Artificial neural network approach for multiphase segmentation of battery electrode nano-CT images
Abstract The segmentation of tomographic images of the battery electrode is a crucial processing step, which will have an additional impact on the results of material characterization and electrochemical simulation. However, manually labeling X-ray CT images (XCT) is time-consuming, and these XCT im...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2022-02-01
|
| Series: | npj Computational Materials |
| Online Access: | https://doi.org/10.1038/s41524-022-00709-7 |
| _version_ | 1818330565134254080 |
|---|---|
| author | Zeliang Su Etienne Decencière Tuan-Tu Nguyen Kaoutar El-Amiry Vincent De Andrade Alejandro A. Franco Arnaud Demortière |
| author_facet | Zeliang Su Etienne Decencière Tuan-Tu Nguyen Kaoutar El-Amiry Vincent De Andrade Alejandro A. Franco Arnaud Demortière |
| author_sort | Zeliang Su |
| collection | DOAJ |
| description | Abstract The segmentation of tomographic images of the battery electrode is a crucial processing step, which will have an additional impact on the results of material characterization and electrochemical simulation. However, manually labeling X-ray CT images (XCT) is time-consuming, and these XCT images are generally difficult to segment with histographical methods. We propose a deep learning approach with an asymmetrical depth encode-decoder convolutional neural network (CNN) for real-world battery material datasets. This network achieves high accuracy while requiring small amounts of labeled data and predicts a volume of billions voxel within few minutes. While applying supervised machine learning for segmenting real-world data, the ground truth is often absent. The results of segmentation are usually qualitatively justified by visual judgement. We try to unravel this fuzzy definition of segmentation quality by identifying the uncertainty due to the human bias diluted in the training data. Further CNN trainings using synthetic data show quantitative impact of such uncertainty on the determination of material’s properties. Nano-XCT datasets of various battery materials have been successfully segmented by training this neural network from scratch. We will also show that applying the transfer learning, which consists of reusing a well-trained network, can improve the accuracy of a similar dataset. |
| first_indexed | 2024-12-13T13:05:58Z |
| format | Article |
| id | doaj.art-f767a38894eb47d3a701adbd70ad92f7 |
| institution | Directory Open Access Journal |
| issn | 2057-3960 |
| language | English |
| last_indexed | 2024-12-13T13:05:58Z |
| publishDate | 2022-02-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | npj Computational Materials |
| spelling | doaj.art-f767a38894eb47d3a701adbd70ad92f72022-12-21T23:44:49ZengNature Portfolionpj Computational Materials2057-39602022-02-018111110.1038/s41524-022-00709-7Artificial neural network approach for multiphase segmentation of battery electrode nano-CT imagesZeliang Su0Etienne Decencière1Tuan-Tu Nguyen2Kaoutar El-Amiry3Vincent De Andrade4Alejandro A. Franco5Arnaud Demortière6Laboratoire de Réactivité et Chimie des Solides (LRCS), CNRS UMR 7314, Université de Picardie Jules Verne, Hub de l’Energie, Rue BaudelocqueMINES ParisTech – PSL Research University, CMM, Center for Mathematical MorphologyLaboratoire de Réactivité et Chimie des Solides (LRCS), CNRS UMR 7314, Université de Picardie Jules Verne, Hub de l’Energie, Rue BaudelocqueLaboratoire de Réactivité et Chimie des Solides (LRCS), CNRS UMR 7314, Université de Picardie Jules Verne, Hub de l’Energie, Rue BaudelocqueAdvanced Photon Source, Argonne National LaboratoryLaboratoire de Réactivité et Chimie des Solides (LRCS), CNRS UMR 7314, Université de Picardie Jules Verne, Hub de l’Energie, Rue BaudelocqueLaboratoire de Réactivité et Chimie des Solides (LRCS), CNRS UMR 7314, Université de Picardie Jules Verne, Hub de l’Energie, Rue BaudelocqueAbstract The segmentation of tomographic images of the battery electrode is a crucial processing step, which will have an additional impact on the results of material characterization and electrochemical simulation. However, manually labeling X-ray CT images (XCT) is time-consuming, and these XCT images are generally difficult to segment with histographical methods. We propose a deep learning approach with an asymmetrical depth encode-decoder convolutional neural network (CNN) for real-world battery material datasets. This network achieves high accuracy while requiring small amounts of labeled data and predicts a volume of billions voxel within few minutes. While applying supervised machine learning for segmenting real-world data, the ground truth is often absent. The results of segmentation are usually qualitatively justified by visual judgement. We try to unravel this fuzzy definition of segmentation quality by identifying the uncertainty due to the human bias diluted in the training data. Further CNN trainings using synthetic data show quantitative impact of such uncertainty on the determination of material’s properties. Nano-XCT datasets of various battery materials have been successfully segmented by training this neural network from scratch. We will also show that applying the transfer learning, which consists of reusing a well-trained network, can improve the accuracy of a similar dataset.https://doi.org/10.1038/s41524-022-00709-7 |
| spellingShingle | Zeliang Su Etienne Decencière Tuan-Tu Nguyen Kaoutar El-Amiry Vincent De Andrade Alejandro A. Franco Arnaud Demortière Artificial neural network approach for multiphase segmentation of battery electrode nano-CT images npj Computational Materials |
| title | Artificial neural network approach for multiphase segmentation of battery electrode nano-CT images |
| title_full | Artificial neural network approach for multiphase segmentation of battery electrode nano-CT images |
| title_fullStr | Artificial neural network approach for multiphase segmentation of battery electrode nano-CT images |
| title_full_unstemmed | Artificial neural network approach for multiphase segmentation of battery electrode nano-CT images |
| title_short | Artificial neural network approach for multiphase segmentation of battery electrode nano-CT images |
| title_sort | artificial neural network approach for multiphase segmentation of battery electrode nano ct images |
| url | https://doi.org/10.1038/s41524-022-00709-7 |
| work_keys_str_mv | AT zeliangsu artificialneuralnetworkapproachformultiphasesegmentationofbatteryelectrodenanoctimages AT etiennedecenciere artificialneuralnetworkapproachformultiphasesegmentationofbatteryelectrodenanoctimages AT tuantunguyen artificialneuralnetworkapproachformultiphasesegmentationofbatteryelectrodenanoctimages AT kaoutarelamiry artificialneuralnetworkapproachformultiphasesegmentationofbatteryelectrodenanoctimages AT vincentdeandrade artificialneuralnetworkapproachformultiphasesegmentationofbatteryelectrodenanoctimages AT alejandroafranco artificialneuralnetworkapproachformultiphasesegmentationofbatteryelectrodenanoctimages AT arnauddemortiere artificialneuralnetworkapproachformultiphasesegmentationofbatteryelectrodenanoctimages |