By Topic

Signal Processing Proceedings, 2000. WCCC-ICSP 2000. 5th International Conference on

Date 21-25 Aug. 2000

Go

Filter Results

Displaying Results 1 - 25 of 159
  • 2000 5th international conference signal processing proceedings

    Publication Year: 2000 , Page(s): 0_2 - 31
    Save to Project icon | Request Permissions | PDF file iconPDF (1582 KB)  
    Freely Available from IEEE
  • Author index

    Publication Year: 2000 , Page(s): 1 - 9
    Save to Project icon | Request Permissions | PDF file iconPDF (455 KB)  
    Freely Available from IEEE
  • Statistical models for multidisciplinary applications of image segmentation and labelling

    Publication Year: 2000 , Page(s): 2103 - 2110 vol.3
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (800 KB)  

    Three classes of statistical techniques used to solve image segmentation and labelling problems are reviewed: (1) supervised and unsupervised pixel classification, (2) exploitation of the probability distribution map as a way to model image structure, (3) Markov random field modelling combined with MAP statistical classification. Diverse examples illustrate the potential of the three approaches that are described as generic methods belonging to a common framework for image segmentation/labelling View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Parameterized wavelet architectures for image processing

    Publication Year: 2000 , Page(s): 2111 - 2114 vol.3
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (352 KB)  

    This paper describes a high level environment dedicated to implement biorthogonal wavelet transforms on the Xilinx XC 4000 series. The system hides the low level hardware details of the FPGA structure and thus allows the user to concentrate more on the experimentation rather than on the low-level architecture. The implementation of the biorthogonal 9/7 wavelet shows the effectiveness of the approach View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A time domain synthesized binary phase code sidelobe suppression filter based on genetic algorithm

    Publication Year: 2000 , Page(s): 1907 - 1910 vol.3
    Cited by:  Papers (1)  |  Patents (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (220 KB)  

    THe binary phase coded waveform (BPCW) is widely used in tracking radars, while the sidelobe suppression filter (SSF) designing methods for BPCW are always limited to short codes such as the Barker code. A new kind of SSF based on time domain synthesis and genetic algorithm (GA) is presented. The filter can be used for various lengthened codes, including the m-sequence etc. Some simulation results of time domain synthesized SSF for the Barker codes and m-sequence codes are given, together with comments on the performance comparison of the new filter with some known filters View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Novel face recognition based on individual eigen-subspaces

    Publication Year: 2000 , Page(s): 1522 - 1525 vol.3
    Cited by:  Papers (2)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (352 KB)  

    In this paper, a novel method of face recognition based on individual eigen-subspaces is presented, which is expected to tackle such problems on face recognition as the ready availability of reference samples for each subject. In the proposed method, multiple face eigensubspaces are created, with each one corresponding to one known subject privately, rather than all individuals sharing one universal subspace as in the traditional eigenface method. Compared with the traditional single subspace face representation, the proposed method captures the extrapersonal difference to the most possible extent, which is crucial to distinguish between individuals, and on the other hand, it throws away the most intrapersonal difference and noise in the input. Our experiments strongly support the proposed idea, in which 20% improvement of performance over the traditional “eigenface” has been observed when tested on the same face base View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Chirp step-frequency moving object echo real time digital generation methods

    Publication Year: 2000 , Page(s): 1838 - 1842 vol.3
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (412 KB)  

    Chirp step-frequency waveform is a new high range resolution (HRR) radar signal, which has the ability to get HRR under low instantaneous signal processing bandwidth. But the matted coupling of time-delay and phase-shift demands that echo modeling has quite accurate delay and phase characteristics, which result in huge computing quantities and become a technical difficult problem of signal processing and echo-simulation. Based on a HRR radar's design the paper put forward two real time digital generation methods based on SHARC: one is derived from signal subsection decompose and modified least square method (MLSM) and another is a method based on iteration for rectilineal object track, which are proved to be feasible and also provide new resolutions for digital generation of wide band radar signal View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Nonlinear blind signal separation: an RBF-based network approach

    Publication Year: 2000 , Page(s): 1739 - 1742 vol.3
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (292 KB)  

    This paper presents a radial basis function (RBF) based approach for blind signal separation in a nonlinear mixture. A cost function, which consists of the mutual information and partial moments of the outputs of the separation system, is defined to extract the independent signals from their nonlinear mixtures. The minimization of the cost function results in the independence of the outputs with desirable moments such that the original sources are separated properly. A learning algorithm for the parametric RBF network is established by using the stochastic gradient descent method. This approach is characterized by high learning convergence rate of weights, modular structure, as well as feasible hardware implementation. A simulation result demonstrates the feasibility, and validity of the proposed approach View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Structured total least squares based on genetic algorithms

    Publication Year: 2000 , Page(s): 1671 - 1674 vol.3
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (156 KB)  

    Structured rank-deficient matrices arise in many applications in signal processing. The inverse iteration algorithm was proposed to solve the so-called structured total least squares (STLS) problems. This algorithm, however, converges to local-minimum under certain conditions. It is well known that genetic algorithms are stochastic optimization techniques that can often outperform classical methods of optimization. Genetic algorithms was utilized here to get the better solution of the STLS problems. Computer simulations show that our method ensures convergence to global minimum View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A clustering algorithm by deterministic annealing and its global convergence

    Publication Year: 2000 , Page(s): 1546 - 1550 vol.3
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (232 KB)  

    The deterministic annealing (DA) is a useful approach to clustering and related optimization problems. With a view to the optimization problem, the clustering algorithm by DA is reformulated in this paper. An important global convergence theorem about this clustering algorithm has been proved View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Design of an obstacle-avoidance strategy based on robot soccer competition

    Publication Year: 2000 , Page(s): 1722 - 1725 vol.3
    Cited by:  Papers (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (252 KB)  

    Design of an obstacle-avoidance strategy is introduced in a robot soccer system and the realizing process of this arithmetic is illustrated in detail in the paper. It makes the robot avoid an obstacle finding a shortest path from starting point to the target point and it has been verified in simulation and a real system, and simulation results are presented in the paper View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Information flow of non-uniform differential quantization

    Publication Year: 2000 , Page(s): 2026 - 2036 vol.3
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (544 KB)  

    Quantization as part of the analog-to-digital measurement procedure does not depend only on the signal dynamics of the measured object, but also on the dynamic behaviour of the measurement channel. A numerical value is attained with successive approximation of the difference between the the reference and measured quantity. For the effectiveness of differential tracking, the non-uniform quantization must fulfill three conditions: partitions into halves, increasing quantization uncertainty with difference, and low overlapping of the quantization intervals. The best trade between the number of decision levels and the settling time is with a pure exponential quantization rule. The fastest response is achievable with base 2. The finite impulse response of the sampler causes the finite information capacity of the channel. The information rate per step is a little below three. The number of bits decreases towards the end of the conversion View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • The associative memory with optimal storage error-correcting performance

    Publication Year: 2000 , Page(s): 1576 - 1579 vol.3
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (244 KB)  

    The character of the associative memory is discussed emphatically from the viewpoint of the storage capacity and the associative ability of the network. The associative memory function can all be realized by utilizing the feedback network or the forward-feed network. The new algorithm, based on adaptive competing-classifying (ACC) algorithm, is presented here. Compared with other associative memories, the neural network determined by the algorithm has high error-correcting capability, large storage capacity, no “rejected-point” and no spurious attractive center. ACC algorithm can decrease the competition-number more than CC algorithm. The obtained network has the very strong ability of correcting errors and exceeds the storage capacity of HAM and BP networks View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • New approach of ultrasonic distance measurement technique in robot applications

    Publication Year: 2000 , Page(s): 2066 - 2069 vol.3
    Cited by:  Papers (5)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (288 KB)  

    This paper deals with the development of a prototype of an airborne ultrasonic distance measurement system applied to an autonomous mobile robot. The pivotal point of this system is the use of a wideband continuous random signal where the detection of the echo delay (time of flight) is done by cross correlation technique. Additionally a Doppler frequency shift caused by robot or obstacle movement is estimated and included in the algorithm. The major advantage of this technique is the multiple sensor interface ability at perfectly low crosstalk between each sensor unit View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Redundant data compression and location accuracy analysis in T/R-R bistatic radar system

    Publication Year: 2000 , Page(s): 1951 - 1955 vol.3
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (300 KB)  

    In modern ECM and ECCM environments, bistatic radar systems have drawn increasing interest in military applications. Bistatic radar system involves two types: T-R and T/R-R. However, the feasibility of compressing redundant data in a bistatic radar system by Markov estimation has received little attention so far. This paper presents a theorem with respect to this particular issue. Theoretical analysis shows that, if there are same components in two measurement sets, redundant data compression in the sense of Markov estimation can not be made. Under the circumstance that the redundant data can be compressed, simulation results show that data compression can improve the location precision greatly in a bistatic radar system, and particularly in the region of the base line, the near region of the transmitter station and the receiver station View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Training support vector machines: an application to well-log data classification

    Publication Year: 2000 , Page(s): 1427 - 1431 vol.3
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (296 KB)  

    In this paper, we investigate the application of support vector machines (SVM) in pattern recognition. SVM is a learning technique developed by Vapnik et al. (1997) that can be seen as a new method for training polynomial, neural network, or radial basis functions classifiers. The decision surfaces are found by solving a linearly constrained quadratic programming problem. We present experimental results of our implementation of SVM, and demonstrate its advantage on well-log data classification problem View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Target location and speed estimation by multistatic radar system using maximum likelihood approach

    Publication Year: 2000 , Page(s): 1964 - 1967 vol.3
    Cited by:  Papers (1)  |  Patents (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (260 KB)  

    This paper describes a short range target location and speed estimation using the maximum likelihood method based on the range difference information of a T-R multistatic radar system transmitting a new grouped waveform of FMCW signal and single carrier pulse. Formulae are derived independent of transmitter position. Performances are simulated with the proposed algorithm and results are given for various situations View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Attribute cluster network and fractal image compression

    Publication Year: 2000 , Page(s): 1613 - 1616 vol.3
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (276 KB)  

    The greatest disadvantage of fractal image compression is the length of time it takes; so this paper advises using attribute cluster network (ACN) to reduce the time complexity. Using ACN can classify the feature vectors of domain blocks and range blocks, and the searching complexity can be reduced. Moreover, the weight values from ACN indicate the importance of all the points, and the new feature vectors can be got only by picking up more important points from a block, which will reduce the needed memory and improve encoding speed and clustering speed. Experiments show that combining the attribute cluster network and fractal block coding together costs less encoding time and clustering time than the other methods; that is to say, the chosen important points can replace the block itself properly View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Estimating the parameters of moving targets in the SAR

    Publication Year: 2000 , Page(s): 1843 - 1846 vol.3
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (240 KB)  

    In this paper, a method of estimating the parameters of moving targets in SAR is given. Theoretical analyses and simulations are carried out. The results are that the method can effectively resolve the number of moving targets and estimate the parameters of the moving targets View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A new MEBML-based algorithm for adjusting parameters online

    Publication Year: 2000 , Page(s): 1714 - 1717 vol.3
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (240 KB)  

    An MEBML (mind-evolution-based machine learning)-based algorithm for adjusting parameters online is proposed. This new AI algorithm can provide laws for parameters adjusted online by adopting the new learning system-MEBML and building the adjusting functions. This new algorithm is applied in adjusting the output scaling factor of the fuzzy logic controller (FLC). In this way, a new FLC is constructed. Conclusions can be drawn from simulation results: 1) MEBML has the rapid convergence rate and high calculation accuracy; 2) the new algorithm is easy and effective; 3) the performance of the new FLC is good View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Parametric optimization in nonlinear system based on genetic algorithms

    Publication Year: 2000 , Page(s): 2037 - 2041 vol.3
    Cited by:  Papers (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (372 KB)  

    A simulating optimization method in nonlinear systems based on genetic algorithms (GAs) is presented and the parametric optimization problem in nonlinear systems is solved. Because GAs search answers in several areas of solution space and can jump out of local optimization at a higher probability, global optimal solutions can be found. Simulation results demonstrate that this method is valid in parametric optimization of nonlinear systems View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Parameter optimization in FCM clustering algorithms

    Publication Year: 2000 , Page(s): 1457 - 1461 vol.3
    Cited by:  Papers (3)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (272 KB)  

    Weighting exponent m is an important parameter in fuzzy c-means (FCM) clustering algorithm, which directly affects the performance of the algorithm and the validity of fuzzy cluster analysis. However, so far the optimal choice of m is still an open problem. A method of selecting the optimal m is proposed in this paper, which is based on the fuzzy decision theory. The experimental results obtained demonstrate its effectiveness and arrive a conclusion that the optimal range of m is [1.5, 2.5] in practical applications View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A multi-moment pattern recognition method based on neural networks

    Publication Year: 2000 , Page(s): 1675 - 1678 vol.3
    Cited by:  Papers (3)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (216 KB)  

    Proposes a fuzzy set theory to preprocess the feature of a Chinese character and uses the multi-moment method to classify the patterns in the neural network. Since it can be used in overlapping classes, this method has practical value View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • De-noising in digital signal spectrum-coding transmitting system with wavelets

    Publication Year: 2000 , Page(s): 1767 - 1770 vol.3
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (196 KB)  

    A new method of denoising with the sym4 wavelet under the soft threshold rule is presented to remove noise in a digital signal spectrum-coding transmitting system. From the simulation of the de-noising process, a conclusion can be drawn that this method is effective in real time de-noising in a digital signal spectrum-coding transmitting system View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A new kind of ISAR autofocusing technique based on entropy criteria

    Publication Year: 2000 , Page(s): 1806 - 1809 vol.3
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (268 KB)  

    In this paper, a new kind of autofocusing technique in inverse synthetic aperture radar (ISAR) imaging is introduced. This non-parametric technique is based on the entropy minimization principle. Some results of real ISAR data confirm the feasibility of this method View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.