Scheduled System Maintenance
On Tuesday, May 22, IEEE Xplore will undergo scheduled maintenance. Single article sales and account management will be unavailable
from 6:00am–5:00pm ET. There may be intermittent impact on performance from noon–6:00pm ET.
We apologize for the inconvenience.

Signal Processing, 1996., 3rd International Conference on

18-18 Oct. 1996

Go

Filter Results

Displaying Results 1 - 25 of 226
  • 1996 3rd International Conference Signal Processing Proceedings

    Publication Year: 1996, Page(s):0_2 - 26
    Request permission for commercial reuse | PDF file iconPDF (1867 KB)
    Freely Available from IEEE
  • A comparison study of image segmentation by clustering techniques

    Publication Year: 1996, Page(s):1280 - 1283 vol.2
    Cited by:  Papers (8)  |  Patents (3)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (491 KB)

    Many image segmentation methods based on clustering are available in the literature. Some of these techniques use classical clustering, some use fuzzy sets. Most of these techniques are not suitable for noisy environments. Some work has been done using the possibilist clustering approach which is robust to noise, but requires knowledge of the number of clusters. The probabilistic approach can be u... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An investigation into the execution of human skilled movements

    Publication Year: 1996, Page(s):1437 - 1440 vol.2
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (357 KB)

    This research is aimed at improving understanding of the execution of human skilled movements in the saggital plane. Towards this aim a model-based approach that consists of the following steps is adopted: (a) a physiologically-motivated model of the lower extremity is developed; (b) a neurophysiological controller with neural activation and intrinsic/peripheral feedback is designed; (c) an optimi... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • The golden number and the extraction of edges: application to trace an implicit function

    Publication Year: 1996, Page(s):1671 - 1674 vol.2
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (314 KB)

    To trace an explicit function, many items of software (for example: Matlab, Mathcad, Mathematica, Gnuplot, ..., under MS-DOS, UNIX, Windows) have been made. The method in theory is simple: if the value of the argument x is given, then we can calculate the value of the function y, as an expression of the given function. To trace an implicit function, there is not at present any efficient method bec... View full abstract»

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

    Publication Year: 1996, Page(s):1 - 11
    Request permission for commercial reuse | PDF file iconPDF (688 KB)
    Freely Available from IEEE
  • Distortion model selecting and accuracy evaluation for CCD camera calibration

    Publication Year: 1996, Page(s):875 - 878 vol.2
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (400 KB)

    The precision of a CCD camera calibration depends on both the distortion models selected and the signal-to-noise. The former is called the intrinsic factor, and latter is the extrinsic factor. So far, none of the fixed criterion has been used for selecting the distortion model, and almost all are personal or experiential. None of the fixed criterion has been used for determining the degree of erro... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A JPEG codec reflecting human sense

    Publication Year: 1996, Page(s):1002 - 1006 vol.2
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (780 KB)

    This paper presents a novel standard baseline JPEG compatible still image compression method that allows to flexibly control the decompression duality of regions in the image. Therefore, it allows both higher quality of important parts and lower bitrate as a whole, and it reflects the sense of human beings. We use the fuzzy technique to determine the important regions in the same manner as human b... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Fast edge detection for color image

    Publication Year: 1996, Page(s):871 - 874 vol.2
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (348 KB)

    A fast edge detection method based on the pyramid structure and vector operator is presented for color image analysis. The pyramid image structure has been used to reduce the computational cost. The vector operator which utilizes the inherent correlation between the RGB channels of a color image is applied to obtain an edge map at the top pyramid level. The edge interpolation, adjustment and smoot... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Still color image coding using local iterated function systems (LIFS)

    Publication Year: 1996, Page(s):998 - 1001 vol.2
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (280 KB)

    In this paper, we introduce several still color image coding schemes using local iterated function systems (LIFS), which have aroused much interest all over the world in the past few years. A lot of experiments based on different schemes and different coding methods are presented and compared. The results show that some schemes presented in the paper can achieve high performance View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Discrete Gabor expansion via Wely-Heisenberg frames

    Publication Year: 1996, Page(s):1211 - 1214 vol.2
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (300 KB)

    A new method to obtain the dual window sequence of the discrete Gabor expansion is presented. We will first generalize the discrete Zak transform and establish the relationships between the continuous and discrete Gabor expansion. Then we will propose a method to compute the discrete Gabor dual window sequence from the continuous Gabor expansion dual window. Numerical examples are given View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Prediction by back-propagation neural network for lossless image compression

    Publication Year: 1996, Page(s):1026 - 1030 vol.2
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (544 KB)

    This paper describes a prediction process produced by a back-propagation neural network for lossless image compression. The predictor is designed by supervised training of a back-propagation neural network using actual image pixels, i.e. using a typical sequence of pixel values. The significance of this approach lies in the fact that it can exploit high-order statistics and the nonlinear function ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Comparison of two moving target detector's performance

    Publication Year: 1996, Page(s):867 - 870 vol.2
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (264 KB)

    This paper compares two kinds of moving target detector: AMTD-CFAR (adaptive moving detection CFAR) and OS-CFAR detector. Their detection performance is investigated for a number of situations such as different distributions, uniform backgrounds, clutter edges and strong interference targets. The results indicate that the AMTD-CFAR detector has a very small loss of CFAR, and its detection performa... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A system for transform vector coding of images

    Publication Year: 1996, Page(s):994 - 997 vol.2
    Cited by:  Papers (2)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (416 KB)

    The author describes a full development system for analysis of selected transform vector quantization algorithms for black and white image coding. The software has been developed under the Windows environment. The system is quite open and can be freely expanded and modified. Up to this time it has been equipped with two basic transform algorithms: discrete Fourier transform (DFT) and discrete cosi... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A new CFAR detector with automatic censoring

    Publication Year: 1996, Page(s):1632 - 1635 vol.2
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (332 KB)

    In this paper a new CFAR detector (MOSTM) based on order statistics (OS), and trimmed mean (TM) is introduced. It takes the mean value of OS and TM local estimations as a noise power estimation to set an adaptive threshold , and it also uses the automatic censoring technique proposed previously (see IEE Proc., Radar Sonar Navig.,vol.141, p.205, 1994). Under Swerling II assumption, the analytic exp... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • DCT-based a new approach of improving image coder performance using the spatial reconstructing noise shaping

    Publication Year: 1996, Page(s):1022 - 1025 vol.2
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (256 KB)

    A modification of the DCT based coding system is proposed that allows to suppress the reconstruction noise in the frequency bands where it is less visible. The modified coder is compatible with the standard DCT system. A 1-D noise feedback filter is proposed which has been designed from the human visual system. The result shows that with noise feedback, the quantization noise can be flexibly contr... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An effective IMM algorithm for maneuvering target tracking

    Publication Year: 1996, Page(s):922 - 925 vol.2
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (232 KB)

    To reduce the complexity of algorithms for tracking maneuvering targets, a two stage filter scheme based on non-zero mean target dynamical model has been developed, and to maintain good tracking performance for non-maneuvering targets, an improved interacting multiple model (IMM) algorithm with the two stage filter has been suggested. The Monte Carlo simulation shows that this scheme posses a good... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A framework for the design of new line parameterizations

    Publication Year: 1996, Page(s):863 - 866 vol.2
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (360 KB)

    Based on four criteria, namely uniqueness, boundedness, uniformness and recursiveness, a framework for the design of new line parameterizations is introduced. It is shown that there does not exist a line parameterization which could satisfy all the above four criteria, and that the line parameterization, formulated as c=ax+(a0-|a|)y a∈[-a0,a0) where a 0 View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Binary tree image coding based on region

    Publication Year: 1996, Page(s):990 - 993 vol.2
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (428 KB)

    A new image coding method based on the image region is proposed. According to the features of the image gray surface, an image is partitioned into many convex polygons which approximate the original image as closely as possible. The experimental results show that a good reconstructed image quality can still be obtained when the data compression ratio is more than 20:1 and the SNR is more than 30 d... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A cosine-modulated Gaussian activation function for hyper-hill neural networks

    Publication Year: 1996, Page(s):1397 - 1400 vol.2
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (348 KB)

    We present here a new class of activation functions for neural networks, which is called the CosGauss function. This function is a cosine-modulated Gaussian function. In contrast to the sigmoidal-, hyperbolic tangent- and Gaussian activation functions, more ridges can be obtained by the CosGauss function. It is proved that this function can be used to approximate polynomials and step functions. Th... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Transient analysis on machinery condition monitoring

    Publication Year: 1996, Page(s):1723 - 1726 vol.2
    Cited by:  Papers (2)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (448 KB)

    In this research, three main time-frequency/time-scale (TF/TS) analysis algorithms-windowed Fourier transform (WFT), Wigner-Ville distribution (WVD), and wavelet analysis (WA), have been studied and implemented. We compare the different algorithms using an identical set of synthetic signals, from which one can realize their individual features. To justify the effectiveness and to highlight the req... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Acquisition and image processing system able to optimize laser cladding process

    Publication Year: 1996, Page(s):1628 - 1631 vol.2
    Cited by:  Papers (1)  |  Patents (7)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (488 KB)

    The laser cladding process, which consists of adding a melt powder to a substrate in order to improve or change the behaviour of the material against corrosion, fatigue and so on, involves a lot of parameters. In order to perform good tracks some parameters need to be controlled during the process. We present here a low cost performant system using two CCD matrix cameras. One camera provides surfa... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Neural approaches for human signature verification

    Publication Year: 1996, Page(s):1346 - 1349 vol.2
    Cited by:  Papers (3)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (344 KB)

    This paper describes three neural network (NN) based approaches for on-line human signature verification: Bayes multilayer perceptrons (BMP), time-delay neural networks (TDNN), and input-oriented neural networks (IONN). The backpropagation algorithm was used for the network training. A signature is input as a sequence of instantaneous absolute velocity (|υ(t)|) extracted from a pair of spatia... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Multiple-codebook run-length vector quantization for image coding

    Publication Year: 1996, Page(s):1019 - 1021 vol.2
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (556 KB)

    It is observed that the utilization of spatial redundancy when using vector quantization (VQ) to code images improves the coding performance. One way of utilizing this redundancy is to segment an image based on textural tendencies, and to use multiple codebooks for each segment. The local persistence of each segment is the spatial redundancy, and this eliminates the need to indicate the type infor... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A study on performance improvement of the interpolation with a half-band filter

    Publication Year: 1996, Page(s):1691 - 1694 vol.2
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (248 KB)

    Usually, in an all-digital receiver, a local clock samples the received waveform at a low sampling rate (fs=2/T or 4/7, T is the symbol interval). Then an interpolator is used for generating the corresponding interpolants which synchronize with the timing of the input signal. By introducing a half-band filter between the matched filter and a simple polynomial-based interpolator, the per... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A new method of radar target recognition based on neural network

    Publication Year: 1996, Page(s):1492 - 1495 vol.2
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (364 KB)

    This paper provides, for the first time, a new method for recognizing the number of aircraft in radar targets based on a neural network. The method converts many duplicate cycle intermediate frequency (IF) narrow bandwidth signals received by the radar which scans the targets into a complex envelope signal, calculates its autocorrelation matrix and autocorrelation matrix characteristic values, and... View full abstract»

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