IEE Proceedings - Vision, Image and Signal Processing

It is shown that morphological opening and closing by reconstruction are powerful tools in the analysis of image sequences. In the paper, they are used to build detection algorithms for fast moving objects, new objects and defects in image sequences. Possible applications are video coding, video surveillance and motion picture restoration View full abstract»

An algorithm is presented to enhance the resilience of block-based video-coding algorithms against channel errors. The error-resilience algorithm described suggests the use of fixed length coding to alleviate the effect of Huffman coding on the synchronisation of the decode under erroneous conditions. The proposed mechanism modifies the order of transmission of the fixed-length video parameters to... View full abstract»

An approach for separating uniform-coloured text strings (i.e. the characters in a text string are of the same colour) from colour graphic background images is presented. This approach overcomes various difficulties that arise due to colour complexity of graphic backgrounds and disturbance from background graphic objects. With this approach, the character-like components in a created binary edge i... View full abstract»

The paper discusses a method for estimating the Hermite coefficients of a discrete-time one-dimensional signal. To estimate the Hermite coefficients a solution based on Gaussian quadratures is used. The paper also looks at various least mean squared (LMS) estimation methods to estimate two further parameters which are incorporated into the orthonormal Hermite basis function; a spread term and a sh... View full abstract»

Forthcoming interactive video services such as video on demand will use pre-encoded bit streams for transmission. A great lack of flexibility arises when heterogeneous networks are used or when the user is allowed to use a bandwidth of his/her own choice. In the paper, a mechanism capable of decoupling video encoders from transmission-network constraints and user demands is proposed. The authors d... View full abstract»

The finding and maintaining of high accuracy foveation points for several types of recognised feature in log polar space such as a line, circular or elliptical arc is considered. Log polar space is preferred over cartesian space as it provides a high resolution and a wide viewing angle; feature invariance in the fovea simplifies foveation; it allows multi-resolution analysis; and rotation and scal... View full abstract»

Wavelet transforms have been one of the important signal processing developments in the last decade, especially for applications such as time-frequency analysis, data compression, segmentation and vision. Although several efficient implementations of wavelet transforms have been derived, their computational burden is still considerable. The paper describes two generic parallel implementations of w... View full abstract»

In modern multimedia systems, video and image signals usually need to be indexed or retrieved according to their contents. Colour characteristics are proposed for use in detection of human faces in colour images with complex backgrounds. The proposed face detection method first uses a neural network to classify the images and then segments the candidate face regions. Then, an energy thresholding m... View full abstract»

Ellipse detection is an interesting issue in machine vision researches and many approaches have been suggested to determine the parameters. Recently, a symmetric axis method has been proposed to determine the centre of an ellipse, but other parameters are determined otherwise. A consistent symmetric axis method is proposed that utilises the information inherent in the symmetric axes throughout the... View full abstract»

A new image compression approach is proposed in which variable block size technique is adopted, using quadtree decomposition, for coding images at low bit rates. In the proposed approach, low-activity regions, which usually occupy large areas in an image, were coded with a larger block size and the block mean is used to represent each pixel in the block. To preserve edge integrity, the classified ... View full abstract»