A system for automatically tracking objects through a video sequence uses a method which automatically tracks an object by estimating the location of the object in a subsequent frame, comparing test windows in the subsequent frame with an object window encompassing the object in the original frame, and selecting the best match window which is most similar to the object window. The location of the best match window corresponds to the location of the object in each frame of the video sequence. The resulting location information can be combined with the video sequence to create interactive video applications in which an user of the application can select individual objects in the interactive video sequence.

An intruding object monitoring method using an intruding object monitoring system having a first intruding object monitoring apparatus for monitoring a monitor area by inputting images of the area and at least one second intruding object monitoring apparatus for performing a following of an intruding object. By this method, first intruding object monitoring apparatus acquires detection information of an object intruding into the monitor area, the second intruding object monitoring apparatus captures the intruding object in accordance with the detection information supplied from the first intruding object monitoring apparatus. After the intruding object is captured, the second intruding object monitoring apparatus performs the following of the captured intruding object independently of the first intruding object monitoring apparatus.

The present invention defines the area where the targeted moving object can move in the three-dimensional real spatial coordinate system, then maps it to the two-dimensional area corresponding to the capturing method and the capturing range of the capturing device, and by limiting the image processing for obtaining the coordinate to that area, obtains the coordinate of the moving object in the three-dimensional real space rapidly by using the signal which is capturing the moving object. Furthermore, the present invention is able to control the capturing direction and the capturing range which correspond to the movement of the capturing target, to overlap the image signal with the data which is always changing by the capturing condition and the movement of the measurement data, to improve the detection precision, and to track automatically and smoothly.

A method and apparatus for detecting objects from a sequence of images of a scene containing an object by using two distinct methods for object detection. One is suited for well-lit scenes (e.g., daytime), while the other is suitable for poorly-lit scenes (e.g., nighttime) where the objects have lights mounted on them. Further, the method and apparatus are adaptive to the image statistics of the objects being detected, and can be programmed to filter out "weak" detections that lie below a certain percentile in the observed statistical distribution of image measures. The specific percentile is determined based on whether the scene has been determined to be well- or poorly-lit and whether it contains shadows or not.

A system and method is provided for processing video signals, tracking objects in view, and blanking fields of view in video surveillance systems. The system includes a camera having two rotational degrees of freedom, a display, and a combination of switches for routing the video signal. The displayed video signal is divided into an array of cells which may be individually masked, blanked, or blurred. The camera microcontroller captures and stores objects in view, and tracks such objects as the camera zooms, pans and tilts in spherical coordinates. Objects in view may be tracked and blanked by obscuring an agglomeration of the cells in response to a masking signal generated by the microcontroller.