Image compression using the Distance Transform on Curved Space (DTOCS) and delaunay triangulation

This paper presents a new image compression method. It is based on the Distance Transform on Curved Space (DTOCS), which calculates an integer approximation of the weighted pseudo-Euclidean distance transform along discrete 8-paths. The decompression is done using the Delaunay triangulation. The overall performance of this scheme is comparable to DCT-based methods, giving better results in the smooth areas of the image. It is shown both analytically and experimentally that the time complexity of the new-algorithms is independent of the number of control points, i.e. the compression ratio. This makes the presented compression algorithm extremely well suitable for time-critical applications. Furthermore, it could well be used in applications where a rapid flashing image is made more accurate as time goes by.