Wire-tracking of bent electric cable using X-ray CT and deep active learning

We have demonstrated a quantification of all component wires in a bent electric cable, which is necessary for discussion of cable products in actual use cases. Quantification became possible for the first time because of our new technologies for image analysis of bent cables. In this paper, various...

Full description

Saved in:
Bibliographic Details
Published inMicroscopy
Main Authors Hoshina, Yutaka, Yamamoto, Takuma, Uemura, Shigeaki
Format Journal Article
LanguageEnglish
Published England 25.05.2024
Subjects
wire-tracking
deep active learning
X-ray CT
electric cable
Online AccessGet more information

Cover

More Information
Summary:We have demonstrated a quantification of all component wires in a bent electric cable, which is necessary for discussion of cable products in actual use cases. Quantification became possible for the first time because of our new technologies for image analysis of bent cables. In this paper, various image analysis techniques to detect all wire tracks in a bent cable are demonstrated. Unique cross-sectional image construction and deep active learning schemes are the most important items in this study. These methods allow us to know the actual state of cables under external loads, which makes it possible to elucidate the mechanisms of various phenomena related to cables in the field and further improve the quality of cable products.
ISSN:2050-5701
DOI:10.1093/jmicro/dfae028