Abstract: A system and method that reduces the time needed to identify infrastructure that has been damaged due to a storm, earthquake, or other event. At a high level, the presently claimed invention includes the following steps. Step 1: Assigned airborne response equipped with high-powered lidar sensors to fly over impacted areas to collect a 3D point cloud. This data focuses on the 3D geometry of the built environment and may be processed in a highly automated fashion to derive the locations of downed poles and wires. Step 2: run automated processes to identify highly impacted areas—providing an output of precise XY locations of downed poles and wires. And step 3: develop unique resource allocation response given the areas of known major damage.

Abstract: An image caption generator analyzes a set of images of a region of an environment captured during or after environmental emergency event to provide a set of utility images selected from the set of images that include a utility asset. The image caption generator also generates descriptive tags for the set of utility images. The descriptive tags each characterize a type and a state of a respective utility asset included in a respective image of the set of utility images. A verification engine cross-references the set of utility images with information from an external data source to confirm an accuracy of the descriptive tags for the set of utility images and providing feedback to the image caption generator based on results of the cross-referencing. The image caption generator adjusts the parameters of the ML model based on the feedback to increase accuracy of the descriptive tags.

Abstract: Systems and methods to generate synthetic images for use in a machine learning training set. The process begins with accessing a database of real-world 3-D images of equipment in a power grid, the 3-D images of equipment include 3-D measurements to create a dimensionally accurate and photorealistic model of the equipment. Optionally, the 3-D images could be aged or weathered using imaging editing software. Next, a database of real-world photographs of scenes in which the equipment is installed is accessed. Optionally, the identical scenes can be captured at different times of day, different times of the year, and at different perspectives. Next, using image editing software, the 3-D images of equipment is inserted into at least one of the scenes to form a synthetic image based on a combination of the equipment and the scene in which each of the equipment and the scene were previously captured independently of each other.

Abstract: A system and method that reduces the time needed to identify infrastructure that has been damaged due to a storm, earthquake, or other event. At a high level, the presently claimed invention includes the following steps. Step 1: Assigned airborne response equipped with high-powered lidar sensors to fly over impacted areas to collect a 3D point cloud. This data focuses on the 3D geometry of the built environment and may be processed in a highly automated fashion to derive the locations of downed poles and wires. Step 2: run automated processes to identify highly impacted areas—providing an output of precise XY locations of downed poles and wires. And step 3: develop unique resource allocation response given the areas of known major damage.

Abstract: A condition generator analyzes a set of utility asset images of a particular utility asset using the ML (machine learning) model identify a type and condition of the particular utility asset depicted in the set of utility asset images. The identified condition is assigned a confidence score, and the set of utility asset images includes at least two images of the particular utility asset captured at different angles. The condition generator generates a descriptive tag for the set of utility asset images based on the identified type and condition. The descriptive tag characterizes an operational status of the particular utility asset. The condition generator stores the set of utility asset images and the generated descriptive tag in a utility asset database. The utility asset database stores images of utility assets.

Abstract: An image caption generator analyzes a set of images of a region of an environment captured during or after environmental emergency event to provide a set of utility images selected from the set of images that include a utility asset. The image caption generator also generates descriptive tags for the set of utility images. The descriptive tags each characterize a type and a state of a respective utility asset included in a respective image of the set of utility images. A verification engine cross-references the set of utility images with information from an external data source to confirm an accuracy of the descriptive tags for the set of utility images and providing feedback to the image caption generator based on results of the cross-referencing. The image caption generator adjusts the parameters of the ML model based on the feedback to increase accuracy of the descriptive tags.

Abstract: Systems and methods for defining bounding polygons in a view of a three-dimensional scene. Rays are defined that each extend from a viewpoint of a virtual three-dimensional model to a vertex of an object of interest in the virtual three-dimensional model. A set of occluded rays is determined that include rays intercepting occluding objects in the virtual three-dimensional model prior to reaching a vertex of the object of interest when extending from the viewpoint. A set of visible rays is defined with respect to the object of interest that excludes the occluded set of rays. A bounding polygon for the object of interest that encompasses each vertex intercepted by the set of visible rays and excludes at least one vertex intercepted by a respective ray in the set of occluded rays is defined in an image of the virtual three-dimensional model that is created from the viewpoint.

Abstract: Systems and methods to generate synthetic images for use in a machine learning training set. The process begins with accessing a database of real-world 3-D images of equipment in a power grid, the 3-D images of equipment include 3-D measurements to create a dimensionally accurate and photorealistic model of the equipment. Optionally, the 3-D images could be aged or weathered using imaging editing software. Next, a database of real-world photographs of scenes in which the equipment is installed is accessed. Optionally, the identical scenes can be captured at different times of day, different times of the year, and at different perspectives. Next, using image editing software, the 3-D images of equipment is inserted into at least one of the scenes to form a synthetic image based on a combination of the equipment and the scene in which each of the equipment and the scene were previously captured independently of each other.