Deep Learning and Bayesian Hyperparameter Optimization: A Data-Driven Approach for Diamond Grit Segmentation toward Grinding Wheel Characterization
Diamond grinding wheels (DGWs) have a central role in cutting-edge industries such as aeronautics or defense and spatial applications. Characterizations of DGWs are essential to optimize the design and machining performance of such cutting tools. Thus, the critical issue of DGW characterization lies...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-12-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/12/24/12606 |
| _version_ | 1797461660035710976 |
|---|---|
| author | Damien Sicard Pascal Briois Alain Billard Jérôme Thevenot Eric Boichut Julien Chapellier Frédéric Bernard |
| author_facet | Damien Sicard Pascal Briois Alain Billard Jérôme Thevenot Eric Boichut Julien Chapellier Frédéric Bernard |
| author_sort | Damien Sicard |
| collection | DOAJ |
| description | Diamond grinding wheels (DGWs) have a central role in cutting-edge industries such as aeronautics or defense and spatial applications. Characterizations of DGWs are essential to optimize the design and machining performance of such cutting tools. Thus, the critical issue of DGW characterization lies in the detection of diamond grits. However, the traditional diamond detection methods rely on manual operations on DGW images. These methods are time-consuming, error-prone and inaccurate. In addition, the manual detection of diamond grits remains challenging even for a subject expert. To overcome these shortcomings, we introduce a deep learning approach for automatic diamond grit segmentation. Due to our small dataset of 153 images, the proposed approach leverages transfer learning techniques with pre-trained ResNet34 as an encoder of U-Net CNN architecture. Moreover, with more than 8600 hyperparameter combinations in our model, manually finding the best configuration is impossible. That is why we use a Bayesian optimization algorithm using Hyperband early stopping mechanisms to automatically explore the search space and find the best hyperparameter values. Moreover, considering our small dataset, we obtain overall satisfactory performance with over 53% IoU and 69% F1-score. Finally, this work provides a first step toward diamond grinding wheel characterization by using a data-driven approach for automatic semantic segmentation of diamond grits. |
| first_indexed | 2024-03-09T17:23:33Z |
| format | Article |
| id | doaj.art-ecc12946f5124892ac96f1d4a7ed83d8 |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-03-09T17:23:33Z |
| publishDate | 2022-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-ecc12946f5124892ac96f1d4a7ed83d82023-11-24T13:01:34ZengMDPI AGApplied Sciences2076-34172022-12-0112241260610.3390/app122412606Deep Learning and Bayesian Hyperparameter Optimization: A Data-Driven Approach for Diamond Grit Segmentation toward Grinding Wheel CharacterizationDamien Sicard0Pascal Briois1Alain Billard2Jérôme Thevenot3Eric Boichut4Julien Chapellier5Frédéric Bernard6Laboratoire Interdisciplinaire Carnot de Bourgogne, PMDM, ICB-UMR6303, CNRS, Université de Bourgogne Franche-Comté, 9 Avenue Alain Savary, BP47870, CEDEX, 21078 Dijon, FranceFEMTO-ST UMR 6174, CNRS, Université de Bourgogne Franche-Comté, UTBM, F-90010 Belfort, FranceFEMTO-ST UMR 6174, CNRS, Université de Bourgogne Franche-Comté, UTBM, F-90010 Belfort, FranceDIAMATEC, Route de Grachaux, 70700 Oiselay-et-Grachaux, FranceDIAMATEC, Route de Grachaux, 70700 Oiselay-et-Grachaux, FranceDIAMATEC, Route de Grachaux, 70700 Oiselay-et-Grachaux, FranceLaboratoire Interdisciplinaire Carnot de Bourgogne, PMDM, ICB-UMR6303, CNRS, Université de Bourgogne Franche-Comté, 9 Avenue Alain Savary, BP47870, CEDEX, 21078 Dijon, FranceDiamond grinding wheels (DGWs) have a central role in cutting-edge industries such as aeronautics or defense and spatial applications. Characterizations of DGWs are essential to optimize the design and machining performance of such cutting tools. Thus, the critical issue of DGW characterization lies in the detection of diamond grits. However, the traditional diamond detection methods rely on manual operations on DGW images. These methods are time-consuming, error-prone and inaccurate. In addition, the manual detection of diamond grits remains challenging even for a subject expert. To overcome these shortcomings, we introduce a deep learning approach for automatic diamond grit segmentation. Due to our small dataset of 153 images, the proposed approach leverages transfer learning techniques with pre-trained ResNet34 as an encoder of U-Net CNN architecture. Moreover, with more than 8600 hyperparameter combinations in our model, manually finding the best configuration is impossible. That is why we use a Bayesian optimization algorithm using Hyperband early stopping mechanisms to automatically explore the search space and find the best hyperparameter values. Moreover, considering our small dataset, we obtain overall satisfactory performance with over 53% IoU and 69% F1-score. Finally, this work provides a first step toward diamond grinding wheel characterization by using a data-driven approach for automatic semantic segmentation of diamond grits.https://www.mdpi.com/2076-3417/12/24/12606deep learningBayesian hyperparameter optimizationcomputer visionsemantic segmentationU-Netdiamond abrasive grits |
| spellingShingle | Damien Sicard Pascal Briois Alain Billard Jérôme Thevenot Eric Boichut Julien Chapellier Frédéric Bernard Deep Learning and Bayesian Hyperparameter Optimization: A Data-Driven Approach for Diamond Grit Segmentation toward Grinding Wheel Characterization Applied Sciences deep learning Bayesian hyperparameter optimization computer vision semantic segmentation U-Net diamond abrasive grits |
| title | Deep Learning and Bayesian Hyperparameter Optimization: A Data-Driven Approach for Diamond Grit Segmentation toward Grinding Wheel Characterization |
| title_full | Deep Learning and Bayesian Hyperparameter Optimization: A Data-Driven Approach for Diamond Grit Segmentation toward Grinding Wheel Characterization |
| title_fullStr | Deep Learning and Bayesian Hyperparameter Optimization: A Data-Driven Approach for Diamond Grit Segmentation toward Grinding Wheel Characterization |
| title_full_unstemmed | Deep Learning and Bayesian Hyperparameter Optimization: A Data-Driven Approach for Diamond Grit Segmentation toward Grinding Wheel Characterization |
| title_short | Deep Learning and Bayesian Hyperparameter Optimization: A Data-Driven Approach for Diamond Grit Segmentation toward Grinding Wheel Characterization |
| title_sort | deep learning and bayesian hyperparameter optimization a data driven approach for diamond grit segmentation toward grinding wheel characterization |
| topic | deep learning Bayesian hyperparameter optimization computer vision semantic segmentation U-Net diamond abrasive grits |
| url | https://www.mdpi.com/2076-3417/12/24/12606 |
| work_keys_str_mv | AT damiensicard deeplearningandbayesianhyperparameteroptimizationadatadrivenapproachfordiamondgritsegmentationtowardgrindingwheelcharacterization AT pascalbriois deeplearningandbayesianhyperparameteroptimizationadatadrivenapproachfordiamondgritsegmentationtowardgrindingwheelcharacterization AT alainbillard deeplearningandbayesianhyperparameteroptimizationadatadrivenapproachfordiamondgritsegmentationtowardgrindingwheelcharacterization AT jeromethevenot deeplearningandbayesianhyperparameteroptimizationadatadrivenapproachfordiamondgritsegmentationtowardgrindingwheelcharacterization AT ericboichut deeplearningandbayesianhyperparameteroptimizationadatadrivenapproachfordiamondgritsegmentationtowardgrindingwheelcharacterization AT julienchapellier deeplearningandbayesianhyperparameteroptimizationadatadrivenapproachfordiamondgritsegmentationtowardgrindingwheelcharacterization AT fredericbernard deeplearningandbayesianhyperparameteroptimizationadatadrivenapproachfordiamondgritsegmentationtowardgrindingwheelcharacterization |