Abstract: A method includes obtaining training data having first image pairs, where each of the first image pairs includes (i) a first training image and (ii) a first ground truth image. The method also includes training a machine learning model to generate realistic images using the first image pairs. The method further includes obtaining additional training data having second image pairs, where each of the second image pairs includes (i) a second training image and (ii) a second ground truth image. At least some of the images in the second image pairs are less aligned or of lower quality than at least some of the images in the first image pairs. In addition, the method includes continuing to train the machine learning model to generate the realistic images using the second image pairs.

Abstract: Devices, systems, and methods for three-dimensional (3D) evaluation point (3DEP) identification; wherein a method can include receiving a first conjugate point of a first real two-dimensional (2D) image, receiving a second conjugate point of a second real 2D image, the first and second conjugate points corresponding to a same geographical location, determining a first set of points of a 3D point set that project to within a specified distance of the first conjugate point in the first real 2D image, determining a second set of points of the 3D point set that project to within the specified distance of the second conjugate point in the second real 2D image, identifying a common point in both the first set of points and the second set of points that satisfies a specified heuristic relative to all other points in both the first set of points and the second set of points, and using the point as the 3DEP.

Abstract: Devices, systems, and methods for three-dimensional (3D) evaluation point (3DEP) identification; wherein a method can include receiving a first conjugate point of a first real two-dimensional (2D) image, receiving a second conjugate point of a second real 2D image, the first and second conjugate points corresponding to a same geographical location, determining a first set of points of a 3D point set that project to within a specified distance of the first conjugate point in the first real 2D image, determining a second set of points of the 3D point set that project to within the specified distance of the second conjugate point in the second real 2D image, identifying a common point in both the first set of points and the second set of points that satisfies a specified heuristic relative to all other points in both the first set of points and the second set of points, and using the point as the 3DEP.

Abstract: Devices, systems, and methods for image processing; wherein a method can include generating a synthetic image based on the 3D point set of a same geographical region as the 2D real image and performing a coarse registration to grossly register the synthetic image to the 2D real image, the coarse registration including determining, for a plurality of synthetic image tiles that span the synthetic and at each of a plurality of offsets in a search region, a correlation score resulting in a plurality of correlation score arrays, determining a weighted combination of scores in the correlation score arrays resulting in a combined correlation score array, identifying an offset of the plurality of offsets based on the combined correlation score array, and moving the synthetic image relative to the 2D real image by the identified offset.

Abstract: A method includes obtaining training data having first image pairs, where each of the first image pairs includes (i) a first training image and (ii) a first ground truth image. The method also includes training a machine learning model to generate realistic images using the first image pairs. The method further includes obtaining additional training data having second image pairs, where each of the second image pairs includes (i) a second training image and (ii) a second ground truth image. At least some of the images in the second image pairs are less aligned or of lower quality than at least some of the images in the first image pairs. In addition, the method includes continuing to train the machine learning model to generate the realistic images using the second image pairs.

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 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: Discussed herein are devices, systems, and methods for multi-image ground control point (GCP) determination. A method can include extracting, from a first image including image data of a first geographical region, a first image template, the first image template including a contiguous subset of pixels from the first image and a first pixel of the first image indicated by the GCP, predicting a first pixel location of the GCP in a second image, the second image including image data of a second geographical overlapping with the first geographical region, extracting, from the second image, a second image template, the second image template including a contiguous subset of pixels from the second image and a second pixel corresponding to the pixel location, identifying a second pixel of the second image corresponding to a highest correlation score, and adding a second pixel location of the identified pixel to the GCP.

Abstract: Discussed herein are devices, systems, and methods for building mask generation. A method can include setting a respective pixel value of an image to a first specified value if the respective pixel corresponds, according to a three-dimensional (3D) point set, to an elevation greater than a specified Z threshold, otherwise setting the respective pixel value to a second, different specified value, grouping contiguous pixels set to the first specified value into one or more groups, determining a feature of each of the one or more groups, comparing the determined feature to a threshold and retaining the group if the feature is greater than a threshold, otherwise removing the group, and providing a building mask that includes pixels of the retained group set to a value and other pixels set to a different value.

Abstract: Discussed herein are devices, systems, and methods for image processing. A method can include generating a synthetic image based on a two-dimensional (2D) image of the geographical region, performing a coarse registration to grossly, register the synthetic image to the 2D image, and performing a fine registration following the coarse registration to improve the registration between the synthetic image and the 2D image.

Abstract: Discussed herein are devices, systems, and methods for multi-image ground control point (GCP) determination. A method can include extracting, from a first image including image data of a first geographical region, a first image template, the first image template including a contiguous subset of pixels from the first image and a first pixel of the first image indicated by the GCP, predicting a first pixel location of the GCP in a second image, the second image including image data of a second geographical overlapping with the first geographical region, extracting, from the second image, a second image template, the second image template including a contiguous subset of pixels from the second image and a second pixel corresponding to the pixel location, identifying a second pixel of the second image corresponding to a highest correlation score, and adding a second pixel location of the identified pixel to the GCP.

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: Discussed herein are devices, systems, and methods for synthetic image generation. A method can include projecting a three-dimensional (3D) point set of a first geographical region to an image space of an image of a second geographical region to generate synthetic image data, identifying control points (CPs) between the image and the synthetic image data, adjusting a geometry of the synthetic image data based on the identified CPs, determining metadata for the synthetic image based on metadata of the image, and associating the determined metadata with the synthetic image data to generate the synthetic image.

Abstract: Discussed herein are devices, systems, and methods for building mask generation. A method can include setting a respective pixel value of an image to a first specified value if the respective pixel corresponds, according to a three-dimensional (3D) point set, to an elevation greater than a specified Z threshold, otherwise setting the respective pixel value to a second, different specified value, grouping contiguous pixels set to the first specified value into one or more groups, determining a feature of each of the one or more groups, comparing the determined feature to a threshold and retaining the group if the feature is greater than a threshold, otherwise removing the group, and providing a building mask that includes pixels of the retained group set to a value and other pixels set to a different value.

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 method can include registering a first image of a region to a three-dimensional (3D) point set of the region to generate a registered first image, registering a second image of the region to the 3D point set to generate a registered second image, identifying, based on the 3D point set, geometric tie points of the registered first image and the registered second image, projecting, using an affine transformation determined based on the identified geometric tie points, pixels of the registered first image to an image space of the registered second image to generate a registered and transformed first image, and displaying the registered and transformed first image and the registered second image simultaneously.

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 method can include receiving, first and second digital elevation models (DEMs) including first elevation data of a geographic location and second elevation data of the geographic location, respectively, identifying differences between the first elevation data and the second elevation data that are greater than a first threshold, determining, for a point in the second elevation data identified to correspond to a difference greater than the first threshold, a slope of the geographic location around and including the point, and altering, in response to determining the difference is greater than a second threshold determined based on the determined slope, elevation data of the second DEM corresponding to the point.

Abstract: A method can include receiving, first and second digital elevation models (DEMs) including first elevation data of a geographic location and second elevation data of the geographic location, respectively, identifying differences between the first elevation data and the second elevation data that are greater than a first threshold, determining, for a point in the second elevation data identified to correspond to a difference greater than the first threshold, a slope of the geographic location around and including the point, and altering, in response to determining the difference is greater than a second threshold determined based on the determined slope, elevation data of the second DEM corresponding to the point.