Title :
Contour based image coding
Author_Institution :
Dept. of Electron. & Electr. Eng., Strathclyde Univ., Glasgow, UK
Abstract :
Illustrates the value of contours of constant intensity in the compression of digital images. Contours have been used for many years in cartography, where a sparse set of smooth connected curves convey a very accurate impression of the landscape. The compression ratios of existing coding techniques appear to have reached a saturation level of about 10:1. It has been suggested, Kunt (1985), that in order to make further significant advances in image coding it will be necessary to adopt radical new approaches in which scene content is explicitly coded. The author suggests the use of contours of constant intensity as the primitives to be extracted both for image coding purposes and image analysis. Images are coded by retaining a set of contours describing the scene content. The images are then reconstructed by interpolating between the contours
Keywords :
encoding; picture processing; compression ratios; constant intensity contours; digital image compression; image analysis; image coding;
Conference_Titel :
Low Bit Rate Image Coding, IEE Colloquium on
Conference_Location :
London