| 摘要 |
The invention relates to an image processing apparatus by means of which an image identification can take place in real time and which, with only very little or even no a priori information, is capable of carrying out a pertinent and reliable identification of objects. To this end, the connection probabilities are determined between two contour points in each case, taking into account the distance between the points, by means of a computation mechanism. Further provided is at least one classifier, which takes sets of calculated connection probabilities and selects from them subsets with at least three connection probabilities for possible links between at least three adjacent contour points, one of which is a previously determined central contour point, and, for each subset, sorts out that contour point which is adjacent to the central contour point and which has a possible link with the lowest connection probability to an adjacent contour point, provided that the link does not connect two points that are adjacent to the central point, and subsequently enters the contour points that have not been sorted out in a contour point list with connectors that identify the remaining links to the central point.
|
