A CONTEXTUAL IMAGE SEGMENTATION SYSTEM USING A-PRIORI INFORMATION FORAUTOMATIC DATA CLASSIFICATION IN NUCLEAR-PHYSICS

This paper presents an original approach to solve an automatic data cl assification problem by means of image processing techniques. The clas sification is achieved using image segmentation techniques for extract ing the meaningful classes. Two types of information are merged for th is purpose: the information contained in experimental images and a pri ori information derived from underlying physics (and adapted to image segmentation problem). This data fusion is widely used at different st ages of the segmentation process. This approach yields interesting res ults in terms of segmentation performances, even in very noisy cases. Satisfactory classification results are obtained in cases where more ' 'classical'' automatic data classification methods fail.