Abstract: Feature extraction of image data using feature extraction modules. The feature extraction modules may be provided in an architecture that allows for modular, decoupled generation and/or operation of the feature extraction modules to generate feature data corresponding to image data. In this regard, the feature extraction modules may communicate with a file system storing image data and feature data by way of a common interface format. Accordingly, regardless of the nature of the execution of the feature extraction module, each feature extraction module may be communicative by way of the common interface format, thereby providing a modular approach that is highly scalable, flexible, and adaptive.

Abstract: A crowdsourced search and locate platform, comprising an application server and a client interface application. The application server: receives connections from crowdsourcing participants; navigates a first crowdsourcing participant to a specific geospatial location; sends an image corresponding to the geospatial location to the first crowdsourcing participant; receives tagging data from the first crowdsourcing participant, the tagging data corresponding to a plurality of objects and locations identified by the first crowdsourcing participant.

Abstract: A system for modular image mining and search, comprising a plurality of image capture devices, a search engine, an indexing engine, a database, and user interface software. A plurality of image capture devices capture images and store them to a database. The indexing engine indexes images to create image data for each image with associated metadata and stores the information to the database. The interface software is used to request search queries via a search engine. The search engine accesses the image data to find images or parts of images that satisfy the search query.

Abstract: Techniques for object detection include training a first classifier to detect an object based on textural features and a second classifier to detect the object based on textural features and spectral features. A classifier relationship between the two classifiers is learned and used with the first classifier to detect other objects. If desired, the performance of the object detector can be improved by comparing the results of detecting objects with the first classifier and the classifier relationship versus detecting objects with the first and second classifier together, and modifying the classifier relationship based upon the comparison.

Abstract: Techniques for atmospheric compensation in satellite imagery that include converting an image including an array of radiance values to an array of surface reflectance values. The conversion is performed in an automated fashion by identifying one or more portions of the image for which the surface reflectance can be estimated and determining the Aerosol Optical Depth (AOD) by iteratively comparing the radiance value captured by the image sensor to a calculated radiance value (based on the known surface reflectance, historical values for other atmospheric parameters, and the AOD) and adjusting the AOD until the calculated radiance value is substantially the same as the captured radiance value.

Abstract: Generation of a plurality of orthomosaic image layers for a geographical area, wherein the layers comprise different spatial resolutions and are radiometrically normalized to facilitate improved radiometric consistency when zooming relative to the geographic area. In applications that facilitate zooming with respect to a geographic area imaged by the orthomosaic layers, radiometric normalization of the orthomosaic image layers may reduce radiometric discontinuities when zooming in, zooming out, and/or panning a displayed portion that is output to the user. The orthomosaic layers may be generated based on one or more orthomosaic image layer or may be developed independently using a source images. In any regard, the radiometric normalization may include generating a normalization function based on image metadata that is calculated and may be independent of the spatial resolution of the image.

Abstract: Techniques for automatically removing slivers from orthomosaics. First, slivers may be identified, which may be based upon user-configurable characteristics such as width and length. Second, slivers may be replaced with portions of another image, such as an older image. Third, boundary slivers may optionally be removed. Fourth, remaining boundary or interior slivers can be inflated by expanding the sliver until it is no longer a sliver, which may include replacing a portion of an adjacent image portion with older imagery.

Abstract: Techniques for improving overhead image bathymetry include obtaining depth information from image data based on one or more of the spectral domain, the angular domain (e.g., stereo or photogrammetry), the temporal domain (e.g., monitoring the movement of waves in a body of water), or any other suitable domain, together with a priori information about the area of interest. These different pieces of depth information from the various different domains are combined together using any combination of Optimal Estimation and Continuity Constraints to improve the accuracy of the results.

Abstract: Processing and analyzing at least one remotely-sensed image to automatically detect and identify parking lots within a region of the remotely-sensed image. This may include: (1) receiving a remotely-sensed image having zero or more parking lots; (2) identifying a set of pixels related to parking lot features within the remotely-sensed image; (3) identifying at least one parking row within the remotely-sensed image; and (3) identifying a parking lot based on the identified parking rows. Identifying at least one parking lot may further include using first, second, third, and fourth level modules, such as a concrete/asphalt detection module, a road marking detection module, a vehicle detection module, and a Hough aggregation for parking row detection module.

Abstract: A Metric Information Network (MIN) with a plurality of Ground Control Points (GCPs) that are selected in an automated fashion. The GCP selection includes clustering algorithms as compared to prior art pair-wise matching algorithms. Further, the image processing that takes place in identifying interest points, clustering, and selecting tie points to be GCPs is all performed before the MIN is updated. By arranging for the processing to happen in this manner, the processing that is embarrassingly parallel (identifying interest points, clustering, and selecting tie points) can be performed in a distributed fashion across many computers and then the MIN can be updated.

Abstract: Utilities (e.g., systems, methods, etc.) for automatically generating high resolution population density estimation data sets through manipulation of low resolution population density estimation data sets with high resolution overhead imagery data (e.g., such as overhead imagery data acquired by satellites, aircrafts, etc. of celestial bodies). Stated differently, the present utilities make use of high resolution overhead imagery data to determine how to distribute the population density of a large, low resolution cell (e.g., 1000 m) among a plurality of smaller, high resolution cells (e.g., 100 m) within the larger cell.

Abstract: A system for automatically extracting interesting structures or areas (e.g., built-up structures such as buildings, tents, etc.) from HR/VHR satellite imagery data using corresponding LR satellite imagery data. The system breaks down HR/VHR input satellite images into a plurality of components (e.g., groups of pixels), organizes the components into a first hierarchical data structure (e.g., a Max-Tree), generates a second hierarchical data structure (e.g., a KD-Tree) from feature elements (e.g., spectral and shape characteristics) of the components, uses LR satellite imagery data to categorize components as being of interest or not, uses the feature elements of the categorized components to train the second data structure to be able to classify all components of the first data structure as being of interest or not, classifies the components of the first data structure with the trained second data structure, and then maps components classified as being of interest into a resultant image.

Abstract: Automatic generation of a mosaic comprising a plurality of geospatial images. An embodiment of the automatic mosaic generation may include automated source image selection that includes comparison of source images to base layer image to determine radiometric similar source images. Additionally, an embodiment of an automatic cutline generator may be provided to automatically determine a cutline when merging two images such that radiometric differences between the images along the cutline are reduced. In this regard, less perceivable outlines may be provided. Further still, an embodiment of a radiometric normalization module may be provided that may determine radiometric adjustments to source images to match certain properties of the base layer image. In some embodiments, when processing source images, the source images may be downsampled during a portion of the processing to reduce computational overhead. Additionally, some highly parallel computations may be performed by a GPU to further enhance performance.

Abstract: A system for oil storage tank monitoring, comprising an extraction module and an analysis module, wherein an extraction module is utilized to determine information from an oil storage tank image and an analysis module is utilized to perform operations on extracted information, such as for determining measurements or values of oil storage tanks.

Abstract: Land classification based on analysis of image data. Feature extraction techniques may be used to generate a feature stack corresponding to the image data to be classified. A user may identify training data from the image data from which a classification model may be generated using one or more machine learning techniques applied to one or more features of the image. In this regard, the classification module may in turn be used to classify pixels from the image data other than the training data. Additionally, quantifiable metrics regarding the accuracy and/or precision of the models may be provided for model evaluation and/or comparison. Additionally, the generation of models may be performed in a distributed system such that model creation and/or application may be distributed in a multi-user environment for collaborative and/or iterative approaches.

Abstract: Techniques for automatically removing slivers from orthomosaics. First, slivers may be identified, which may be based upon user-configurable characteristics such as width and length. Second, slivers may be replaced with portions of another image, such as an older image. Third, boundary slivers may optionally be removed. Fourth, remaining boundary or interior slivers can be inflated by expanding the sliver until it is no longer a sliver, which may include replacing a portion of an adjacent image portion with older imagery.

Abstract: Feature extraction of image data using feature extraction modules. The feature extraction modules may be provided in an architecture that allows for modular, decoupled generation and/or operation of the feature extraction modules to generate feature data corresponding to image data. In this regard, the feature extraction modules may communicate with a file system storing image data and feature data by way of a common interface format. Accordingly, regardless of the nature of the execution of the feature extraction module, each feature extraction module may be communicative by way of the common interface format, thereby providing a modular approach that is highly scalable, flexible, and adaptive.

Abstract: Automatic generation of a mosaic comprising a plurality of geospatial images. An embodiment of the automatic mosaic generation may include automated source image selection that includes comparison of source images to base layer image to determine radiometric similar source images. Additionally, an embodiment of an automatic cutline generator may be provided to automatically determine a cutline when merging two images such that radiometric differences between the images along the cutline are reduced. In this regard, less perceivable cutlines may be provided. Further still, an embodiment of a radiometric normalization module may be provided that may determine radiometric adjustments to source images to match certain properties of the base layer image. In some embodiments, when processing source images, the source images may be downsampled during a portion of the processing to reduce computational overhead. Additionally, some highly parallel computations may be performed by a GPU to further enhance performance.

Abstract: Automatic generation of a mosaic comprising a plurality of geospatial images. An embodiment of the automatic mosaic generation may include automated source image selection that includes comparison of source images to base layer image to determine radiometric similar source images. Additionally, an embodiment of an automatic cutline generator may be provided to automatically determine a cutline when merging two images such that radiometric differences between the images along the cutline are reduced. In this regard, less perceivable outlines may be provided. Further still, an embodiment of a radiometric normalization module may be provided that may determine radiometric adjustments to source images to match certain properties of the base layer image. In some embodiments, when processing source images, the source images may be downsampled during a portion of the processing to reduce computational overhead. Additionally, some highly parallel computations may be performed by a GPU to further enhance performance.

Abstract: A crowdsourced search and locate platform, comprising an application server and a client interface application. The application server: receives connections from crowdsourcing participants; navigates a first crowdsourcing participant to a specific geospatial location; sends an image corresponding to the geospatial location to the first crowdsourcing participant; receives tagging data from the first crowdsourcing participant, the tagging data corresponding to a plurality of objects and locations identified by the first crowdsourcing participant. The client interface application: displays elements enabling an administrative user to configure a crowdsourced search and locate campaign; displays other elements enabling an administrative user to configure and run a crowdrank algorithm; and upon the user's selecting a particular object type icon, displays tags of the type selected by placing visual indicators at each tagged location of objects of the selected type within the plurality of image strips.