Skip to main content
U.S. Department of Energy
Office of Scientific and Technical Information

Uncertainty-refined image segmentation under domain shift

Patent ·
OSTI ID:1924873

A method for 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. The label of any image element with an uncertainty value above a predefined threshold is replaced with a new label corresponding to the class with the next highest value.

Research Organization:
Sandia National Laboratories (SNL-NM), Albuquerque, NM (United States)
Sponsoring Organization:
USDOE
DOE Contract Number:
NA0003525
Assignee:
National Technology & Engineering Solutions of Sandia, LLC (Albuquerque, NM)
Patent Number(s):
11,379,991
Application Number:
16/887,311
OSTI ID:
1924873
Country of Publication:
United States
Language:
English

References (1)

V-Net: Fully Convolutional Neural Networks for Volumetric Medical Image Segmentation conference October 2016

Similar Records

Uncertainty-refined image segmentation under domain shift
Patent · Mon Dec 16 23:00:00 EST 2024 · OSTI ID:2542924

Bayesian SegNet for Semantic Segmentation with Improved Interpretation of Microstructural Evolution During Irradiation of Materials
Journal Article · Tue Jul 01 00:00:00 EDT 2025 · Computational Materials Science · OSTI ID:2569566

Automated segmentation of porous thermal spray material CT scans with predictive uncertainty estimation
Journal Article · Wed May 31 00:00:00 EDT 2023 · Computational Mechanics · OSTI ID:1975896

Related Subjects