next up previous contents index
Next: bicontrast Up: Reference Previous: amle   Contents   Index

amle_init

$ \bigcirc$Name


amle_init Compute initial data for the level line image interpolation scheme (AMLE)




$ \bigcirc$Command Synopsis


amle_init in delta out



in : Input uniformly quantized fimage

delta : Width step of the uniform quantization used for the input image

out : Output fimage as initial data (input) for the AMLE model




$ \bigcirc$Function Summary


void amle_init (in , delta , out )

Fimage in , out ;

float delta ;




$ \bigcirc$Description


This module prepares an uniformly quantized image to be the input of the module amle which reconstructs an image by interpolating the missing level lines from the ones kept (see the main reference [CMS98] and [Aro67][Cao98][Fro99] for related works).

The preparation consists of building an image with 0 everywhere but at pixel locations (x, y) adjoining the remaining level lines (which are borders of level sets). At these locations, the value of the pixels out(x, y) is given by the value $ \lambda$ of the nearest level line in in(x, y). Because of the quantization process, adjacent level lines are multiple of delta = $ \delta$ where $ \delta$ corresponds the quantization step, and in the input image only the last level line of value $ \mu$ > = $ \lambda$ is recorded. The value $ \lambda$ is therefore recovered using the following scheme:

if in(x, y) > $ \mu$ then out(x, y) = $ \mu$ + $ \delta$, else out(x, y) = in(x, y).




$ \bigcirc$Version 1.1


Last Modification date : Thu Nov 29 20:23:56 2001


$ \bigcirc$Author


Jacques Froment






next up previous contents index
Next: bicontrast Up: Reference Previous: amle   Contents   Index
mw 2004-05-05