| 摘要 |
A method for the detection of a marker affixed to the surface of an object, includes: acquiring a photographic image of a zone containing at least one marker; extracting contours from the photographic image; detecting a marker-positioning contour from among the aforementioned contours, which corresponds to a set of confocal conics having a predetermined form; modelling an image of the confocal conics in the form of a matrix having at least two symmetric matrices of order 3; analyzing the matrices in order to determine a perspective correction for the photographic image; detecting, in the perspective-corrected photographic image, an indexing element associated graphically with the positioning contour of the marker, and rotation-correcting the photographic image; detecting, in the perspective and rotation-corrected photographic image, encoded data graphically associated with the positioning contour of the marker, and decoding the encoded data. A marker-detection system using the method is also described. |
