Abstract: A method and system for improving the accuracy and timeliness of video metadata by incorporating information related to the motion of the camera as derived from the video imagery itself. Frame-to-frame correspondences are used to accurately estimate changes in camera pose. While the method and system do not require geo-registration, geo-registration results, if available, may be considered in processing the video images and generating improved camera pose estimates.

Abstract: Apparatus and a concomitant method of identifying the direction of motion within a scene that is represented by a sequence of images, e.g., a series of frames in a video sequence. The apparatus contains a flow field generator, a flow field segmentor, and alarm detector and an alarm processor. The flow field generator processes the sequence of images and generates a substantially distortionless flow field representing the motion within a scene. The flow field is a vector representation of the motion of the scene that represents both the magnitude and the direction of the motion. The flow field is generated by correlating at least two frames in the sequence of images. This flow field is analyzed by a flow field segmentor to determine the magnitude and direction of motion with the scene and segment the motion information from the static portions of the scene. An alarm detector then processes the motion information to determine if an alarm should be generated based upon the motion information.

Abstract: A method for fixating a camera, mounted on a motorized mount, on a target point. The method includes receiving a plurality of images representative of a scene; selecting, within the plurality of images, a first display reference image containing the target point at a first coordinate location; estimating, in a current image within the plurality of images, a transformation that maps the current image to the first display reference image; estimating a second coordinate location of the target point in the current image using the transformation; computing an image slip between the target point in the current image and the target point in the first display reference image; converting the image slip into an angular correction for fixating the camera; modifying the angular correction by using closed-loop control with the motorized mount; and warping the current image using the transformation to align the current image with the first display reference image.

Abstract: An image stabilizer selectively adds image data from a background image to the current image to compensate for data in the current image that is missing due to a sudden shift in the current image relative to the previous images. The current image is warped into the coordinate system of the background image and then the warped current image is merged with the background image to replace any blank areas in the current image with corresponding pixel values from the background image. The image data from the background image which is to be substituted into the warped current image is subject to a low-pass filtering operation before it is merged with the warped current image. The warped current image is merged with the background image to form a modified background image which is then merged with the warped current image. The background image is, itself, warped to track camera motion in obtaining the current image before the background image is merged with the warped current image.

Abstract: A real-time modular video processing system (VPS) which can be scaled smoothly from relatively small systems with modest amounts of hardware to very large, very powerful systems with significantly more hardware. The modular video processing system includes a processing module containing at least one general purpose microprocessor which controls hardware and software operation of the video processing system using control data and which also facilitates communications with external devices. One or more video processing modules are also provided, each containing parallel pipelined video hardware which is programmable by the control data to provide different video processing operations on an input stream of video data. Each video processing module also contains one or more connections for accepting one or more daughterboards which each perform a particular image processing task.

Abstract: Digital signal processing circuitry implemented in ASICs or FPGAs is built by combining multi-component constructs (e.g. macrocells). These circuits may be modified to include a timing channel by augmenting selected ones of the constructs to include a path which propagates a timing signal with a delay that compensates for the signal processing delay through the construct. The selected constructs are those that are used in a critical processing path in the digital signal processing circuitry. A timing compensation circuit may also be defined as a construct. This block receives two digital data signals having accompanying timing signals and delays the first signal that provides valid data until the second signal also provides valid data, as determined by their timing signals. A configurable arithmetic and logic unit (ALU) made using these techniques includes a timing compensation circuit, a look-up table and an accumulator.

Abstract: Apparatus and a concomitant method of identifying the direction of motion within a scene that is represented by a sequence of images, e.g., a series of frames in a video sequence. The apparatus contains a flow field generator, a flow field segmentor, and alarm detector and an alarm processor. The flow field generator processes the sequence of images and generates a substantially distortionless flow field representing the motion within a scene. The flow field is a vector representation of the motion of the scene that represents both the magnitude and the direction of the motion. The flow field is generated by correlating at least two frames in the sequence of images. This flow field is analyzed by a flow field segmentor to determine the magnitude and direction of motion with the scene and segment the motion information from the static portions of the scene. An alarm detector then processes the motion information to determine if an alarm should be generated based upon the motion information.

Abstract: A pipelined parallel processor (PPP) integrated onto a single integrated circuit. The PPP includes a filtering unit, internal routing circuitry such as a crosspoint switch, an internal frame store controller, and one or more function circuits. The function circuits may be, for example, arithmetic units, lookup tables, timing compensators, adders/subtractors, statistics modules, image shifting circuitry, and other useful processing devices. The components of the PPP are interconnected with the crosspoint switch which routes data between the frame store controller, the filtering unit, function circuits, external input channels, and external output channels.