Abstract: A security system comprises a computer network and a plurality of sensors each connected to the computer network at a respective network address and each generating sensing data. A managing component is connected with the network and communicates with the sensors by access thereof through the associated network address on the network, and processes sensor information received from said sensors. The managing component has a display with an interface screen showing to a user all the sensors in the security system, and an input device through which the user can enter interactive instructions to the managing component. The managing component controls communications to and from the sensors and has a rules engine storing rules therein. Each of the rules being associated with a respective device on the network, and causing the managing computer to take an action in response to output from at least one of the devices.

Abstract: Method and apparatus for dynamically placing sensors in a 3D model is provided. Specifically, in one embodiment, the method selects a 3D model and a sensor for placement into the 3D model. The method renders the sensor and the 3D model in accordance with sensor parameters associated with the sensor and parameters desired by a user. In addition, the method determines whether an occlusion to the sensor is present.

Abstract: A method and apparatus for high-resolution 3D imaging ladar system which can penetrate foliage and camouflage to sample fragments of concealed surfaces of interest is disclosed. Samples collected while the ladar moves can be integrated into a coherent object shape. In one embodiment, a system and method for automatic data-driven registration of ladar frames, comprises a coarse search stage, a pairwise fine registration stage using an iterated closest points algorithm, and a multi-view registration strategy. After alignment and aggregation, it is often difficult for human observers to find, assess and recognize objects from a point cloud display. Basic display manipulations, surface fitting techniques, and clutter suppression to enhance visual exploitation of 3D imaging ladar data may be utilized.

Abstract: Method for tracking an object recorded within a selected frame of a sequence of frames of video data, using a plurality of layers, where at least one object layer of the plurality of layers represents the object includes initializing layer ownership probabilities for pixels of the selected frame using a non-parametric motion model, estimating a set of motion parameters of the plurality of layers for the selected frame using a parametric maximization algorithm and tracking the object. The non-parametric motion model is optical flow and includes warping the mixing probabilities, the appearances of the plurality of layers, and the observed pixel data from the pixels of the preceding frame to the pixels of the selected frame to initialize the layer ownership probabilities for the pixels of the selected frame.

Abstract: The present invention provides a system and method for processing real-time rapid capture, annotation and creation of an annotated hyper-video map for environments. The method includes processing video, audio and GPS data to create the hyper-video map which is further enhanced with textual, audio and hyperlink annotations that will enable the user to see, hear, and operate in an environment with cognitive awareness. Thus, this annotated hyper-video map provides a seamlessly navigable, situational awareness and indexable high-fidelity immersive visualization of the environment.

Abstract: A system for tracking at least one object, comprising: a plurality of communicatively connected visual sensing units configured to capture visual data related to the at least one object; and a manager component communicatively connected to the plurality of visual sensing units, where the manager component is configured to assign one visual sensing unit to act as a visual sensing unit in a master mode and at least one visual sensing unit to act as a visual sensing unit in a slave mode, transmit at least one control signal to the plurality of visual sensing units, and receive the visual data from the plurality of visual sensing units.

Abstract: A sentient system combines detection, tracking, and immersive visualization of a cluttered and crowded environment, such as an office building, terminal, or other enclosed site using a network of stereo cameras. A guard monitors the site using a live 3D model, which is updated from different directions using the multiple video streams. As a person moves within the view of a camera, the system detects its motion and tracks the person's path, it hands off the track to the next camera when the person goes out of that camera's view. Multiple people can be tracked simultaneously both within and across cameras, with each track shown on a map display. The track system includes a track map browser that displays the tracks of all moving objects as well as a history of recent tracks and a video flashlight viewer that displays live immersive video of any person that is being tracked.

Abstract: The present invention is embodied in a video flashlight method. This method creates virtual images of a scene using a dynamically updated three-dimensional model of the scene and at least one video sequence of images. An estimate of the camera pose is generated by comparing a present image to the three-dimensional model. Next, relevant features of the model are selected based on the estimated pose. The relevant features are then virtually projected onto the estimated pose and matched to features of the image. Matching errors are measured between the relevant features of the virtual projection and the features of the image. The estimated pose is then updated to reduce these matching errors. The model is also refined with updated information from the image. Meanwhile, a viewpoint for a virtual image is selected. The virtual image is then created by projecting the dynamically updated three-dimensional model onto the selected virtual viewpoint.

Abstract: A method and apparatus for detecting objects (e.g., bags, vehicles, etc.) left in a field of view are disclosed. A long-term representation and a short-term representation of the field of view are constructed, and a difference between the long-term representation and the short-term representation is calculated. One or more criteria may optionally be applied to this difference to determine whether the difference represents an object that was left in the field of view.

Abstract: A method and apparatus for high-resolution 3D imaging ladar system which can penetrate foliage and camouflage to sample fragments of concealed surfaces of interest is disclosed. Samples collected while the ladar moves can be integrated into a coherent object shape. In one embodiment, a system and method for automatic data-driven registration of ladar frames, comprises a coarse search stage, a pairwise fine registration stage using an iterated closest points algorithm, and a multi-view registration strategy. After alignment and aggregation, it is often difficult for human observers to find, assess and recognize objects from a point cloud display. Basic display manipulations, surface fitting techniques, and clutter suppression to enhance visual exploitation of 3D imaging ladar data may be utilized.

Abstract: A scalable architecture for providing real-time multi-camera distributed video processing and visualization. An exemplary system comprises at least one video capture and storage system for capturing and storing a plurality of input videos, at least one vision based alarm system for detecting and reporting alarm situations or events, and at least one video rendering system (e.g., a video flashlight system) for displaying an alarm situation in a context that speeds up comprehension and response. One advantage of the present architecture is that these systems are all scalable, such that additional sensors (e.g., cameras, motion sensors, infrared sensors, chemical sensors, biological sensors, temperature sensors and like) can be added in large numbers without overwhelming the ability of security forces to comprehend the alarm situation.

Abstract: Method and apparatus for dynamically placing sensors in a 3D model is provided. Specifically, in one embodiment, the method selects a 3D model and a sensor for placement into the 3D model. The method renders the sensor and the 3D model in accordance with sensor parameters associated with the sensor and parameters desired by a user. In addition, the method determines whether an occlusion to the sensor is present.

Abstract: A system and method for accurately mapping between image coordinates and geo-coordinates, called geo-spatial registration. The system utilizes the imagery and terrain information contained in the geo-spatial database to precisely align geodetically calibrated reference imagery with an input image, e.g., dynamically generated video images, and thus achieve a high accuracy identification of locations within the scene. When a sensor, such as a video camera, images a scene contained in the geo-spatial database, the system recalls a reference image pertaining to the imaged scene. This reference image is aligned very accurately with the sensor's images using a parametric transformation. Thereafter, other information that is associated with the reference image can easily be overlaid upon or otherwise associated with the sensor imagery.

Abstract: A system and method for accurately mapping between camera coordinates and geo-coordinates, called geo-spatial registration, using a Euclidean model. The system utilizes the imagery and terrain information contained in the geo-spatial database to precisely align geographically calibrated reference imagery with an input image, e.g., dynamically generated video images, and thus achieve a high accuracy identification of locations within the scene. When a sensor, such as a video camera, images a scene contained in the geo-spatial database, the system recalls a reference image pertaining to the imaged scene. This reference image is aligned very accurately with the sensor's images using a parametric transformation produced by a Euclidean model. Thereafter, other information that is associated with the reference image can easily be overlaid upon or otherwise associated with the sensor imagery.

Abstract: A method and apparatus for providing immersive surveillance wherein a remote security guard may monitor a scene using a variety of imagery sources that are rendered upon a model to provide a three-dimensional conceptual view of the scene. Using a view selector, the security guard may dynamically select a camera view to be displayed on his conceptual model, perform a walk through of the scene, identify moving objects and select the best view of those moving objects and so on.

Abstract: The present invention is embodied in a video flashlight method. This method creates virtual images of a scene using a dynamically updated three-dimensional model of the scene and at least one video sequence of images. An estimate of the camera pose is generated by comparing a present image to the three-dimensional model. Next, relevant features of the model are selected based on the estimated pose. The relevant features are then virtually projected onto the estimated pose and matched to features of the image. Matching errors are measured between the relevant features of the virtual projection and the features of the image. The estimated pose is then updated to reduce these matching errors. The model is also refined with updated information from the image. Meanwhile, a viewpoint for a virtual image is selected. The virtual image is then created by projecting the dynamically updated three-dimensional model onto the selected virtual viewpoint.

Abstract: A system and method for accurately mapping between image coordinates and geo-coordinates, called geo-spatial registration. The system utilizes the imagery and terrain information contained in the geo-spatial database to precisely align geodetically calibrated reference imagery with an input image, e.g., dynamically generated video images, and thus achieve a high accuracy identification of locations within the scene. When a sensor, such as a video camera, images a scene contained in the geo-spatial database, the system recalls a reference image pertaining to the imaged scene. This reference image is aligned very accurately with the sensor's images using a parametric transformation. Thereafter, other information that is associated with the reference image can easily be overlaid upon or otherwise associated with the sensor imagery.

Abstract: A method and apparatus of operating a CT imaging apparatus having a cone beam radiation source and a 2D detector arrangement, for exactly reconstructing an image of a 3D region of interest (ROI) in an object. Measurement data acquired by the 2D detector is processed to develop sub-sets of Radon data. Each of the sub-sets of Radon data is targeted for reconstructing a corresponding local 2D ROI in a 2D parallel projection of the object. Accordingly, after a sub-set of Radon data is completely developed, it is subjected to a first inversion processing step for developing the corresponding local 2D ROI. Multiple ones of the local ROIs are then grouped together and subjected to a second inversion processing step to develop an image reconstruction of a part of the 3D ROI in the object.

Abstract: A computed tomographic imaging apparatus performs three-dimensional (3D) image reconstruction of a region of interest of an object using a plurality of processors for processing successive sets of cone beam measurement data that are acquired by scanning about the object with a cone beam radiation source and an area detector. A central memory has stored therein a plurality of subsets of pre-calculated image processing information, and the plurality of processors are responsive to a given sequence of successive ones of the subsets of pre-calculated image processing information for converting the cone beam measurement data to Radon derivative data on a plurality of Radon .phi.-planes.

Abstract: A method and apparatus are provided for three dimensional computerized tomographic imaging of an object, wherein cone beam energy from a cone beam source is applied to at least a portion of an object to be imaged by causing relative rotational motion between the source and the object at a plurality of source positions along a path scanning trajectory. An area detector receives cone beam energy from the source on a plurality of detector elements arranged in an array of rows and columns, so that its detector elements develop a set of measurement signals at each of the plurality of source positions. Derivative Radon transform data is directly developing by multiplication of the measurement signals of each set by pre-calculated and stored weight factors. Thereafter, an image of the object is reconstructed using the directly developed Radon derivative data.