Image processing of blurred medical images

Continued research into several rheumatic diseases such as scleroderma, dermatomysitis and mixed connective tissue diseases is attempting to piece together the links between capillary structure and these diseases. A diagnostic technique that has become popular is Nailfold Capillaroscopy, which analyses the landscape of the capillary bed at the base of the finger nail using a video camera to store pictures. The problem to which this work is devoted is one of artefact removal. The captured images are not always of a good enough quality for diagnosis to be undertaken, often blurred and vague images of poor contrast are produced. The principal objectives of this research are to produce an image processing system that is usable and portable and independent of specialised image processing hardware. Secondly, to provide a variety of generalised image processing filters that may be applied to captured images dependent on the user's requirements, which is user friendly. Furthermore, to suggest a deblurring philosophy that will be effective on nailfold capillaries. Various techniques are presented for removing and reducing this blur in images that have been obtained by acoustical or other techniques. A number of algorithms have been developed and coded in the programming language 'C'. These algorithms allow the user of the system to process the image with the minimum of effort and computer literacy.<
>