摘要:

An adaptive filter that enables coefficient computation of the Gabor expansion while employing the least mean-squared algorithm for complex signals is described. A significant simplification of the filter structure and the use of FFT algorithms contribute to a fast and efficient determination of the coefficients when compared with known algorithms for Gabor computation. Based on these results, we propose a hardware realisation using a CORDIC processor. Computer simulations indicate that sufficient precision and a fast convergence of the coefficients can be obtained.

DOI:10.1049/ip-i-2.1993.0002

出版商:IEE    

年代:1993

数据来源: IET

摘要:

Two closely related methods for skeletonisation via a Euclidean distance transform are presented. First from an observation on the response of the human visual system to a grey level image of the transform, convolution with the Marr-Hildreth operator is proposed. Secondly, it is suggested that skeletonisation can be performed by convolution of the distance transform with a function designed to detect the occurrence of a specific geometric model, or feature, in a way similar to that used in contemporary edge detection algorithms. Examples of the technique are shown and the results discussed.

DOI:10.1049/ip-i-2.1993.0003

出版商:IEE    

年代:1993

数据来源: IET

摘要:

New results, which have been achieved by using neural network architectures for two-dimensional image classification based on the goal-seeking neuron (GSN), are presented. A number of important practical issues concerning mapping topologies and the parallel implementation of GSN-based architectures are also investigated, together with a proposal for the development of a related neurally based feature extractor to be used as a front-end processor in a fully integrated Boolean network architecture.

DOI:10.1049/ip-i-2.1993.0004

出版商:IEE    

年代:1993

数据来源: IET

摘要:

This paper describes VHD (very high definition) image media competing in spatial resolution with the familiar film media as a new digital image media concept, and discusses a problem of image acquisition as a key technology for achieving VHD image systems. Towards the development of a VHD image acquisition system, the work presents a new signal-processing based approach using multiple different cameras. The image acquisition approach, processing and integrating multiple images taken simultaneously with multiple different cameras, produces an improved spatial resolution image with sufficiently high signal-to-noise ratio. Theoretical analysis and experimental simulations show clearly the improvement in the high frequencies and details.

DOI:10.1049/ip-i-2.1993.0005

出版商:IEE    

年代:1993

数据来源: IET

摘要:

Model-based image coding applied to interpersonal communication achieves very low bit-rate image transmission. To accomplish it, accurate three-dimensional (3-D) motion estimation of a speaker is necessary. A new method of 3-D motion estimation is presented, consisting of two steps. In the first, facial contours and feature points of a speaker are extracted using filtering and Snake algorithms. Five feature points on a speaker's facial image are tracked between consecutive picture frames, which gives 2-D motion vectors of the feature points. Then, in the second step, the 3-D motion of a speaker's head is estimated using a three-layered neural network model, after training with many possible motion patterns of the human head using an existing 3-D general shape model. Experimental results show that our method not only achieves good results but is also more robust than existing methods, even when the motion of an object is rather large or complicated. Accurately estimated 3-D motion parameters can realise image transmission at a very low bit rate.

DOI:10.1049/ip-i-2.1993.0006

出版商:IEE    

年代:1993

数据来源: IET

摘要:

The segmentation of colour pictures in a graph-theoretical context is considered. The procedure aims at identifying, extracting and classifying visually important features on the image plane, such as regions of homogeneous colour and chromatic transitions. Well established principles of colour theory and graph theory are combined to obtain a unified representation of a colour picture. The picture is represented by means of a weighted graph, constructed so as to reflect the specification of the colour space employed as well as important relationships between picture elements. A spanning tree of the graph is obtained by iteratively minimising a specific picture distortion measure. This tree structure describes a hierarchy of partitions on the image plane. Each partition comprises disjoint regions containing elements with similar attribute. Due to the fact that region identification and edge detection form dual problems from the graph-theoretical viewpoint, region contours defined by such partitions form a hierarchy. To avoid artificial contouring, a specific type of artefact introduced by the segmentation algorithm, the use of higher level information, is considered. It is shown that, when texture (which is taken into account at an intermediate stage of picture segmentation) is combined with colour as joint similarity attributes of regions, an improved hierarchical description of contours is possible. This facilitates the progressive elimination of the undesirable contours and leads to the visual enhancement of the segmentation obtained.

DOI:10.1049/ip-i-2.1993.0007

出版商:IEE    

年代:1993

数据来源: IET

摘要:

In this paper, Bayesian networks of Markov random fields (BN-MRFs) are proposed as a technique for representing and applying apriori knowledge at different abstraction levels inside a distributed image processing framework. It is shown that this approach, thanks to the common probabilistic basis of the two techniques, is able to combine in a natural way causal inference properties at different abstraction levels as provided by Bayesian networks with optimisation criteria usually applied to find the best configuration for an MRF. Examples of two-level BN-MRFs are given, where each node uses a coupled Markov random field which has to solve a coupled restoration and segmentation problem. Experiments are concerned with expert-driven registered segmentation and tracking of regions from image sequences.

DOI:10.1049/ip-i-2.1993.0008

出版商:IEE    

年代:1993

数据来源: IET

摘要:

A fast lattice-based gain/shape vector-quantisation method (FLGSVQ) is presented. The proposed technique combines the basic ideas of gain/shape vector quantisation (GSVQ) with lattice codes to offer an encoding process which is extremely fast compared with the conventional GSVQ. The main aim is to show how vector quantisers designed using the LBG algorithm can be substituted by lattice-based ones with vast reductions in encoding complexity, at a competitive image quality. A simple single-/two-stage coding system based on lattice quantisation is tested for interframe predictive coding of image sequences. Simulation results are provided to compare the performance of the proposed method with the conventional LBG-designed vector quantisation. These results, along with subjective evaluations, clearly demonstrate that lattice quantisation can be competitive with vector quantisation in terms of signal-to-noise performance, only at a very small fraction (4%) of the vector-quantiser complexity.

DOI:10.1049/ip-i-2.1993.0009

出版商:IEE    

年代:1993

数据来源: IET

摘要:

The general theory of designing multidimensional quadrature mirror filters (QMF) for use in sub-band coding systems using the McClellan transform [1] is presented. It was recently shown that the McClellan transform could be used to generate 2-D diamond-shape QMF filters [2]. The proofs of the diamond-shape case are formalised and generalise to other shapes, sampling rasters and dimensions. It is possible and even more convenient to design QMF filter banks by performing transformations on a class of real-valued polynomials. Examples are given of two-dimensional diamond and fan-shape filters and three-dimensional tetrad filters designed using this transformation technique.

DOI:10.1049/ip-i-2.1993.0010

出版商:IEE    

年代:1993

数据来源: IET

摘要:

This paper is concerned with the problem of model based recognition of polyhedral objects from a single perspective view. A hypothesise-verify paradigm based on the use of high level knowledge constraints derived from local shape properties is presented. In the recognition system, two high level features, namely triangle pair and quadrilateral are employed as key features for model invocation and hypothesis generation. A verification process for performing a detailed check on the model to scene correspondences is developed. To reduce the number of implausible hypotheses generated from scene to model intermediate feature assignments, two geometrical contraints, namely distance and angle constraints are employed. A list of closed polygons and C-triple pairs extracted from a 2D intensity image by means of edge and intermediate feature detection process is used as an input to the matching system. The intermediate feature grouping process starts by identifying junctions created by pairs of line segments and then forms triples by combining pairs of junctions which share a common line. These triples are then scanned by procedure which connects them into meaningful geometric structures. As a byproduct of the recognition method, the relative pose of the 3D polyhedral objects with respect to the camera is recovered. Extensive experimental results are reported to confirm the feasibility of the proposed method.

DOI:10.1049/ip-i-2.1993.0011

出版商:IEE    

年代:1993

数据来源: IET