A transferred multitask regularization convolutional neural network (TrMR-CNN) for laser-induced breakdown spectroscopy quantitative analysis

Cui, Jiacheng; Song, Weiran; Hou, Zongyu; Gu, Weilun; Wang, Zhe

doi:10.1039/d2ja00182a

Cited by 10 publications

(6 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Davari et al [82] fused the Lasso model with a one-dimensional CNN method to analyze LIBS spectra of single-crystal silicon and measured the oxygen-related impurities in the samples with LOD as low as 0-16 ppm. Cui et al [83] proposed a migration-learning multitask regularized CNN model for quantitative analysis of coal samples on a dataset with small sample size. Compared with PLS, SVM regression, and ordinary CNN networks, the RMSE of this method was reduced by 19.9%, 5.9%, and 7.7%, respectively.…”

Section: Convolutional Neural Network (Cnn)mentioning

confidence: 99%

recent advances in machine learning methodologies for LIBS quantitative analysis

Liu,

Han,

Yang

et al. 2024

Pulsed Laser Processing of Materials

View full text Add to dashboard Cite

The mapping between LIBS spectral data to the quantitative results can become highly complicated and nonlinear due to experimental conditions, sample surface state, matrix effect, self-absorption, etc. Therefore, the accurate quantitative analysis is the longstanding dream of the LIBS community. The advantages of machine learning in dealing with high-dimensional and nonlinear problems have made it a cutting-edge hot topic in quantitative LIBS in recent years. This chapter introduces the current bottlenecks in quantitative LIBS, sorts out the data processing methods, and reviews the research status and progress of conventional machine learning methods such as PLS, SVM, LSSVM, Lasso, and artificial neural network-based methods. By comparing the results of different methods, the perspective of future developments on learning-based methods is discussed. This chapter aims to review the applications of the combination of quantitative LIBS and machine learning methods and demonstrate the performance of different machine learning methods based on experimental results.

show abstract

Section: Convolutional Neural Network (Cnn)mentioning

confidence: 99%

recent advances in machine learning methodologies for LIBS quantitative analysis

Liu,

Han,

Yang

et al. 2024

Pulsed Laser Processing of Materials

View full text Add to dashboard Cite

show abstract

“…The techniques establish connections between source and target domains that have different but related distributions and can share features between domains. Cui et al 15 proposed a transfer learning method called transferred multitask regularization convolutional neural network (TrMR-CNN) to address the small-sample problem in LIBS quantification analysis and improve the generalization ability of transfer learning. Compared with other machine learning methods, TrMR-CNN with high accuracy has great power in LIBS quantification analysis.…”

Section: Introductionmentioning

confidence: 99%

Rapid quantitative analysis of raw rocks by LIBS coupled with feature-based transfer learning

Rao,

Ren,

Kong

et al. 2024

J. Anal. At. Spectrom.

View full text Add to dashboard Cite

show abstract

“…Zhang et al [16] used wavelet neural network (WNN) to classify coal ash, and the results showed that WNN showed better classification performance. Cui et al [17] used CNN, combined with multi-task regularization, proposed a transfer learning method which significantly reduced the RMSEP compared with the baseline method. Chen et al [18] proposed a moisture-spectral intensity correction model, and established an ANN model to improve the model prediction capability.…”

Section: Introductionmentioning

confidence: 99%

Application of Semi-Supervised Learning Model to Coal Sample Classification

Wang,

Xu,

Gao

et al. 2024

Applied Sciences

View full text Add to dashboard Cite

As an extremely important energy source, improving the efficiency and accuracy of coal classification is important for industrial production and pollution reduction. Laser-induced breakdown spectroscopy (LIBS) is a new technology for coal classification which has the ability to rapidly analyze coal compared with traditional coal analysis methods. In the practical application of LIBS, a large amount of labeling data is usually required, but it is quite difficult to obtain labeling data in industrial sites. In this paper, to address the problem of insufficient labeled data, a semi-supervised classification model (SGAN) based on adversarial neural network is proposed, which can utilize unlabeled data to improve the classification accuracy. The effects of labeled and unlabeled samples on the classification accuracy of the SGAN model are investigated, and the results show that the number of labeled and unlabeled samples are positively correlated, and the highest average classification accuracy that the model can achieve is 98.5%. In addition, the classification accuracies of SGAN and other models (e.g., CNN, RF) are also compared, and the results show that, with the same number of labeled samples in the three models, SGAN performs better after the number of unlabeled samples reaches a certain level, with an improvement of 0.7% and 2.5% compared to the CNN and RF models, respectively. This study provides new ideas for the application of semi-supervised learning in LIBS.

show abstract

A transferred multitask regularization convolutional neural network (TrMR-CNN) for laser-induced breakdown spectroscopy quantitative analysis

Abstract: Laser-induced breakdown spectroscopy (LIBS) combined with machine learning has demonstrated great capabilities for quantitative elemental analysis. When the distributions of training and test data differ due to changes in measurement...

Cited by 10 publications

References 30 publications

recent advances in machine learning methodologies for LIBS quantitative analysis

recent advances in machine learning methodologies for LIBS quantitative analysis

Rapid quantitative analysis of raw rocks by LIBS coupled with feature-based transfer learning

Application of Semi-Supervised Learning Model to Coal Sample Classification

Contact Info

Product

Resources

About