MLTiming: A Machine Learning Framework for Gamma-Ray Pulse Time Pick-Up. DOI

Julian Avellaneda,

V. Sánchez-Tembleque, L. M. Fraile

et al.

Published: Sept. 25, 2024

Language: Английский

Online sequential Extreme learning Machine (OSELM) based denoising of encrypted image DOI

Biniyam Ayele Belete,

Demissie Jobir Gelmecha, Ram Sewak Singh

et al.

Expert Systems with Applications, Journal Year: 2025, Volume and Issue: unknown, P. 126999 - 126999

Published: Feb. 1, 2025

Language: Английский

Citations

0
Time-of-Flight Based One-Dimensional Position Estimation of Radioactive Sources Using Artificial Neural Network Model DOI Creative Commons
Jin‐Hong Kim,

Siwon Song,

Jae Hyung Park

et al.

Nuclear Engineering and Technology, Journal Year: 2025, Volume and Issue: unknown, P. 103662 - 103662

Published: April 1, 2025

Language: Английский

Citations

0
Multiclass small target detection algorithm for surface defects of chemicals special steel DOI Creative Commons
Yuanyuan Wang,

Shaofeng Yan,

Hauwa Suleiman Abdullahi

et al.

Frontiers in Physics, Journal Year: 2024, Volume and Issue: 12

Published: Aug. 30, 2024

Introduction: Chemical special steels are widely used in chemical equipment manufacturing and other fields, small defects on its surface (such as cracks punches) easy to cause serious accidents harsh environments. Methods: In order solve this problem, paper proposes an improved defect detection algorithm for steel based YOLOv8. Firstly, effectively capture local global information, a ParC2Net (Parallel-C2f) structure is proposed feature extraction, which can accurately the subtle features of defects. Secondly, loss function adjusted MPD-IOU, dynamic non-monotonic focusing characteristics overfitting problem bounding box low-quality targets. addition, RepGFPN fuse multi-scale features, deepen interaction between semantics spatial significantly improve efficiency cross-layer information transmission. Finally, RexSE-Head (ResNeXt-Squeeze-Excitation) design adopted enhance positioning accuracy Results discussion: The experimental results show that [email protected] model reaches 93.5%, number parameters only 3.29M, realizes high precision response performance steels, highlights practical application value model. code available at https://github.com/improvment/prs-yolo .

Language: Английский

Citations

2
MLTiming: A Machine Learning Framework for Gamma-Ray Pulse Time Pick-Up. DOI

Julian Avellaneda,

V. Sánchez-Tembleque, L. M. Fraile

et al.

Published: Sept. 25, 2024

Language: Английский

Citations

0