MORPHOLOGICAL IMAGE ENLARGEMENTS

This paper presents a method based in mathematical morphology to enlar ge images. It does not make the low pass assumption which is common to all linear interpolation methods and which does not often hold for im ages. Pixels in smooth areas are properly interpolated by linear metho ds while those at the edges are not. The method begins with a linear i nterpolation and a gradient computation. The gradient serves as a meas ure of confidence about the linear interpolation. Then, the proposed a lgorithm processes the pixels in a certain order: first pixels with hi gh confidence (smooth zones) of the image and those with a low one (ed ges) at the end. By doing so, it preserves both slow variations and sh arp edges. The method can be applied to other image processing problem s, such as edge enhancement or motion vector estimation, where there i s an image and confidence information about each pixel. (C) 1997 Acade mic Press.