Searching for objects in digital images using colour

In order to perform a visual search for an object two capabilities must be available: the ability to recognise the object when it comes into view; and a way of selecting these views. Although visual search is often complicated by object occlusion and low spatial resolution, colour may be used as a cue since it is salient and stable under reduced resolution. This book explores the use of colour in visual search for objects in digital images at low and high spatial resolutions. The book describes the visual search process and its confounding factors, as well as reviews several colour constancy, image segmentation and object recognition algorithms from the literature. It then presents three visual search algorithms that can represent 2 or 3- dimensional planar or deformable objects which may be: affine or perspectively distorted, occluded and under fluorescence or tungsten lighting. This book caters to researchers, practitioners and students interesting in: visual search, colour and colour constancy, colour object recognition, syntactic neural networks and pattern matching.

Autorentext

Dr. Paul Walcott holds a Ph.D. in Information Engineering from City University (UK) and is a Computer Science Lecturer at the University of the West Indies (Barbados). He teaches courses in software engineering, information systems and e-commerce and conducts research in: computer vision; computers and education; and e-commerce.