19,242 research outputs found

    A Comparison of Adaptive Beamforming Implementations for Wideband Scenarios

    No full text
    In this paper, we compare low cost implementations of wideband adaptive beamformers utilising the generalised sidelobe canceller (GSC) in combination with the least-mean squares algorithms. DFT based techniques suffer from unjustified narrowband assumptions. We therefore derive and investigate an overlap-save implementation of the GSC, which offers a steady state suppression of jammers equivalent to a time domain GSC but may be prone to slow convergence. Finally, subband techniques offer a more robust convergence trade-off for a somewhat higher computational cost. Analysis and simulation results revealing some of the algorithms' properties are presented

    Model-based stroke extraction and matching for handwritten Chinese character recognition

    No full text
    This paper proposes a model-based structural matching method for handwritten Chinese character recognition (HCCR). This method is able to obtain reliable stroke correspondence and enable structural interpretation. In the model base, the reference character of each category is described in an attributed relational graph (ARG). The input character is described with feature points and line segments. The strokes and inter-stroke relations of input character are not determined until being matched with a reference character, The structural matching is accomplished in two stages: candidate stroke extraction and consistent matching. All candidate input strokes to match the reference strokes are extracted by line following and then the consistent matching is achieved by heuristic search. Some structural postprocessing operations are applied to improve the stroke correspondence. Recognition experiments were implemented on an image database collected in KAIST, and promising results have been achieved. (C) 2001 Pattern Recognition Society. Published by Elsevier Science Ltd. All rights reserved

    Multiresolution locally expanded HONN for handwritten numeral recognition

    No full text
    In this paper, we propose a neural network architecture, multiresolution locally expanded high order neural network (MRLHONN) to solve the problem of handwritten numeral recognition. In this recognition scheme, the multiresolution representation of character image is input into a high order neural network (HONN), while in each resolution, only neighboring pixels are expanded to produce high order input. The property of this architecture is that, the local expansion alleviate the problem of large connecting weight set, and the multiresolution representation remedy the inadequacy of local expansion. Two forms of multiresolution representations, quadtree representation and Gaussian pyramid, were used in experiments. The recognition results demonstrate the efficiency of the proposed architecture. (C) 1997 Elsevier Science B.V
    corecore