Process for restoring the inner volumes of a solid with a view to computations and associated application

DETAILED DESCRIPTION The object of the present invention is to provide a remedy for the drawbacks mentioned above.
The invention particularly allows complex volumes to be accurately restored and permits digital computations to be made from such volumes.
The present application is advantageously, but not exclusively, applied to concealed volumes located inside a solid.
It is therefore possible to restore and to analyze volumes whose geometry is not precisely known.
The invention allows for example computations to be performed on pipes embedded in a solid for displaying any wear or a layer formed during operation.
It is thus possible to precisely restore the shape of inner pipes after such a pipe has been used for a certain time, and to carry out computations directly from these concealed and a prior unknown volumes.
The invention is here applied to the field of non-destructive testing.
The main advantage of the invention is that it allows complex volumes to be processed with considerable less time when compared with the prior art.
In fact, as contained in the description above, when complex volumes are to be subjected to scientific computations, it takes a long time to integrate their geometry into storage in the computer memory.
The present invention allows a nearly automatic fast acquisition of data representing any geometry.
The main objects and advantages which have been mentioned above are obtained by means of an automatic reconstitution process for volumes located notably inside solids, in order to carry out computations relating to the volumes.
According to the invention, the process comprises the following stages: obtaining digitized images of several parallel cross-sections of the space surrounding the volume; processing the digitized images to create a digital contour of the volume; and digitally processing the contour.
More precisely the processing of the digital images to create a digital contour comprises: determining the contour of the volume of each digitized image; approximating with polygons the determined contours; locating vertices of the polygons; and creating, from data representing the vertices, a surface mesh pattern approximating a real contour of the volume