Abstract: A feature-based method and system for blur estimation in eye images. A blur estimation can be performed from eye/iris images in order to produce de-blurred images that are more useful for biometric identification. The eye/iris region, in particular the edge between the iris and pupil regions, can be utilized. The pattern of shutter motion or a characterization of the optical system can be utilized. By capturing a burst of images, or a video stream, one can use eye position in the images before and after a given capture to predict the motion of the eye within that capture. Because the before/after image frames need only contain the information necessary to locate the eye, and need not contain sufficient information to perform matching, the capture of these images can be accomplished with a wider range of settings.

Abstract: Methods and apparatus are provided for dividing an image into a plurality of image chips for presentation on a display. Potential objects of interest are detected within an image by detecting features therein that correspond to objects of interest. The image is uniformly divided into a plurality of preliminary image chips. Triage image chips are generated by automatically adjusting each preliminary image chip such that the potential objects of interest detected within each preliminary image chip are at least substantially centered in each preliminary image chip.

Abstract: A method, system and program product for the detection of removed objects in video images is disclosed. An object (or objects) is within a fixed camera's FOV (Field of View), is designated to be monitored. If an object is not designated, an algorithm can be used to segment and identify all the objects in the image that is acquired by a fixed camera. In such case, all these objects can be monitored. A reference frame around the monitored object that also includes the surrounding background can be determined. The object can be extracted out of the reference frame, thereby creating a “reference with hole” image and an object reference image. The hole in the “reference with hole” image can be filled by either an in-painting algorithm or a texture synthesis algorithm depending on the complexity of the surrounding background.

Abstract: A detection system which provides for continuous background estimation removal from a sequence of spectra. A panoramic field of regard may be partitioned into a large number of fields of view (FOVs). An FOV may have a chemical vapor cloud. The small FOV may maximize detection of the cloud. Such detection may require removing the spectral characteristics other than that of the target cloud. This may amount to removal of background spectra with an estimated background developed from one or more FOVs which may or may not be similar to the background of the FOV with the target cloud. A number of estimated background spectra of the other FOVs may be used individually to greatly increase the detection probability of the target chemical.

Abstract: In one embodiment, a method is disclosed that includes obtaining at least one measurement in a spectral domain of a sample and computing one or more measurements of the salient features in the spectral domain. The salient features correspond to at least one peak within the spectral domain. This method also includes classifying the computed salient features against a feature signature of nitric acid. In addition, this method includes determining if the chemical is present in the sample.

Abstract: A system and method for providing information for autonomous vehicle navigation are disclosed. The system comprises at least one laser scanner configured to perform one or more range and intensity scans of an area around the autonomous vehicle, and a geo-location unit comprising one or more global positioning system sensors and inertial navigation system sensors. The system also includes at least one processor in operative communication with the laser scanner and the geo-location unit.

Abstract: A method and system provide road and obstacle detection in navigating an autonomous vehicle. The method comprises scanning a distance ahead of the autonomous vehicle to obtain a current range scan, and obtaining navigation data, including dynamics, position, and orientation measurements of the autonomous vehicle. The current range scan is transformed to world coordinates with respect to a reference location based on the navigation data, and the transformed current range scan is input into a distance-based accumulator. The transformed current range scan is added to a variable size buffer when the autonomous vehicle is deemed to be non-stationary. A ground plane is estimated from the transformed current range scan and prior range scans stored in the variable size buffer. The estimated ground plane is represented as a constrained quadratic surface, which is classified into one or more of a traversable area, a non-traversable area, or an obstacle area for navigation of the autonomous vehicle.

Abstract: A combination of inter-frame and intra-frame and sequential classification methods provide for robust and flexible object classification. Intra-frame classification determines the object type based on features extracted from a single image. Inter-frame classification determines object types based on the features extracted from several sequential frames. A sequential classification combines the results of the inter- and intra-frame classifications over a sequence of images.

Abstract: An image is processed by a sensed-feature-based classifier to generate a list of objects assigned to classes. The most prominent objects (those objects whose classification is most likely reliable) are selected for range estimation and interpolation. Based on the range estimation and interpolation, the sensed features are converted to physical features for each object. Next, that subset of objects is then run through a physical-feature-based classifier that re-classifies the objects. Next, the objects and their range estimates are re-run through the processes of range estimation and interpolation, sensed-feature-to-physical-feature conversion, and physical-feature-based classification iteratively to continuously increase the reliability of the classification as well as the range estimation. The iterations are halted when the reliability reaches a predetermined confidence threshold.

Abstract: A debinarizer and a hardware-based method for debinarizing a binary bit stream into symbols comprising predefined binary sequences are disclosed in which predefined binary sequences and their corresponding symbols are represented as tree structures. A binary sequence is debinarized by traversing one of the tree structures to determine a corresponding symbol. Similar tree structures may be combined to form a single tree structure. The tree structures may be implemented as a finite state machine.

Abstract: A video analytics system for characterization of a flare. A video of a flare may be taken for obtaining information so as to appropriately control the flare in an interest of reducing emissions not necessarily favorable to the environment. The system may incorporate a control scenario involving one or more parameters of a flare which are to be controlled in view of a flare characterization from an algorithmic analysis of the video.

Abstract: In an embodiment, a video processor is configured to identify anomalous or abnormal behavior. A hierarchical behavior model based on the features of the complement of the abnormal behavior of interest is developed. For example, if the abnormal behavior is stealing or shoplifting, a model is developed for the actions of normal shopping behavior (i.e., not stealing or not shoplifting). Features are extracted from video data and applied to an artificial intelligence construct such as a dynamic Bayesian network (DBN) to determine if the normal behavior is present in the video data (i.e, the complement of the abnormal behavior). If the DBN indicates that the extracted features depart from the behavior model (the complement of the abnormal behavior), then the presence of the abnormal behavior in the video data may be assumed.

Abstract: A method, system and program product for the detection of removed objects in video images is disclosed. An object (or objects) is within a fixed camera's FOV (Field of View), is designated to be monitored. If an object is not designated, an algorithm can be used to segment and identify all the objects in the image that is acquired by a fixed camera. In such case, all these objects can be monitored. A reference frame around the monitored object that also includes the surrounding background can be determined. The object can be extracted out of the reference frame, thereby creating a “reference with hole” image and an object reference image. The hole in the “reference with hole” image can be filled by either an in-painting algorithm or a texture synthesis algorithm depending on the complexity of the surrounding background.

Abstract: A method for motion detection includes capturing a plurality of frame images. The plurality of frame images is preprocessed. A first background reference is generated based on a first subset of the plurality of frame images. A second background reference is generated based on a second subset of the plurality of frame images, wherein the second subset of the plurality of frame images is a subset of the first subset of the plurality of frame images. A first motion result is generated based on the first background reference and a second motion result is generated based on the second background reference. The first motion result and the second motion result are combined to generate a combined result. Object motion is detected based on the combined result.

Abstract: An automated tailgating detection system for an entry through a door or other entry port, including an entry authorization sensor detecting valid entry credentials and transmitting the number of valid entry credentials detected. An automated video surveillance system detects, counts and tracks the persons who enter through the door. A tailgate detection processor receives the number of valid entry credentials detected and the number of people detected and sounds an alarm when the credentials number is less than the number of persons. A door contact sensor determines the state of the door, and sends a signal when the door is opened and closed.

Abstract: A sensor network provides the ability to detect, classify and identify a diverse range of agents over a large area, such as a geographical region or building. The network possesses speed of detection, sensitivity, and specificity for the diverse range of agents. Different functional level types of sensors are employed in the network to perform early warning, broadband detection and highly specific and sensitive detection. A high probability of detection with low probability of false alarm is provided by the processing of information provided from multiple sensors. A Bayesian net is utilized to combine probabilities from the multiple sensors in the network to reach a decision regarding the presence or absence of a threat. The network is field portable and capable of autonomous operation. It also is capable of providing automated output decisions.

Abstract: The present invention pertains to video playback systems, devices and methods for searching events contained within a video image sequence. A video playback system in accordance with an illustrative embodiment of the present invention includes a video playback device adapted to run a sequential searching algorithm for sequentially presenting video images within an image sequence, and a user interface that can be used by an operator to detect the occurrence of an event contained within the image sequence. Methods of searching for an event of interest contained within an image sequence are also disclosed herein.

Abstract: A sensor network provides the ability to detect, classify and identify a diverse range of agents over a large area, such as a geographical region or building. The network possesses speed of detection, sensitivity, and specificity for the diverse range of agents. Different functional level types of sensors are employed in the network to perform early warning, broadband detection and highly specific and sensitive detection. A high probability of detection with low probability of false alarm is provided by the processing of information provided from multiple sensors. A Bayesian net is utilized to combine probabilities from the multiple sensors in the network to reach a decision regarding the presence or absence of a threat. The network is field portable and capable of autonomous operation. It also is capable of providing automated output decisions.

Abstract: A sensor network provides the ability to detect, classify and identify a diverse range of agents over a large area, such as a geographical region or building. The network possesses speed of detection, sensitivity, and specificity for the diverse range of agents. Different functional level types of sensors are employed in the network to perform early warning, broadband detection and highly specific and sensitive detection. A high probability of detection with low probability of false alarm is provided by the processing of information provided from multiple sensors. A Bayesian net is utilized to combine probabilities from the multiple sensors in the network to reach a decision regarding the presence or absence of a threat. The network is field portable and capable of autonomous operation. It also is capable of providing automated output decisions.

Abstract: A sensor network provides the ability to detect, classify and identify a diverse range of agents over a large area, such as a geographical region or building. The network possesses speed of detection, sensitivity, and specificity for the diverse range of agents. Different functional level types of sensors are employed in the network to perform early warning, broadband detection and highly specific and sensitive detection. A high probability of detection with low probability of false alarm is provided by the processing of information provided from multiple sensors. A Bayesian net is utilized to combine probabilities from the multiple sensors in the network to reach a decision regarding the presence or absence of a threat. The network is field portable and capable of autonomous operation. It also is capable of providing automated output decisions.