Processor-supported detection of selective target object in image

摘要

The method involves using a colour table including predetermined colours to form images of objects. A target object is classified from the objects in a received image using characteristics of the target object. The spatial position of the target object is determined. A rectangle is pref. formed around the target to delimit a block of data including the object, and the user can draw around the rectangle. An image of the background is pref. taken prior to the process, and the object is classified as a target when the background does not contain the target object. A direction-dependent hysteresis filter may be used, and a movement prediction may be performed for the object.