Raytheon Company

Franklin R Tanner

Philip A Sallee

Jonathan Goldstein

James Mullen

Jeffrey S Klein

Darrell Young

Stephen J Raif

James Talamonti

Steven A Israel

Shane A Zabel

Lisa A Mccoy

Christopher A Eccles

Grayson D Gunderson

Discussed herein are devices, systems, and methods for autonomous, dynamic navigation of scenarios. A method can include implementing a path generation machine learning (ML) technique to determine paths between a device and a goal, determining a node of the paths as an intersection of at least two of the paths, and implementing an executive ML technique to determine which of the at least two paths to take at a node of the graph to reach the goal.

Graph construction and execution ML techniques

Systems, devices, methods, and computer-readable media for evaluation and visualization of machine learning data drift. A method can include receiving a series of data indicating accuracy and confidence associated with classification of respective batches of input samples, and dynamically displaying, on the GUI, a concurrent plot of the accuracy and confidence as the series of data are received.

Architecture for dynamic ML model drift evaluation and visualization on a GUI

Subject matter regards improving machine learning techniques using informed pseudolabels. A method can include receiving previously assigned labels indicating an expected classification for data, the labels having a specified uncertainty, generating respective pseudolabels for the data based on the previously assigned labels, the data, a class vector determined by an ML model, and a noise model indicating, based on the specified uncertainty, a likelihood of the previously assigned label given the class, and substituting the pseudolabels for the previously assigned labels in a next epoch of training the ML model.

Machine learning using informed pseudolabels

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.

Target identification in large image data

A machine receives, for a given target in a given image, a plurality of annotations, each annotation indicating a range of pixels corresponding to the given target. The machine computes a number of pixels in an intersection of the plurality of annotations. The machine computes an average pixels on target (APOT) for the given target by computing, for each annotation, a number of pixels that corresponds to the given target and taking a mean of those numbers. The machine computes a precision value of the plurality of annotations as the number of pixels in the intersection divided by the APOT. The machine provides an output representing the precision value.

Calculating the precision of image annotations

A neural network apparatus includes processing circuitry and memory. The memory stores a plurality of images of a target. The processing circuitry is configured to: access, from the memory, a first image and an identification of a centroid pixel of the target within the first image; generate, based on a geometry of the target and the centroid pixel, a confidence map indicating, for each pixel in the first image, a confidence value that the pixel includes the target; train, using the plurality of images of the target, including the first image and the confidence map, an artificial neural network to identify the target in visual data; and provide an output representing the trained artificial neural network.

Computer architecture for object detection using point-wise labels

A modeling engine (ME) for generating or 'bootstrapping' a three dimensional edge model (3DEM) from two stereo pairs of images and correcting engine (CE) for correcting a camera model associated with an image are provided. The ME back projects edge detected images into 3DEMs using camera models associated with the stereo images, updates and stereo updates voxel probabilities in the 3DEM s with back projections of the edge detected images, and then merges the 3DEMs together to create a 'sparse' 3DEM. The CE calculates a registration solution mapping an edge detected image of an image to an expected edge image. The expected edge image is a projection of a 3DEM using a camera model associated with the image. The CE corrects the camera model by applying the registration solution to the camera model based on determining whether the registration solution is a high confidence registration solution.

System for correcting RPC camera model pointing errors using 2 sets of stereo image pairs and probabilistic 3-dimensional models

A modeling engine (ME) for generating or 'bootstrapping' a three dimensional edge model (3DEM) from two stereo pairs of images and correcting engine (CE) for correcting a camera model associated with an image are provided. The ME back projects edge detected images into 3DEMs using camera models associated with the stereo images, updates and stereo updates voxel probabilities in the 3DEMs with back projections of the edge detected images, and then merges the 3DEMs together to create a 'sparse' 3DEM. The CE calculates a registration solution mapping an edge detected image of an image to an expected edge image. The expected edge image is a projection of a 3DEM using a camera model associated with the image. The CE corrects the camera model by applying the registration solution to the camera model based on determining whether the registration solution is a high confidence registration solution.

Growing community of inventors

Asburn, VA, United States of America

Franklin R Tanner