Shigeki Yumoto
Papers
1
Total Citations
8
H-Index
1
About
Shigeki Yumoto is a researcher at the forefront of applying generative adversarial networks (GANs) to infrastructure inspection. His key research areas include anomaly detection, computer vision, and robotic inspection systems for aging civil infrastructure. Yumoto’s most notable contribution is his pioneering work on using GANs to identify defects in sewer pipe images captured by earthworm-type inspection robots. His 2023 paper, "Anomaly detection from images in pipes using GAN," which has garnered 8 citations, addresses a critical gap in automated inspection methods for pipes that have exceeded their service life. By leveraging GANs’ ability to learn normal pipe surface patterns and flag anomalies, Yumoto has provided a foundation for more reliable, cost-effective infrastructure maintenance. His work bridges robotics, deep learning, and civil engineering, offering a scalable solution to a pressing societal problem. Yumoto’s research is particularly impactful for students and engineers interested in applying AI to real-world challenges, demonstrating how cutting-edge techniques can transform routine safety inspections and prevent costly failures.
Research Focus
Key Achievements
Top Papers
- 1Anomaly detection from images in pipes using GAN8 citations · 2023