By Topic

Vision, Image and Signal Processing, IEE Proceedings -

Issue 2 • Date 21 April 2003

Filter Results

Displaying Results 1 - 7 of 7
  • Enhancement of a genetic algorithm for affine invariant planar object shape matching using the migrant principle

    Page(s): 107 - 113
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (739 KB)  

    The use of the migrant principle has proved to be effective in reducing the impact of the initial populations of genetic algorithms in optimising simple linear functions. Analytical and empirical results have also suggested that the method could be applied to locate an optimal solution in larger search space with more complex landscape. In the paper, an attempt has been made to develop an enhanced object matching technique that is based on the integration of the migrant principle and an existing genetic algorithm for affine invariant object recognition. As the latter had been taken as the foundation of a series of research works, any improvement on the scheme will directly benefit subsequent developments. The problem being addressed is highly nonlinear, which requires well-formed initial populations to attain successful matching of object shapes. Experimental results reveal that, for the same population size and mutation rate, the proposed method demonstrates significant improvement, as compared with its precedent, and that it is insensitive to the initial population. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Stochastic gradient based third-order Volterra system identification by using nonlinear Wiener adaptive algorithm

    Page(s): 90 - 98
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (708 KB)  

    The nonlinear Wiener stochastic gradient adaptive algorithm for third-order Volterra system identification application with Gaussian input signals is presented. The complete self-orthogonalisation procedure is based on the delay-line structure of the nonlinear discrete Wiener model. The approach diagonalises the autocorrelation matrix of an adaptive filter input vector which dramatically reduces the eigenvalue spread and results in more rapid convergence speed. The relationship between the autocorrelation matrix and cross-correlation matrix of filter input vectors of both nonlinear Wiener and Volterra models is derived. The algorithm has a computational complexity of O(M3) multiplications per sample input where M represents the length of memory for the system model, which is comparable to the existing algorithms. It is also worth noting that the proposed algorithm provides a general solution for the Volterra system identification application. Computer simulations are included to verify the theory. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Watermark detection based on the properties of error control codes

    Page(s): 115 - 121
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (640 KB)  

    Watermark detection is a topic which is seldom addressed in the watermarking literature. Most authors concentrate on developing novel watermarking algorithms. In a practical watermarking system, however, one must be able to distinguish between watermarked and unwatermarked documents. Many existing systems belong to the class of so called 'yes/no' watermarks, where the detector correlates the candidate image with some known sequence to determine whether a mark is present. Unfortunately, these watermarks often carry no extra information and are not very useful. On the other hand, multi-bit watermarking schemes typically use a separate reference watermark and the payload of the watermark is decoded only when this reference watermark is successfully detected in the received image. It is shown that it is not necessary to use a reference watermark for detection purposes if the watermark payload is encoded with an error control code. One can then put all the energy into the payload watermark and increase its robustness. The turbo code is used as an example of error control codes in the work presented, and simulation results using an algorithm based on the authors' previous work verifies their theory. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Fractional Fourier transform of the Gaussian and fractional domain signal support

    Page(s): 99 - 106
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (670 KB)  

    The fractional Fourier transform (FrFT) provides an important extension to conventional Fourier theory for the analysis and synthesis of linear chirp signals. It is a parameterised transform which can be used to provide extremely compact representations. The representation is maximally compressed when the transform parameter, α, is matched to the chirp rate of the input signal. Existing proofs are extended to demonstrate that the fractional Fourier transform of the Gaussian function also has Gaussian support. Furthermore, expressions are developed which allow calculation of the spread of the signal representation for a Gaussian windowed linear chirp signal in any fractional domain. Both continuous and discrete cases are considered. The fractional domains exhibiting minimum and maximum support for a given signal define the limit on joint time-frequency resolution available under the FrFT. This is equated with a restatement of the uncertainty principle for linear chirp signals and the fractional Fourier domains. The calculated values for the fractional domain support are tested empirically through comparison with the discrete transform output for a synthetic signal with known parameters. It is shown that the same expressions are appropriate for predicting the support of the ordinary Fourier transform of a Gaussian windowed linear chirp signal. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Stereoscopic dual-energy X-ray imaging for target materials identification

    Page(s): 122 - 130
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (1144 KB)  

    An experimental stereoscopic dual-energy X-ray imaging system is employed to extract the effective atomic number from a target layer of multiple layers of different materials. Further work investigated utilising stereoscopic parallax data to calculate layer thickness which, when combined with dual-energy data, enables the mass density of the target to be established. The research is part of an ongoing programme of work in collaboration with the UK Home Office to discriminate and identify substances in X-ray security screening applications. Initial work utilised a basis materials decomposition (BMD) technique to compute the characteristic angle, an indicator of atomic composition. The problem of overlapping materials masking a target was solved by further refinement of the BMD method, which can also be used determine mass density when layer thickness is known. The empirical investigation concentrated on computing the characteristic angle for different target materials in various overlapping arrangements. Also, the results from employing manually extracted parallax data to calculate the thickness of a target material are presented together with the resultant estimation of the targets' mass density. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Robust energy-to-peak filtering with improved LMI representations

    Page(s): 82 - 89
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (597 KB)  

    The authors revisit the problem of robust energy-to-peak filtering for linear systems with parametric uncertainty residing in a polytope. Based on two results that appeared recently, they derive new L2-L performance criteria which allow the use of parameter-dependent Lyapunov functions for analysis and synthesis problems. Robust L2-L filters are then designed upon the new conditions and by means of the linear matrix inequality (LMI) technique, with the result that less conservativeness is achieved compared with earlier results that are based on a quadratic framework. Both continuous and discrete-time cases are considered and numerical examples illustrate the feasibility and advantage of the proposed designs. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Generalisation of the Dirac-delta impulse extending Laplace and z transform domains

    Page(s): 69 - 81
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (855 KB)  

    A generalisation of the Dirac-delta impulse and its derivatives as two generalised distributions, namely, the xi and zeta impulses, and their derivatives, defined on the complex s-plane and z-plane of continuous-time and discrete-time functions, respectively, is proposed. The generalised impulses extend the existence of Laplace and z transforms to a large class of infinite duration two-sided functions, which hitherto had no transform or had only a Fourier transform in the form of distributions. The proposed generalised impulses are shown to bridge the gap between the theory of generalised functions and both the unilateral and bilateral Laplace and z transforms. The generalised impulses extend the existence of Laplace and z transforms to include both functions that have a Fourier transform as a distribution as well as exponentially rising infinite duration two-sided functions that have no Fourier transform. It is shown that a modulation theory can now be added to the properties of bilateral transforms. No such theorem has hitherto existed for these transforms. The proposed generalised impulses and the resulting extended Laplace and z transforms are shown to lead to new complex-plane operations, such as spatial convolution, and to simplify operations such as ordinary convolution, sampling and the solution of differential and difference equations. Bilateral Laplace and z transforms may receive greater attention now that these transforms can be applied to a new, large and basic class of functions, such as two-sided infinite duration exponentials and rising trigonometric and hyperbolic functions. View full abstract»

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