US 12,169,962 B2
Uncertainty-refined image segmentation under domain shift
Carianne Martinez, Albuquerque, NM (US); Kevin Matthew Potter, Albuquerque, NM (US); Emily Donahue, Albuquerque, NM (US); Matthew David Smith, Albuquerque, NM (US); Charles J. Snider, Albuquerque, NM (US); John P. Korbin, Albuquerque, NM (US); Scott Alan Roberts, Albuquerque, NM (US); and Lincoln Collins, Albuquerque, NM (US)
Assigned to National Technology & Engineering Solutions of Sandia, LLC, Albuquerque, NM (US)
Filed by National Technology & Engineering Solutions of Sandia, LLC, Albuquerque, NM (US)
Filed on Jun. 3, 2022, as Appl. No. 17/832,477.
Application 17/832,477 is a continuation in part of application No. 16/887,311, filed on May 29, 2020, granted, now 11,379,991.
Prior Publication US 2022/0301291 A1, Sep. 22, 2022
Int. Cl. G06V 10/776 (2022.01); G06T 7/11 (2017.01); G06V 10/764 (2022.01); G06V 10/82 (2022.01)
CPC G06V 10/776 (2022.01) [G06T 7/11 (2017.01); G06V 10/764 (2022.01); G06V 10/82 (2022.01); G06T 2207/20081 (2013.01); G06T 2207/20084 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A computer-implemented method for digital image segmentation, the method comprising:
using a number of processors to perform the steps of:
training a neural network for image segmentation with a labeled training dataset from a first domain, wherein a subset of nodes in the neural net are dropped out during training;
receiving, by the neural network, image data from a second, different domain, wherein the image data comprises a number of image elements;
calculating, by the neural network, a vector of N values that sum to 1 for each image element, wherein each of the N values represents an image segmentation class;
assigning, by the neural network, a segmentation label to each image element, wherein the segmentation label corresponds to a segmentation class with a highest value in the vector calculated for the image element;
performing, by the neural network with active dropout layers, multiple inferences for each image element;
generating, by the neural network, an uncertainty value for each image element according to the inferences;
resolving uncertainty according to expected characteristics based on domain knowledge; and
replacing the segmentation label of any image element with an uncertainty value above a predefined threshold with a new segmentation label corresponding to a segmentation class according to the domain knowledge for that image element.