Unveiling human eye temperature with deep learning-powered segmentation DOI

J. Persiya,

A. Sasithradevi

Biomedical Signal Processing and Control, Journal Year: 2024, Volume and Issue: 102, P. 107256 - 107256

Published: Nov. 27, 2024

Language: Английский

Deep Learning-based Fracture Mode Determination in Composite Laminates DOI
Muhammad Muzammil Azad, Atta ur Rehman Shah, M. Prabhakar

et al.

Journal of the Computational Structural Engineering Institute of Korea, Journal Year: 2024, Volume and Issue: 37(4), P. 225 - 232

Published: Aug. 31, 2024

Language: Английский

Citations

10
Deep learning-based autonomous morphological fracture analysis of fiber-reinforced composites DOI
Muhammad Muzammil Azad, Atta ur Rehman Shah, M. Prabhakar

et al.

Engineering Failure Analysis, Journal Year: 2025, Volume and Issue: unknown, P. 109292 - 109292

Published: Jan. 1, 2025

Language: Английский

Citations

1
Multimodal Data Fusion Enhanced Deep Learning Prediction of Crack Path Segmentation in CFRP Composites DOI
Peng Zhang, Keke Tang, Guangxu Chen

et al.

Composites Science and Technology, Journal Year: 2024, Volume and Issue: 257, P. 110812 - 110812

Published: Aug. 13, 2024

Language: Английский

Citations

6
A deep learning model to identify the initial parameters of dissimilar welded joints under impact fracture DOI Creative Commons
Carlos Avilés‐Cruz, Miriam Aguilar, B. Vargas-Arista

et al.

Welding in the World, Journal Year: 2025, Volume and Issue: unknown

Published: Jan. 24, 2025

Language: Английский

Citations

0
Research on the performance of the SegFormer model with fusion of edge feature extraction for metal corrosion detection DOI Creative Commons

Bingnan Yan,

Conghui Wang,

Xiaolong Hao

et al.

Scientific Reports, Journal Year: 2025, Volume and Issue: 15(1)

Published: March 8, 2025

Addressing the challenge that existing deep learning models face in accurately segmenting metal corrosion boundaries and small areas. In this paper, a SegFormer detection method based on parallel extraction of edge features is proposed. Firstly, to solve boundary ambiguity problem images, an edge-feature module (EEM) introduced construct spatial branch network assist model extracting shallow details information from images. Secondly, mitigate loss target feature during reconstruction decoder, paper adopts gradual upsampling decoding layer design. It introduces fusion (FFM) achieve hierarchical progressive fusion, thereby enhancing corroded Experimental results show proposed outperforms other semantic segmentation achieving accuracy 86.56% public surface image dataset reaching mean intersection over union (mIoU) 91.41% BSData defect dataset. On Self-built tubing pit dataset, utilizes only 3.60 MB parameters 96.52%, confirming effectiveness performance advantages practical applications.

Language: Английский

Citations

0
Multi-scale semantic segmentation for fiber identification and 3D reconstruction of unidirectional composite DOI
Peng Zhang, Xun Zhou, Ruifeng Liang

et al.

Composites Science and Technology, Journal Year: 2025, Volume and Issue: unknown, P. 111160 - 111160

Published: March 1, 2025

Language: Английский

Citations

0
Parallel simulation and prediction techniques for digital twins in urban underground spaces DOI

Haofeng Gong,

Dong Su,

Shiqi Zeng

et al.

Automation in Construction, Journal Year: 2025, Volume and Issue: 175, P. 106212 - 106212

Published: April 21, 2025

Language: Английский

Citations

0
A GAN-based stepwise full-field mechanical prediction model for architected metamaterials DOI
Yujie Xiang, Jixin Hou, Xianyan Chen

et al.

International Journal of Mechanical Sciences, Journal Year: 2024, Volume and Issue: 284, P. 109771 - 109771

Published: Oct. 8, 2024

Language: Английский

Citations

3
Graphical Feature Construction-Based Deep Learning Model for Fatigue Life Prediction of AM Alloys DOI Open Access
Hao Wu, Anbin Wang, Zhiqiang Gan

et al.

Materials, Journal Year: 2024, Volume and Issue: 18(1), P. 11 - 11

Published: Dec. 24, 2024

Fatigue failure poses a serious challenge for ensuring the operational safety of critical components subjected to cyclic/random loading. In this context, various machine learning (ML) models have been increasingly explored, due their effectiveness in analyzing relationship between fatigue life and multiple influencing factors. Nevertheless, existing ML hinge heavily on numeric features as inputs, which encapsulate limited information process interest. To cure deficiency, novel model based upon convolutional neural networks is developed, where are transformed into graphical ones by introducing two enrichment operations, namely, Shapley Additive Explanations Pearson correlation coefficient analysis. Additionally, attention mechanism introduced prioritize important regions image-based inputs. Extensive validations using experimental results laser powder bed fusion-fabricated metals demonstrate that proposed possesses better predictive accuracy than conventional models.

Language: Английский

Citations

1
Unveiling human eye temperature with deep learning-powered segmentation DOI

J. Persiya,

A. Sasithradevi

Biomedical Signal Processing and Control, Journal Year: 2024, Volume and Issue: 102, P. 107256 - 107256

Published: Nov. 27, 2024

Language: Английский

Citations

0