Uncertainty-refined image segmentation under domain shift

Patent · Tue Dec 17 04:00:00 EST 2024

OSTI ID:2542924

Martinez, Carianne; Potter, Kevin Matthew; Donahue, Emily; Smith, Matthew David; Snider, Charles J.; Korbin, John P.; Roberts, Scott Alan; Collins, Lincoln

Digital image segmentation is provided. The method comprises 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. The neural network receives image data from a second, different domain. A vector of N values that sum to 1 is calculated for each image element, wherein each value represents an image segmentation class. A label is assigned to each image element according to the class with the highest value in the vector. Multiple inferences are performed with active dropout layers for each image element, and an uncertainty value is generated for each image element. Uncertainty is resolved according to expected characteristics. The label of any image element with an uncertainty above a threshold is replaced with a new label corresponding to a segmentation class based on domain knowledge.

• Research Organization: Sandia National Laboratories (SNL-NM), Albuquerque, NM (United States)
• Sponsoring Organization: USDOE National Nuclear Security Administration (NNSA)
• DOE Contract Number: NA0003525
• Assignee: National Technology & Engineering Solutions of Sandia, LLC (Albuquerque, NM)
• Patent Number(s): 12,169,962
• Application Number: 17/832,477
• OSTI ID: 2542924
• Country of Publication: United States
• Language: English