| 摘要 |
The segmentation procedure of the invention consists in making up a data base (learning) from several populations of objects which are representative of classes to be discriminated and for which an estimate is made or a statistical law is available which defines them in terms of characteristic measurements. The data base is segmented by fixing cut-offs between the classes taken two by two, the cut-offs being calculated with the aid of descriptive parameters of the statistical laws for each class. Each new object is assigned to the most probable class. |
