Related Experiment Video
Updated: Jun 25, 2025
10:56
Long-term Behavioral Tracking of Freely Swimming Weakly Electric Fish
Published on: March 6, 2014
12.5K
Wire-tracking of bent electric cable using X-ray CT and deep active learning
Yutaka Hoshina1, Takuma Yamamoto1, Shigeaki Uemura1
1Analysis Technology Research Center R&D Unit, Sumitomo Electric Industries, Ltd., 1-1-3 Shimaya, Konohana-ku, Osaka 554-0024, Japan.
Microscopy (Oxford, England)
|May 25, 2024
Summary
New image analysis technologies enable precise quantification of wires within bent electric cables. This breakthrough allows for better understanding and improvement of cable products used in real-world applications.
Area of Science:
- Electrical Engineering
- Materials Science
- Computer Vision
Background:
- Assessing the condition of electric cables under load is crucial for understanding field performance.
- Existing methods lack the precision to quantify individual wires within complex, bent cable structures.
Purpose of the Study:
- To develop and demonstrate novel image analysis techniques for quantifying all component wires in bent electric cables.
- To enable a deeper understanding of cable behavior under operational stress.
Main Methods:
- Development of unique cross-sectional image construction techniques.
- Implementation of deep active learning schemes for wire track detection.
- Application of advanced image analysis for bent cable assessment.
Main Results:
- Successful quantification of all component wires in bent electric cables.
- Demonstration of various image analysis techniques for wire track detection.
- Establishment of methods to assess cable state under external loads.
Conclusions:
- The developed image analysis technologies provide unprecedented quantification of wires in bent cables.
- This capability is essential for analyzing cable products in actual use cases.
- The findings facilitate elucidation of field-related cable phenomena and quality improvement.
