Abstract: Described is a system for object recognition. The system generates a training image set of object images from multiple image classes. Using a training image set and annotated semantic attributes, a model is trained that maps visual features from known images to the annotated semantic attributes using joint sparse representations with respect to dictionaries of visual features and semantic attributes. The trained model is used for mapping visual features of an unseen input image to its semantic attributes. The unseen input image is classified as belonging to an image class, and a device is controlled based on the classification of the unseen input image.

Abstract: Described herein is a method for enhancing image data that includes dividing an image into multiple regions. The method includes measuring variations in pixel intensity distribution of the image to determine high pixel intensity variations for identifying an intensity-changing region. The method includes calculating a histogram of intensity distribution of pixel intensity values for the intensity-changing region without calculating a histogram of intensity distribution of pixel intensity values for each region of the multiple regions. The method also includes determining a transformation function based on the intensity distribution for the intensity-changing region. The method includes applying the transformation function to modify an intensity for each pixel in the image to produce an enhanced image in real time. The method also includes detecting in the enhanced image a horizon for providing to an operator of a vehicle an indication of the horizon in the image on a display in the vehicle.

Abstract: Described is system and method for visual media reasoning. An input image is filtered using a first series of kernels tuned to represent objects of general categories, followed by a second series of sparse coding filter kernels tuned to represent objects of specialized categories, resulting in a set of sparse codes. Object recognition is performed on the set of sparse codes to generate object and semantic labels for the set of sparse codes. Pattern completion is performed on the object and semantic labels to recall relevant meta-data in the input image. Bi-directional feedback is used to fuse the input data with the relevant meta-data. An annotated image with information related to who is in the input image, what is in the input image, when the input image was captured, and where the input image was captured is generated.

Abstract: Described is a system for library-based spectral demixing. The system simultaneously separates and identifies spectral elements in a set of noisy, cluttered spectral elements using Sparse Representation-based Classification (SRC) by modeling the set of noisy, cluttered spectral elements. The spectral library models each spectral element in the set of noisy, cluttered spectral elements, each spectral element having a corresponding wavenumber measurement. Wavenumber measurements are classified, resulting in salient wavenumber measurements. Target spectral elements representing a target of interest are identified in the set of noisy, cluttered spectral elements using the salient wavenumber measurements.

Abstract: Described herein is a method of enhancing an image includes determining a level of environmental artifacts at a plurality of positions on an image frame of image data. The method also includes adjusting local area processing of the image frame, to generate an adjusted image frame of image data, based on the level of environmental artifacts at each position of the plurality of positions. The method includes displaying the adjusted image frame.

Abstract: Described herein is a method of displaying an object that includes detecting a first location of the object on a first image frame of image data. The method also includes determining a second location of the object on a second image frame of the image data on which the object is not detectable due to an obstruction. The method includes displaying a representation of the object on the second image frame.

Abstract: Examples include methods, systems, and articles for localizing a vehicle relative to an imaged surface configuration. Localizing the vehicle may include selecting pairs of features in an image acquired from a sensor supported by the vehicle having corresponding identified pairs of features in a reference representation of the surface configuration. A three-dimensional geoarc may be generated based on an angle of view of the sensor and the selected feature pair in the reference representation. In some examples, a selected portion of the geoarc disposed a known distance of the vehicle away from the portion of the physical surface configuration may be determined. Locations where the selected portions of geoarcs for selected feature pairs overlap may be identified. In some examples, the reference representation may be defined in a three-dimensional space of volume elements (voxels), and voxels that are included in the highest number of geoarcs may be determined.

Abstract: The present disclosure generally relates to systems and methods for spectral demixing. An example technique includes obtaining empirical spectroscopic data representing a plurality of frequencies of electromagnetic energy that has interacted with a specimen, accessing a computer readable representation of a hierarchal spectral cluster tree representing a spectral library, demixing, with data on each of a plurality of levels of the hierarchal spectral cluster tree, foveated spectroscopic data derived from the empirical spectroscopic data, identifying at least one node in the hierarchal spectral cluster tree as corresponding to the empirical spectroscopic data, and outputting an endmember abundance assessment of the specimen corresponding to at least the at least one node.

Abstract: A method includes receiving image data at a first tracking system. The image data represents a region in an image of a sequence of images. The method also includes generating a first tracking fingerprint based on the image data. The method further includes comparing the first tracking fingerprint and a second tracking fingerprint. The method also includes providing an output from the first tracking system to a second tracking system based on a result of the comparison of the first tracking fingerprint and the second tracking fingerprint. The output includes an instruction associated with an object model stored at the second tracking system.

Abstract: Described is a system for recognition of unseen and untrained patterns. A graph is generated based on visual features from input data, the input data including labeled instances and unseen instances. Semantic representations of the input data are assigned as graph signals based on the visual features. The semantic representations are aligned with visual representations of the input data using a regularization method applied directly in a spectral graph wavelets (SGW) domain. The semantic representations are then used to generate labels for the unseen instances. The unseen instances may represent unknown conditions for an autonomous vehicle.

Abstract: Examples include methods, systems, and articles for localizing a vehicle relative to an imaged surface configuration. Localizing the vehicle may include selecting pairs of features in an image acquired from a sensor supported by the vehicle having corresponding identified pairs of features in a reference representation of the surface configuration. A three-dimensional geoarc may be generated based on an angle of view of the sensor and the selected feature pair in the reference representation. In some examples, a selected portion of the geoarc disposed a known distance of the vehicle away from the portion of the physical surface configuration may be determined. Locations where the selected portions of geoarcs for selected feature pairs overlap may be identified. In some examples, the reference representation may be defined in a three-dimensional space of volume elements (voxels), and voxels that are included in the highest number of geoarcs may be determined.

Abstract: A method and apparatus for recognizing target objects. The method comprises identifying a group of target objects in an image. Further, the method comprises forming a group of chips encompassing the group of target objects identified in the image. Yet further, the method comprises recognizing the group of target objects in the group of chips using a group of filters, wherein the group of filters was created using a group of models for reference objects and environmental information for a location where the group of target objects was located when the image was generated, wherein a filter in the group of filters comprises a group of reference images for a reference object in the reference objects.

Abstract: A method includes receiving image data at a first tracking system. The image data may represent a region in an image of a sequence of images. The method includes generating a first tracking fingerprint based on the image data. The method includes comparing the first tracking fingerprint and a second tracking fingerprint. The method further includes providing an output from the first tracking system to a second tracking system based on a result of the comparison of the first tracking fingerprint and the second tracking fingerprint.

Abstract: Described herein is a method for enhancing image data that includes dividing an image into multiple regions. The method includes measuring variations in pixel intensity distribution of the image to determine high pixel intensity variations for identifying an intensity-changing region. The method includes calculating a histogram of intensity distribution of pixel intensity values for the intensity-changing region without calculating a histogram of intensity distribution of pixel intensity values for each region of the multiple regions. The method also includes determining a transformation function based on the intensity distribution for the intensity-changing region. The method includes applying the transformation function to modify an intensity for each pixel in the image to produce an enhanced image in real time. The method also includes detecting in the enhanced image a horizon for providing to an operator of a vehicle an indication of the horizon in the image on a display in the vehicle.

Abstract: Examples include methods, systems, and articles for localizing a vehicle relative to an imaged surface configuration. Localizing the vehicle may include selecting pairs of features in an image acquired from a sensor supported by the vehicle having corresponding identified pairs of features in a reference representation of the surface configuration. A three-dimensional geoarc may be generated based on an angle of view of the sensor and the selected feature pair in the reference representation. In some examples, a selected portion of the geoarc disposed a known distance of the vehicle away from the portion of the physical surface configuration may be determined. Locations where the selected portions of geoarcs for selected feature pairs overlap may be identified. In some examples, the reference representation may be defined in a three-dimensional space of volume elements (voxels), and voxels that are included in the highest number of geoarcs may be determined.

Abstract: Examples include methods, systems, and articles for localizing a vehicle relative to an imaged surface configuration. Localizing the vehicle may include selecting pairs of features in an image acquired from a sensor supported by the vehicle having corresponding identified pairs of features in a reference representation of the surface configuration. A three-dimensional geoarc may be generated based on an angle of view of the sensor and the selected feature pair in the reference representation. In some examples, a selected portion of the geoarc disposed a known distance of the vehicle away from the portion of the physical surface configuration may be determined. Locations where the selected portions of geoarcs for selected feature pairs overlap may be identified. In some examples, the reference representation may be defined in a three-dimensional space of volume elements (voxels), and voxels that are included in the highest number of geoarcs may be determined.

Abstract: A method for automatic target recognition in synthetic aperture radar (SAR) data, comprising: capturing a real SAR image of a potential target at a real aspect angle and a real grazing angle; generating a synthetic SAR image of the potential target by inputting, from a potential target database, at least one three-dimensional potential target model at the real aspect angle and the real grazing angle into a SAR regression renderer; and, classifying the potential target with a target label by comparing at least a portion of the synthetic SAR image with a corresponding portion of the real SAR image using a processor.

Abstract: An apparatus, system, and method for increasing points in a point cloud. In one illustrative embodiment, a two-dimensional image of a scene and the point cloud of the scene are received. At least a portion of the points in the point cloud are mapped to the two-dimensional image to form transformed points. A fused data array is created using the two-dimensional image and the transformed points. New points for the point cloud are identified using the fused data array. The new points are added to the point cloud to form a new point cloud.

Abstract: Described is a system for dynamic background estimation which utilizes Particle Swarm Optimization (PSO). The present invention comprises a system, method, and computer program product for accurate estimation of a background mask corresponding to a dynamically changing scene. The system is configured to construct a background template model of a scene, and then capture an image of a current view of the scene with a camera. Thereafter, the system generates an image-based template matching cost function as an optimization problem, where the objective is to identify and fit a corresponding subregion of the background template model to the current camera view. The cost function is optimized using a PSO search algorithm. Finally, the system is configured to generate the corresponding subregion of the background template model for display. The inherent efficiency of PSO makes this system conducive for use in applications requiring real-time background estimation.

Abstract: A method includes receiving first tracking data and a first confidence value at a tracker selection system from a first tracking system. The method includes receiving second tracking data and a second confidence value at the tracker selection system from a second tracking system. The tracking systems may be configured to track an object in a sequence of images, and the first and second tracking data may indicate locations of regions where the corresponding tracking system has tracked the object in an image of the sequence of images. The confidence values may indicate likelihoods that a corresponding tracking system is tracking the object. The method further includes providing output data to the first tracking system and to the second tracking system. The output data may include data selected from the first tracking data and the second tracking data based on a result of a comparison of the confidence values.