. Finally, a simulation example is given to illustrate the effectiveness of the proposed design.
In this paper, we investigate exponential stability of delayed recurrent neural networks. By using the delay partitioning method, some sufficient conditions are established to guarantee exponential stability of delaye...
详细信息
In this paper, we investigate exponential stability of delayed recurrent neural networks. By using the delay partitioning method, some sufficient conditions are established to guarantee exponential stability of delayed recurrent neural networks under two different conditions with constructing new Lyapunov–Krasvoskii functional. This partitioning approach can reduce the conservatism comparing with some previous results of stability. At last, numerical examples are given out to show the effectiveness and advantage of our results.
This paper is concerned with the problems of stability for a class of impulsive positive systems. An impulsive positive system model is introduced for the first time and a necessary and sufficient condition guaranteei...
详细信息
This paper is concerned with the problems of stability for a class of impulsive positive systems. An impulsive positive system model is introduced for the first time and a necessary and sufficient condition guaranteeing the positivity of this kind of system is proposed. Several sufficient criteria of global exponential stability and global asymptotical stability for impulsive positive systems are established respectively by using a linear copositive Lyapunov function. Two numerical examples are given to illustrate the effectiveness and applicability of the proposed results.
Virtual machine(VM) placement is a key technologyto improve data center *** works consider VM placement problem only with respect to physical machine(PM) or network resource ***,efficient VM placement should be implem...
详细信息
Virtual machine(VM) placement is a key technologyto improve data center *** works consider VM placement problem only with respect to physical machine(PM) or network resource ***,efficient VM placement should be implemented by joint optimization of above two *** this paper,a multi-objective VM placement model to minimize the number of active PMs,minimize communication traffic and balance multi-dimensional resource use simultaneously within the data center is *** improved evolutionary multi-objective algorithm: NS-GGA is also designed to tackle this problem,which incorporates the fast nondominated sorting of NSGA-II into the Grouping Genetic *** simulation results show that,in most cases,our model and algorithm gains significantly in all aspects and yields better solutions compared to the existing methods.
A 5-kW dynamic solid oxide fuel cell(SOFC)system with a second air bypass has been developed with a model to perform both steady-state and dynamic analysis in this *** identifies and addresses the control challenges a...
详细信息
A 5-kW dynamic solid oxide fuel cell(SOFC)system with a second air bypass has been developed with a model to perform both steady-state and dynamic analysis in this *** identifies and addresses the control challenges associated with simultaneous power and thermal management under current-based control.
暂无评论