Email alert
2001 Vol. 23, No. 1
column
Display Method:
2001, 23(1): 1-8.
Abstract:
The filtering scales of different edge models axe different in edge detection. The effective scales of different edge models based on compact,symmetry and smooth function are discussed in detail. The effective scale ranges of Gauss function and cubic B spline function to restrain noise are analyzed also. The conclusions of this paper are important in multiscale edge detection, and are the theoretic base of scale choice in edge detection.
The filtering scales of different edge models axe different in edge detection. The effective scales of different edge models based on compact,symmetry and smooth function are discussed in detail. The effective scale ranges of Gauss function and cubic B spline function to restrain noise are analyzed also. The conclusions of this paper are important in multiscale edge detection, and are the theoretic base of scale choice in edge detection.
2001, 23(1): 9-15.
Abstract:
The definition, property and time-domain expression of random fractal signal are introduced in brief. The fractional difference equation of random fractal series is induced by substituting difference operator for differential operator, and the fractional differenced whiten-ing filter, through which random fractal signal is converted to white noise, is analyzed and constructed. The method for parameters estimation based on fractional differenced whitening filter is proposed, the essentiality of which is to find out a factor by calculating the extremum of the statistic test expression for the output. Experimental results demonstrate its effectiveness.
The definition, property and time-domain expression of random fractal signal are introduced in brief. The fractional difference equation of random fractal series is induced by substituting difference operator for differential operator, and the fractional differenced whiten-ing filter, through which random fractal signal is converted to white noise, is analyzed and constructed. The method for parameters estimation based on fractional differenced whitening filter is proposed, the essentiality of which is to find out a factor by calculating the extremum of the statistic test expression for the output. Experimental results demonstrate its effectiveness.
2001, 23(1): 16-22.
Abstract:
A new signal processing algorithm based on a local nonlinear prediction modei (LNLP) is presented. With the same implementation, the speech predicting and coding based on LNLP gives improved performance and the computational efforts for LNLP analysis is not increased over that for the conventional algorithms.
A new signal processing algorithm based on a local nonlinear prediction modei (LNLP) is presented. With the same implementation, the speech predicting and coding based on LNLP gives improved performance and the computational efforts for LNLP analysis is not increased over that for the conventional algorithms.
2001, 23(1): 23-29.
Abstract:
The multi-level LUT method for VLD is a fast, memory efficient method and is widely implemented in digital video and image decoders. This paper discusses the design and optimization for multi-level LUT and its state-machine in details, provides a formula for calculating the items of LUT and FSM states, and then proves the algorithm with an experiment of MPEG1. According to the result of this paper, the items in LUT can be fully saved in hardware design.
The multi-level LUT method for VLD is a fast, memory efficient method and is widely implemented in digital video and image decoders. This paper discusses the design and optimization for multi-level LUT and its state-machine in details, provides a formula for calculating the items of LUT and FSM states, and then proves the algorithm with an experiment of MPEG1. According to the result of this paper, the items in LUT can be fully saved in hardware design.
2001, 23(1): 30-36.
Abstract:
The paper analyses the influence of scale factor on discrete moment invariant characteristics. A modified attributive center characteristics with scale, translation and rotating invariances are proposed. The moment invariant characteristics are used in target recognition. Finally through simulation, its recognition rate with that before modification is compared. The results show this way is feasible.
The paper analyses the influence of scale factor on discrete moment invariant characteristics. A modified attributive center characteristics with scale, translation and rotating invariances are proposed. The moment invariant characteristics are used in target recognition. Finally through simulation, its recognition rate with that before modification is compared. The results show this way is feasible.
2001, 23(1): 37-44.
Abstract:
The two Fuzzy C-Spherical Shell (FCSS) clustering algorithms based on the same object function are analyzed and a new algorithm is proposed. The properties of the new algorithm in both the speed and the accuracy are appropriate.
The two Fuzzy C-Spherical Shell (FCSS) clustering algorithms based on the same object function are analyzed and a new algorithm is proposed. The properties of the new algorithm in both the speed and the accuracy are appropriate.
2001, 23(1): 45-52.
Abstract:
In this paper, global asymptotic stability for cellular neural networks with time delay is discussed using a novel Liapunov function. Some novel sufficient conditions for global asymptotic stability are obtained. Those results are simple and practical than given by P. P. Civalleri et al, and have a leading importance to design cellular neural networks with time delay.
In this paper, global asymptotic stability for cellular neural networks with time delay is discussed using a novel Liapunov function. Some novel sufficient conditions for global asymptotic stability are obtained. Those results are simple and practical than given by P. P. Civalleri et al, and have a leading importance to design cellular neural networks with time delay.
2001, 23(1): 53-59.
Abstract:
In this paper, the service performance model for Web system based on HTTP is presented, in which the Web server and the communication network are considered as one system to construct an open Markov queue network. This model is also extended to multiple- server system. After analysing the effects of the parameters on the performance, the method to avoid deadlock and the speed rule to add new server are investigated. Finally, several schemes for improving Web server performance based on different congestion conditions are evaluated and compared. Simulation results confirmed the validity of the model.
In this paper, the service performance model for Web system based on HTTP is presented, in which the Web server and the communication network are considered as one system to construct an open Markov queue network. This model is also extended to multiple- server system. After analysing the effects of the parameters on the performance, the method to avoid deadlock and the speed rule to add new server are investigated. Finally, several schemes for improving Web server performance based on different congestion conditions are evaluated and compared. Simulation results confirmed the validity of the model.
2001, 23(1): 60-67.
Abstract:
In this paper, composite and degenerated multi-output feedforward functions are discussed and some cryptographic properties of them are given.
In this paper, composite and degenerated multi-output feedforward functions are discussed and some cryptographic properties of them are given.
2001, 23(1): 68-74.
Abstract:
This paper theoretically presents the geocoding principle and method for airborne SAR system image, which is converted only by the positional and directional data from the aircraft mounted SAR. The calculated equations and error functions of the geocoding SAR system are derived and the differences of the geocoding SAR image converted by INS data, GPS data or INS/GPS combined data are also analyzed. Besides, the simulated results are suggested in this paper.
This paper theoretically presents the geocoding principle and method for airborne SAR system image, which is converted only by the positional and directional data from the aircraft mounted SAR. The calculated equations and error functions of the geocoding SAR system are derived and the differences of the geocoding SAR image converted by INS data, GPS data or INS/GPS combined data are also analyzed. Besides, the simulated results are suggested in this paper.
2001, 23(1): 75-80.
Abstract:
Based on the transmission function theory for voltage-mode CMOS circuits and an algebra for current-mode CMOS multivalued circuits, this paper presents an algebraic theory for hybrid CMOS multivalued circuits and uses it to design quaternary decoder and encoder. The design result shows the design based on this algebraic theory is better than the traditional design.
Based on the transmission function theory for voltage-mode CMOS circuits and an algebra for current-mode CMOS multivalued circuits, this paper presents an algebraic theory for hybrid CMOS multivalued circuits and uses it to design quaternary decoder and encoder. The design result shows the design based on this algebraic theory is better than the traditional design.
2001, 23(1): 81-84.
Abstract:
This paper proposes a novel cathode-on-membrane vacuum microelectronic pressure sensor. The cathode-on-membrane structure has been fabricated. The properties of the silicon membrane under pressure are simulated, and the effects of the parameters of membrame structure on the sensitivity are discussed.
This paper proposes a novel cathode-on-membrane vacuum microelectronic pressure sensor. The cathode-on-membrane structure has been fabricated. The properties of the silicon membrane under pressure are simulated, and the effects of the parameters of membrame structure on the sensitivity are discussed.
2001, 23(1): 88-91.
Abstract:
A new image fusion algorithm is presented in this paper. This algorithm is used to implement fusion of SAR image and optical image,on the basis of the reciprocal characteristic of SAR image and optical image. The algorithm of image fusion presented in the paper uses a multiscale analysis and the singularity detection principle of wavelet transformation of SAR irnage and optical image, achieves a better result in experiment.
A new image fusion algorithm is presented in this paper. This algorithm is used to implement fusion of SAR image and optical image,on the basis of the reciprocal characteristic of SAR image and optical image. The algorithm of image fusion presented in the paper uses a multiscale analysis and the singularity detection principle of wavelet transformation of SAR irnage and optical image, achieves a better result in experiment.
2001, 23(1): 92-95.
Abstract:
A plane surface approximation method for image coding and its quad tree scheme are presented in this paper. For a given image block, an optimal plane is found under the min-imum square error criteria, and the pixels of the block are approximated by the corresponding points on the plane. As a result, only three description parameters of the plane are needed to. be encoded. The coding quality is related to the smoothness and the size of the block. Therefore, variable block size can be chosen with a quad tree structure based on the quality requirement.
A plane surface approximation method for image coding and its quad tree scheme are presented in this paper. For a given image block, an optimal plane is found under the min-imum square error criteria, and the pixels of the block are approximated by the corresponding points on the plane. As a result, only three description parameters of the plane are needed to. be encoded. The coding quality is related to the smoothness and the size of the block. Therefore, variable block size can be chosen with a quad tree structure based on the quality requirement.
2001, 23(1): 96-99.
Abstract:
An increase in chip area and circuit complexity leads to a reduction in the yield of production. In order to solve the problem of low yield by defects in process of large scale integrated circuits manufacture, the fault-tolerant technique is introduced into the integrated circuits design. A system must be provided with a certain quantity of spare elements to have the ability of fault-tolerance. In this paper, the problem of optimal spare allocation for fault-tolerant VLSI is solved effectively by means of genetic algorithm.
An increase in chip area and circuit complexity leads to a reduction in the yield of production. In order to solve the problem of low yield by defects in process of large scale integrated circuits manufacture, the fault-tolerant technique is introduced into the integrated circuits design. A system must be provided with a certain quantity of spare elements to have the ability of fault-tolerance. In this paper, the problem of optimal spare allocation for fault-tolerant VLSI is solved effectively by means of genetic algorithm.
2001, 23(1): 85-87.
Abstract:
The idea about undulating propagation presented by the paper Undulating propagation of the electromagnetic wave is discussed. This paper point out the mistake of the idea, which is the result of wrong mathematicaJ formula of the polarization wave and wrong calculation.
The idea about undulating propagation presented by the paper Undulating propagation of the electromagnetic wave is discussed. This paper point out the mistake of the idea, which is the result of wrong mathematicaJ formula of the polarization wave and wrong calculation.
