| 摘要 |
The invention relates to a classification method of an object's image belonging to a category of objects, said method comprising a prior step consisting of obtaining a subset (E<SUB>j</SUB>) of object's images of said category, said subset being associated to a classification criterion, and said process being characterized in that it also comprises: - a calculation step (b1) of the distance between said image and said subset (E<SUB>j</SUB>) according to the formula: Dist(q, E<SUB>j</SUB>) = DMC({v<SUB>1</SUB>, V<SUB>2</SUB>,..., v<SUB>N</SUB>}U{q}) - DMC({v<SUB>1</SUB>, V<SUB>2</SUB>,..., v<SUB>N</SUB>}) in which: - Dist(q, E<SUB>j</SUB>) is the distance between an image vector q corresponding to said image and the vector of the images of the subset E<SUB>j</SUB>, - (V<SUB>1</SUB>, v<SUB>2</SUB>,..., v<SUB>N</SUB>} are the vectors V<SUB>1</SUB> to v<SUB>N</SUB> of the images of said subset E<SUB>j</SUB>, - and DMC(x<SUB>1</SUB>,...,x<SUB>p</SUB>}) is an operator which gives the determinant of the covariance matrix of the vectors X<SUB>1</SUB> to X<SUB>p</SUB>, assimilated to P observations of a random vector variable X<SUB>i</SUB>, i being an index varying from 1 to P, - and a comparison step (b2) of the thus calculated distance having a decision threshold (dj). |
