By Topic

Vision, Image and Signal Processing, IEE Proceedings -

Issue 1 • Date Feb 1996

Filter Results

Displaying Results 1 - 9 of 9
  • Source coding of wavelet-transformed digital HDTV for recording applications

    Page(s): 1 - 9
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (2692 KB)  

    A novel intraframe source-coding algorithm suitable for the recording of digital high-definition television signals is presented. A multilayered, hierarchical description of the source signal is obtained by means of a quad-tree, halfband wavelet transform. This transform decomposes the input signal into a collection of spectrally nonoverlapping sub-bands. These are used to influence the allocation of the available coding resources to the various layers, so that the integrity of visually significant features can be preserved. Individual sub-bands are quantised and entropy coded by using a predictive arithmetic coding technique. The algorithm is tuned to achieve bit-rate reduction ratios in the range 8:1-4:1 which is most useful for recording applications. Results obtained from simulating the coding algorithm show noticeable improvement over the current state-of-the-art international standard algorithm for still picture encoding, both in terms of subjective quality and of measured mean-square error View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Variable statistical wordlength in digital filters

    Page(s): 62 - 66
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (428 KB)  

    Statistical wordlength calculations estimate the wordlength to which a digital filter's coefficients can be rounded and still be likely to meet the specification. The value of this wordlength has generally been dominated by the sensitivity of the magnitude response to a small subset of filter coefficients. If variable wordlengths are used, more bits can be assigned to sensitive coefficients and fewer to insensitive coefficients, allowing the average wordlength (and hence implementation complexity) to be reduced. The paper investigates the variable wordlength technique as applied to minimax IIR filter designs. Significant reductions in filter complexity over the uniform wordlength method result View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Adaptive step edge model for self-consistent training of neural network for probabilistic edge labelling

    Page(s): 41 - 50
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (1056 KB)  

    The authors present a robust neural network edge labelling strategy in which a network is trained with data from an imaging model of an ideal step edge. They employ the Sobel operator and other preprocessing steps on image data to exploit the known invariances due to lighting and rotation and so reduce the complexity of the mapping which the network has to learn. The composition of the training set to achieve labelling of the image lattice with Bayesian posterior probabilities is described. The back propagation algorithm is used in network training with a novel scheme for constructing the desired training set; results are shown for real images and comparisons are made with the Canny (1986) edge detector. The effects of adding zero-mean Gaussian image noise are also shown. Several training sets of different sizes generated from the step edge model have been used to probe the network generalisation ability and results for both training and testing sets are shown. To elucidate the roles of the Sobel operator and the network, a probabilistic Sobel labelling strategy has been derived; its results are inferior to those of the neural network View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Bound for Minkowski metric or quadratic metric applied to VQ codeword search

    Page(s): 67 - 71
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (440 KB)  

    A bound for a Minkowski metric based on Lp distortion measure is proposed and evaluated as a means to reduce the computation in vector quantisation. This bound provides a better criterion than the absolute error inequality (AEI) elimination rule on the Euclidean distortion measure. For the Minkowski metric of order n, this bound contributes the elimination criterion from the L1 metric to L n metric. This bound can also be an extended quadratic metric which can be a hidden Markov model (HMM) with a Gaussian mixture probability density function (PDF). In speech recognition, the HMM with the Gaussian mixture VQ codebook PDF has been shown to be a promising method View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Statistical syntactic methods for high-performance OCR

    Page(s): 23 - 30
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (916 KB)  

    The paper describes a new method for language modelling and reports its application to handwritten OCR. Images of characters are first chain-coded to convert them to strings. A novel language modelling method is then applied to build a statistical model for strings of each class. The language modelling method is based on a probabilistic version of an n-tuple classifier which is scanned along the entire string for both training and recognition. This method is extremely fast and robust, and concentrates all the computational effort on the portion of the image where the information is, i.e. the edges left by the trace of the pen. Results on the CEDAR handwritten digit database show the new method to be almost as accurate as the best methods reported so far, while offering a significant speed advantage View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Modified neocognitron for improved 2-D pattern recognition

    Page(s): 31 - 40
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (1108 KB)  

    Some modifications to an existing neural network, the neocognitron, are proposed in order to overcome some of its limitations and to achieve an improved recognition of patterns (for instance, characters). Motivation for the present work arose from the results of extensive simulation experiments on the neocognitron. Inhibition during training is dispensed with, including it only during the testing phase of the neocognitron. Even during testing, inhibition is totally discarded in the initial layer because it leads, otherwise, to some undesirable results. However, inhibition, which is feature-based, is incorporated in the later stages. The number of network parameters which are to be set manually during training is reduced. The training is made simple without involving multiple training patterns of the same nature. A new layer has been introduced after the C-layer (of the neocognitron) to scale down the network size. Finally, the response of the S-cell has been simplified, and the blurring operation between the S- and the C-layers has been changed. The new architecture, which is robust with respect to small variations in the value of the network parameters, and the associated training are believed to be simpler and more efficient than those of the neocognitron View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Design of nonseparable 3-D filter banks/wavelet bases using transformations of variables

    Page(s): 51 - 61
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (1072 KB)  

    The authors present a technique to design two-channel filter banks in three dimensions where the sampling is on the FCO (face centred orthorhombic) lattice, The ideal 3-D sub-band is of the truncated octahedron shape. The design technique is based transformation of variable method equivalent to the generalised McClellan transformation. The filters are FIR, have linear phase and achieve perfect reconstruction. Although the sub-band shape is quite complicated, the ideal frequency characteristics are well approximated. This is illustrated with an example. The technique provides the flexibility of controlling the frequency characteristics of the filters with ease. The filters can be implemented quite efficiently due to the highly symmetrical nature of the coefficients of the transformation. The authors also modify and extend the basic design technique to impose the zero property (the number of zeros of the filter transfer function at the aliasing frequency) on the sub-band filters. This property is important when the filter bank is used iteratively in a tree-structured manner as a discrete wavelet transform system and the issue of regularity arises. Several design examples are presented to illustrate the design technique View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Region-based image coding using polynomial intensity functions

    Page(s): 15 - 22
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (1636 KB)  

    The vast majority of coded images are real-world images. These images consist of distinct objects within a scene, where each object has its own reflective, textural and lighting characteristics. Region-based image coding encodes these images by partitioning the scene into objects, and then describing each object's characteristics using a set of parameters. The paper uses orthonormal polynomial functions to describe the lighting and reflective characteristics of each object. The coefficients of these polynomials are coded with linear quantisers that have their decision boundaries spaced according to rate-distortion considerations. The textural component of each object is coded using vector quantisation of the autocorrelation coefficients of the residual. The partitioning of the image into distinct objects is achieved with a segmentation algorithm which attempts to maximise the rate-distortion performance of the encoding procedure as a whole. In doing so, the segmentation algorithm partitions the image into distinct objects as well as providing estimates for the optimal bit allocations among the polynomial coefficients. Results generated by this method show reconstructions with quality superior to other region-based methods, both objectively and subjectively View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Optimal quantisation strategy for DCT image compression

    Page(s): 10 - 14
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (1104 KB)  

    The authors present a strategy for generating optimal quantisation tables for use in JPEG image compression and its extension to general block sizes. Directly optimised quantisation tables were obtained by simulated annealing. A composite cost function minimised the RMS error between original and recovered images while keeping the compression ratio close to some desired value. Examination of these tables led to a simple model giving quantisation coefficients in terms of (x, y) position in the table and three model parameters. Annealing on the model parameters for several compressions yielded an expression for each parameter as a function of compression ratio. This approach was extended to general block sizes, and psychovisual evaluation determined the visually optimal block size for each compression ratio. The authors demonstrate significant improvements over JPEG coding due to the use of optimal quantisation rather than default tables. Use of general block size effectively extends the JPEG approach to higher compressions than are feasible with standard JPEG coding View full abstract»

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