| 摘要 |
The invention concerns a method for processing a two-dimensional or three-dimensional image representing an object whereof the contours are disturbed by structured or unstructured noise. The invention is characterised in that it consists in extracting from the image, the active contour of the object by the following steps: a) analysing the unstructured noise, that is the background noise; b) analysing the structured noise, that is detecting the multiple reflections and the signature of the object; c) identifying the edges of the objects from the image obtained in b); and d) identifying the surfaces of the object from the raw image (or initial image) and from the images obtained at steps b) and c). |
