2022
DOI: 10.1007/s42243-021-00719-7
|View full text |Cite
|
Sign up to set email alerts
|

A deep learning-based method for segmentation and quantitative characterization of microstructures in weathering steel from sequential scanning electron microscope images

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1

Citation Types

0
4
0

Year Published

2022
2022
2024
2024

Publication Types

Select...
8

Relationship

1
7

Authors

Journals

citations
Cited by 16 publications
(4 citation statements)
references
References 38 publications
0
4
0
Order By: Relevance
“…One of the reasons for using ML is that it can handle variances well, as long as they are represented in the training data. However, although ML has already demonstrated its potential for a variety of tasks in microstructural analysis, the majority of publications deal with well-curated datasets, generated under laboratory conditions and exhibiting little variation (e.g., [17,[24][25][26][27][28][29])-and a general understanding of robustness and generalization, as well as dataset size, occurring variances and the maximum variance an ML model can handle, is therefore still lacking.…”
Section: Domain Challenges and The Role Of The Ground Truthmentioning
confidence: 99%
“…One of the reasons for using ML is that it can handle variances well, as long as they are represented in the training data. However, although ML has already demonstrated its potential for a variety of tasks in microstructural analysis, the majority of publications deal with well-curated datasets, generated under laboratory conditions and exhibiting little variation (e.g., [17,[24][25][26][27][28][29])-and a general understanding of robustness and generalization, as well as dataset size, occurring variances and the maximum variance an ML model can handle, is therefore still lacking.…”
Section: Domain Challenges and The Role Of The Ground Truthmentioning
confidence: 99%
“…The location information of extracted features was restored to the aluminum alloy section, and the quantitative statistical results with spatial distribution information could be obtained. For verification of the method accuracy and results, refer to the previous results published by our research group [46]. The network we constructed in the dotted box based on the original U-Net architecture consisted of a contracting path and an expanding path [43][44][45].…”
Section: Experimental Samplesmentioning
confidence: 99%
“…The location information of extracted features was restored to the aluminum alloy section, and the quantitative statistical results with spatial distribution information could be obtained. For verification of the method accuracy and results, refer to the previous results published by our research group [46].…”
Section: Experimental Samplesmentioning
confidence: 99%
“…Han [ 14 ] proposed a high-throughput characterization method based on deep learning, rapid acquisition techniques, and mathematical statistics to identify, segment, and quantify the microstructure of weathering steel. The segmentation accuracies of 89.95% and 90.86% for non-metallic inclusions and pearlite phases, respectively, and the detection time are significantly reduced.…”
Section: Introductionmentioning
confidence: 99%