METHOD AND SYSTEM FOR RECOGNISING AN OBJECT, AND METHOD AND SYSTEM FOR GENERATING A MARKING IN A SCREEN REPRESENTATION BY MEANS OF A NON-CONTACT GESTURE-CONTROLLED SCREEN POINTER

摘要

The invention relates to a method for recognising an object in a current image of a sequence of images, wherein the object was recognised at least in an image section of an earlier image of the sequence of images, comprising the following steps: selecting (746) an image section of the current image; determining (748) a degree of similarity between the image section of the current image and the image section of the earlier image in which the object was recognised, by means of a comparison of the same; and recognising (750) the object in the image section of the current image (620) if the degree of similarity meets a specified condition of similarity.