Abstract: A method for capturing surveillance images includes of a secured area has a first camera with a first field of view and a second camera with a second field of view. An optical center is determined for the second field of view. A magnification ratio is determined between a target located in the first field of view relative to the second field of view. A pan and tilt model is determined that provides a relationship between a first camera coordinates and pan and tilt values to position the second camera at the optical center. A target is captured in the first field of view and the second camera is controlled to record the target in the second field of view with the second field of view being an enhanced view relative to the first field of view.

Abstract: A system and method for filtering video noise is provided. Video generated under low lighting conditions is susceptible to increased noise and graininess, which increases the required storage space required for recording the noisy video. The provided system and method for filtering video noise reduces noise in low-light video by estimating the noise profile associated with at least one video frame; discriminating between chrominance noise and luminance noise in said noise profile; applying predetermined filter criteria to said chrominance noise to estimate the parameters for luminance noise filtering; applying said luminance noise filtering parameters to filter the luminance noise; and generating a first filtered output of a frame by logical combination of outputs obtained from luminance noise filter and chrominance noise filter.

Abstract: A method is presented for enhancing an image from an initial image, comprising computing a first luminance level frequency distribution corresponding to plurality of pixel constructing said initial image, representing said first luminance level frequency distribution as a resultant of Gaussian model mixtures to assess relative utilization of overall luminance level across said initial image, computing a desired luminance level frequency distribution as a function of the relative utilization, computing a transfer function to adjust the first luminance level frequency distribution to an enhanced luminance level as a function of desired level frequency distribution estimation, and applying said transfer function globally to said initial image to provide an enhanced image is. The first luminance level can be adapted to be linearized in a logarithmic form. In one embodiment, the model comprises one or more Gaussian functions. The initial image can be a background image estimated from a sequence of images.

Abstract: Real-time image stabilization using computationally efficient corner detection and correspondence is provided. The real-time image stabilization performs a scene learning process on a first frame of an input video to obtain reference features and a detection threshold value. The presence of jitter is determined in a current frame of the input video by comparing features of the current frame against the reference features using the detection threshold value. If the current frame is found to be unstable, corner points are obtained from the current frame. The obtained corner points are matched against reference corner points of the reference features. If the number of matched corner points is not less than a match point threshold value, the current frame is modeled using random sample consensus. The current frame is corrected to compensate for the jitter based on the results of the modeling.

Abstract: A method is presented for enhancing an image from an initial image, comprising computing a first luminance level frequency distribution corresponding to plurality of pixel constructing said initial image, representing said first luminance level frequency distribution as a resultant of Gaussian model mixtures to assess relative utilization of overall luminance level across said initial image, computing a desired luminance level frequency distribution as a function of the relative utilization, computing a transfer function to adjust the first luminance level frequency distribution to an enhanced luminance level as a function of desired level frequency distribution estimation, and applying said transfer function globally to said initial image to provide an enhanced image is. The first luminance level can be adapted to be linearized in a logarithmic form. In one embodiment, the model comprises one or more Gaussian functions. The initial image can be a background image estimated from a sequence of images.

Abstract: A video camera is located at a secure access point and is connected to a computer network. The camera is used as a conventional security camera and also provides automated assistance with access control. When an identity badge is presented to the camera by a person requesting access to the facility, a computer that is monitoring the camera recognizes a bar code or other symbols on the badge and accesses a database to locate information associated with that identity badge. This information is then used to determine if the person should be granted access.

Abstract: Real-time image stabilization using computationally efficient corner detection and correspondence is provided. The real-time image stabilization performs a scene learning process on a first frame of an input video to obtain reference features and a detection threshold value. The presence of jitter is determined in a current frame of the input video by comparing features of the current frame against the reference features using the detection threshold value. If the current frame is found to be unstable, corner points are obtained from the current frame. The obtained corner points are matched against reference corner points of the reference features. If the number of matched corner points is not less than a match point threshold value, the current frame is modeled using random sample consensus. The current frame is corrected to compensate for the jitter based on the results of the modeling.

Abstract: A method for capturing surveillance images includes of a secured area has a first camera with a first field of view and a second camera with a second field of view. An optical center is determined for the second field of view. A magnification ratio is determined between a target located in the first field of view relative to the second field of view. A pan and tilt model is determined that provides a relationship between a first camera coordinates and pan and tilt values to position the second camera at the optical center. A target is captured in the first field of view and the second camera is controlled to record the target in the second field of view with the second field of view being an enhanced view relative to the first field of view.

Abstract: The invention consists of a system and method for transferring video images from video sources to video receivers using a video server as an intermediary. The video server creates a list of all the requests from the video receivers for a video image from a video source. The video server requests the video image from the video source and then sends the video image to each video receiver on the list. The video source is only required to send one image for any number of requests from video receivers and the video sources and video receivers have no direct connection to each other. The load of handling many video sources and many video receivers can be distributed amongst multiple video servers.

Abstract: A system and method for filtering video noise is provided. Video generated under low lighting conditions is susceptible to increased noise and graininess, which increases the required storage space required for recording the noisy video. The provided system and method for filtering video noise reduces noise in low-light video by estimating the noise profile associated with at least one video frame; discriminating between chrominance noise and luminance noise in said noise profile; applying predetermined filter criteria to said chrominance noise to estimate the parameters for luminance noise filtering; applying said luminance noise filtering parameters to filter the luminance noise; and generating a first filtered output of a frame by logical combination of outputs obtained from luminance noise filter and chrominance noise filter.

Abstract: An array of processor elements has multiple instruction streams and multiple data streams broadcast to all of the processor elements. The processor elements are each connected to multiple neighbouring processor elements within a cruciate neighbourhood. The architecture is suitable for use in fine-grained applications. The array may have a processor element for each pixel of an image. The array is preferably provided on a single integrated circuit having 10,000 or more processor elements.

Abstract: The invention consists of a system and method for transferring video images from video sources to video receivers using a video server as an intermediary. The video server creates a list of all the requests from the video receivers for a video image from a video source. The video server requests the video image from the video source and then sends the video image to each video receiver on the list. The video source is only required to send one image for any number of requests from video receivers and the video sources and video receivers have no direct connection to each other. The load of handling many video sources and many video receivers can be distributed amongst multiple video servers.

Abstract: A video camera is located at a secure access point and is connected to a computer network. The camera is used as a conventional security camera and also provides automated assistance with access control. When an identity badge is presented to the camera by a person requesting access to the facility, a computer that is monitoring the camera recognizes a bar code or other symbols on the badge and accesses a database to locate information associated with that identity badge. This information is then used to determine if the person should be granted access.