Deep stochastic image segmentation for autonomous robotic inspection

Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review

Abstract

Corrosion detection inside a vessel’s ballast tank is a dangerous task that is normally performed by human specialists who, due to standards that are not easily quantifiable, can cause a high level of subjectivity in the inspection process. The Inspectrone project aims to automate this process using a drone flying through the different ballast tanks to perform the inspection, avoiding possible risks for the crew. Due to the subjectivity of the task a standard, deterministic, model for semantic segmentation can not be used for this scenario. On the contrary, a model that performs stochastic segmentation is required to enable the encapsulation of multiple differentiating human specialist opinions. The architecture presented in this paper combines a segmentation model with a latent distribution to encode the segmentation variants of the dataset to solve this task by being able to produce, for the same input image, different, but all plausible segmentations. The designed model is defined as Probabilistic GSCNN since it employs the state-of-the-art Gated Shape CNN to perform segmentation and a fixed prior distribution to produce a probabilistic output. The proposed method was able to outperform the baseline method, the Probabilistic U-Net, with a 9% increase in accuracy and without the need to train the CVAE module allowing a faster and less resource-demanding development.
Original languageEnglish
Title of host publicationProceedings of 2023 IEEE International Conference on Imaging Systems and Techniques
Number of pages6
PublisherIEEE
Publication date19 Oct 2023
Article number10355654
ISBN (Print)979-8-3503-3084-7
DOIs
Publication statusPublished - 19 Oct 2023
Event2023 IEEE International Conference on Imaging Systems and Techniques - Technical University of Denmark, Copenhagen, Denmark
Duration: 17 Oct 202319 Oct 2023

Conference

Conference2023 IEEE International Conference on Imaging Systems and Techniques
LocationTechnical University of Denmark
Country/TerritoryDenmark
CityCopenhagen
Period17/10/202319/10/2023

Keywords

  • Training
  • Image segmentation
  • Shape
  • Stochastic processes
  • Inspection
  • Logic gates
  • Probabilistic logic

Fingerprint

Dive into the research topics of 'Deep stochastic image segmentation for autonomous robotic inspection'. Together they form a unique fingerprint.

Cite this