| 摘要 |
Video surveillance systems are used to observe oftentimes large-area, broken or complex surveillance areas. The image data streams recorded with the surveillance camera are usually brought together in a surveillance center or the like and are either automated there or controlled by surveillance personnel. The invention relates to an image processing module (3) for estimating an object position of a surveillance object or subareas thereof in a surveillance area for a surveillance system (1) for surveilling at least said surveillance area by means of at least one surveillance camera (2). Said image processing module comprises a model input interface (6) for receiving a model or partial model of the surveillance area (referred to in the following as model), a camera input interface (7) for receiving a camera model of the surveillance camera (2), an object input interface for receiving an object point of the surveillance object, said object points being determined based on a surveillance image recorded from one or more image points of the surveillance object by means of the surveillance camera. The image processing module (5) is adapted to determine the object position of the surveillance object or subareas thereof in the surveillance area by calculation of the model, the camera model and the object point, the model being configured as a 3D model so that the object position can be determined as a 3D object position. |
