A new shape adaptive predictive lossless image coder is proposed. Three classes of block shapes are delineated with associated “optimum” predctors. Each image is partitioned into sub-blocks that are classified into one of the three classes using vector quantisation. The encoder then employs the predictor corresponding to the class of the block under consideration. Performance evaluation of the ...

The hypothesis that for a given amount of training data a speaker model has an optimum number of components is examined. This is investigated with regard to Gaussian mixture models (GMM) with and without world model adaptation. Results show that maximising the number of components in a speaker model can improve speaker recognition results. Comparisons with vector quantisation (V Q) indicate tha...

A mathematical quantisation of a Random Access Memory (RAM) is proposed starting from its matrix representation. This quantum RAM (q-RAM) is employed as the neural unit of q-RAM-based Neural Networks, q-RbNN, which can be seen as the quantisation of the corresponding RAM-based ones. The models proposed here are direct realisable in quantum circuits, have a natural adaptation of the classical le...

We propose novel compression algorithms for time-varying channel state information (CSI) in wireless communications. The proposed scheme combines (lossy) vector quantisation and (lossless) compression. First, the new technique is based on a class of parametrised companders applied each component normalised CSI vector. Our algorithm chooses suitable compander an intuitively simple way whenever e...

Permanent distortion is one of the main drawbacks of all the irreversible watermarking schemes. Attempts to recover the original signal after the signal passing the authentication process are being made starting just a few years ago. Some common problems, such as salt-andpepper artefacts due to intensity wraparound and low embedding capacity, can now be resolved. However, we point out in this w...

In this paper, we investigate the use of a switched split vector quantiser (SSVQ) for coding linear predictive coding (LPC) parameters. The SSVQ is applied to quantise the LPC parameters in terms of line spectral frequencies from the TIMIT database and its performance is compared with the split vector quantiser. Experimental results show that the SSVQ provides a better trade-off between bit-rat...

JPEG XR is considered as a lossy sample data compression scheme in the context of iris recognition techniques. It is shown that by optimising the JPEG XR quantisation strategy, JPEG XR default quantisation as well as JPEG2000 based iris recognition can be improved in terms of EER. The optimised JPEG XR quantisation strategy shows good performance across a wide range of iris feature extraction t...

In quantum physics, the operators associated with the position and the momentum of a particle are unbounded operators and C∗-algebraic quantisation does therefore not deal with such operators. In the present article, I propose a quantisation of the Lie-Poisson structure of the dual of a Lie algebroid which deals with a big enough class of functions to include the above mentioned example. As an ...

In this paper, we present a non-linear method of embedding a signature in colour and monochrome images and demonstrate its recovery. The embedding process can be viewed as pseudo-random perturbations to angles between vector elements. The derived angles are dithered by the addition of a watermark, and encoded as a pseudo-noise sequence of dither angle offsets. This is followed by a re-quantisat...