Abstract: Systems and methods for VIIRS image processing. The method can include receiving image data of immediately adjacent VIIRS image scans including a first image scan and a second image scan. The first image scan and the second image scan provide a partially overlapping view of a geographic area. The method can further involve resampling columns of pixels of the first image scan and the second image scan. The resampling can include selecting, in the first image scan and the second image scan, a subset of pixel values in each column that correspond to a specified geographic distance. The method can further involve upsampling the selected pixels to an equal number of pixels in each column resulting in upsampled pixel values and interpolating the upsampled pixel values to produce modified first and second image scans.

Abstract: Subject matter regards generating a 3D point cloud and registering the 3D point cloud to the surface of the Earth (sometimes called “geo-locating”). A method can include capturing, by unmanned vehicles (UVs), image data representative of respective overlapping subsections of the object, registering the overlapping subsections to each other, and geo-locating the registered overlapping subsections.

Abstract: Subject matter regards colorizing a three-dimensional (3D) point set. A method of colorizing a 3D point can include voxelizing 3D points including the 3D point into voxels such that a voxel of the voxels including the 3D point includes a voxel subset of the 3D points, projecting the voxel subset to respective image spaces of first and second images used to generate the 3D points, and associating a color value, determined based on a respective number of pixels of the first and second images to which the voxel subset projects, with the 3D point.

Abstract: Systems and methods for VIIRS image processing. The method can include receiving image data of immediately adjacent VIIRS image scans including a first image scan and a second image scan. The first image scan and the second image scan provide a partially overlapping view of a geographic area. The method can further involve resampling columns of pixels of the first image scan and the second image scan. The resampling can include selecting, in the first image scan and the second image scan, a subset of pixel values in each column that correspond to a specified geographic distance. The method can further involve upsampling the selected pixels to an equal number of pixels in each column resulting in upsampled pixel values and interpolating the upsampled pixel values to produce modified first and second image scans.

Abstract: A method can include identifying a geolocation of an object in an image, the method comprising receiving data indicating a pixel coordinate of the image selected by a user, identifying a data point in a targetable three-dimensional (3D) data set corresponding to the selected pixel coordinate, and providing a 3D location of the identified data point.

Abstract: A machine receives a large image having large image dimensions that exceed memory threshold dimensions. The large image includes metadata. The machine adjusts an orientation and a scaling of the large image based on the metadata. The machine divides the large image into a plurality of image tiles, each image tile having tile dimensions smaller than or equal to the memory threshold dimensions. The machine provides the plurality of image tiles to an artificial neural network. The machine identifies, using the artificial neural network, at least a portion of the target in at least one image tile. The machine identifies the target in the large image based on at least the portion of the target being identified in at least one image tile.

Abstract: Subject matter regards improving image segmentation or image annotation. A method can include receiving, through a user interface (UI), for each class label of class labels to be identified by the ML model and for a proper subset of pixels of the image data, data indicating respective pixels associated with the class label, partially training the ML model based on the received data, generating, using the partially trained ML model, pseudo-labels for each pixel of the image data for which a class label has not been received, and receiving, through the UT, a further class label that corrects a pseudo-label of the generated pseudo-labels.

Abstract: Subject matter regards generating a 3D point cloud and registering the 3D point cloud to the surface of the Earth (sometimes called “geo-locating”). A method can include capturing, by unmanned vehicles (UVs), image data representative of respective overlapping subsections of the object, registering the overlapping subsections to each other, and geo-locating the registered overlapping subsections.

Abstract: A system receives digital images of a geographic location, associates each digital image with ground control points in a set of reference stereo images, and associates each digital image to each other digital image via image to image tiepoints. The system updates a geometry of each image via a bundle adjustment, and uses a prioritized stacking order to establish piecewise linear seam lines between each of the images. The system finally builds a prioritized map in a mosaic space specifying the source image pixels that are used in each region of the output mosaic, and forms the mosaic image using the prioritized map.

Abstract: Subject matter regards colorizing a three-dimensional (3D) point set. A method of colorizing a 3D point can include voxelizing 3D points including the 3D point into voxels such that a voxel of the voxels including the 3D point includes a voxel subset of the 3D points, projecting the voxel subset to respective image spaces of first and second images used to generate the 3D points, and associating a color value, determined based on a respective number of pixels of the first and second images to which the voxel subset projects, with the 3D point.

Abstract: A system receives digital images of a geographic location, associates each digital image with ground control points in a set of reference stereo images, and associates each digital image to each other digital image via image to image tiepoints. The system updates a geometry of each image via a bundle adjustment, and uses a prioritized stacking order to establish piecewise linear seam lines between each of the images. The system finally builds a prioritized map in a mosaic space specifying the source image pixels that are used in each region of the output mosaic, and forms the mosaic image using the prioritized map.

Abstract: A machine receives a large image having large image dimensions that exceed memory threshold dimensions. The large image includes metadata. The machine adjusts an orientation and a scaling of the large image based on the metadata. The machine divides the large image into a plurality of image tiles, each image tile having tile dimensions smaller than or equal to the memory threshold dimensions. The machine provides the plurality of image tiles to an artificial neural network. The machine identifies, using the artificial neural network, at least a portion of the target in at least one image tile. The machine identifies the target in the large image based on at least the portion of the target being identified in at least one image tile.

Abstract: A method can include identifying a geolocation of an object in an image, the method comprising receiving data indicating a pixel coordinate of the image selected by a user, identifying a data point in a targetable three-dimensional (3D) data set corresponding to the selected pixel coordinate, and providing a 3D location of the identified data point.

Abstract: An image processing technique uses requirements for a geospatial distribution of image tie points for a triangulation of images. The images are correlated, thereby generating candidate tie points across the images. Statistical consistency checks are applied to the images to identify and dispose of the candidate tie points that are local outliers, and a geometric identification technique is applied to the images to identify and dispose of the candidate tie points that are global outliers. The candidate tie points that are not local outliers or global outliers are spatially down-selected such that the spatially down-selected candidate tie points satisfy the one or more requirements for the geospatial distribution.

Abstract: An image processing technique uses requirements for a geospatial distribution of image tie points for a triangulation of images. The images are correlated, thereby generating candidate tie points across the images. Statistical consistency checks are applied to the images to identify and dispose of the candidate tie points that are local outliers, and a geometric identification technique is applied to the images to identify and dispose of the candidate tie points that are global outliers. The candidate tie points that are not local outliers or global outliers are spatially down-selected such that the spatially down-selected candidate tie points satisfy the one or more requirements for the geospatial distribution.

Abstract: A computer vision method, executed by one or more processors, for generating a single 3D model view of a geographic scene includes: receiving image data for the scene from a plurality of sensors located at different angles with respect to the geographic scene; dividing the image data into a plurality of image spatial regions; correlating the image data in each image spatial region to obtain a score for each image data in each image spatial region; grouping the image data in each image spatial region into two or more image clusters, based on the scores for each image; performing a multi-ray intersection within each image cluster to obtain a 3D reference point for each region; for each region, combining the one or more clusters, based on the 3D reference point for the region; and registering the combined clusters for each region to obtain a single 3D model view of the scene.

Abstract: A computer vision method, executed by one or more processors, for generating a single 3D model view of a geographic scene includes: receiving image data for the scene from a plurality of sensors located at different angles with respect to the geographic scene; dividing the image data into a plurality of image spatial regions; correlating the image data in each image spatial region to obtain a score for each image data in each image spatial region; grouping the image data in each image spatial region into two or more image clusters, based on the scores for each image; performing a multi-ray intersection within each image cluster to obtain a 3D reference point for each region; for each region, combining the one or more clusters, based on the 3D reference point for the region; and registering the combined clusters for each region to obtain a single 3D model view of the scene.

Abstract: Subject matter regards improving image segmentation or image annotation. A method can include receiving, through a user interface (UI), for each class label of class labels to be identified by the ML model and for a proper subset of pixels of the image data, data indicating respective pixels associated with the class label, partially training the ML model based on the received data, generating, using the partially trained ML model, pseudo-labels for each pixel of the image data for which a class label has not been received, and receiving, through the UT, a further class label that corrects a pseudo-label of the generated pseudo-labels.

Abstract: A system and method of generating point clouds from passive images. Image clusters are formed, wherein each image cluster includes two or more passive images selected from a set of passive images. Quality of the point cloud that could be generated from each image cluster is predicted for each image cluster based on a performance prediction score for each image cluster. A subset of image clusters is selected for further processing based on their performance prediction scores. A mission-specific quality score is determined for each point cloud generated and the point cloud with the highest quality score is selected for storage.

Abstract: The system and methods described herein operate on a plurality of images that include multiple views of the same scene, typically from slightly different viewing angles and/or lighting conditions. One of the images is selected as a reference image. For each image ray in a non-reference image, the system and methods resample a local region from the non-reference image's space to the reference image's space. The resampling is performed multiple times, each time with a different surface orientation hypothesis. The system and methods run cross-correlation style correlators on the resampled images, evaluate correlation scores for each of the resampled images, and select the surface orientation hypothesis associated with the highest correlation score. The system and methods project a peak of the correlation surface back through a geometry model for the selected surface orientation hypothesis to determine a three-dimensional (ground) location for the image ray.