A Comparative Study of State-of-the-Art Deep Learning Models for Semantic Segmentation of Pores in Scanning Electron Microscope Images of Activated Carbon
Accurate measurement of the microspores, mesopores, and macropores on the surface of the activated carbon is essential due to its direct influence on the material’s adsorption capacity, surface area, and overall performance in various applications like water purification, air filtration,...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2024-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10478488/ |
| _version_ | 1797213580511150080 |
|---|---|
| author | Bishwas Pokharel Deep Shankar Pandey Anjuli Sapkota Bhimraj Yadav Vasanta Gurung Mandira Pradhananga Adhikari Lok Nath Regmi Nanda Bikram Adhikari |
| author_facet | Bishwas Pokharel Deep Shankar Pandey Anjuli Sapkota Bhimraj Yadav Vasanta Gurung Mandira Pradhananga Adhikari Lok Nath Regmi Nanda Bikram Adhikari |
| author_sort | Bishwas Pokharel |
| collection | DOAJ |
| description | Accurate measurement of the microspores, mesopores, and macropores on the surface of the activated carbon is essential due to its direct influence on the material’s adsorption capacity, surface area, and overall performance in various applications like water purification, air filtration, and gas separation. Traditionally, Scanning Electron Microscopy (SEM) images of activated carbons are collected and manually annotated by a human expert to differentiate and measure different pores on the surface. However, manual analysis of such surfaces is costly, time-consuming, and resource-intensive, as it requires expert supervision. In this paper, we propose an automatic deep-learning-based solution to address this challenge of activated carbon surface segmentation. Our deep-learning approach optimizes pore analysis by reducing time and resources, eliminating human subjectivity, and effectively adapting to diverse pore structures and imaging conditions. We introduce a novel SEM image segmentation dataset for activated carbon, comprising 128 images that capture the variability in pore sizes, structures, and imaging artifacts. Challenges encountered during dataset creation, irregularities in pore structures, and the presence of impurities were addressed to ensure robust model performance. We then evaluate the state-of-the-art deep learning models on the novel semantic segmentation task that shows promising results. Notably, DeepLabV3Plus, DeepLabV3, and FPN emerge as the most promising models based on semantic segmentation test results, with DeepLabV3Plus achieving the highest test Dice coefficient of 68.68%. Finally, we outline the key research challenges and discuss potential research directions to address these challenges. |
| first_indexed | 2024-04-24T11:00:32Z |
| format | Article |
| id | doaj.art-2fd9b337a1e84e53b0b5327a253f38ab |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-04-24T11:00:32Z |
| publishDate | 2024-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-2fd9b337a1e84e53b0b5327a253f38ab2024-04-11T23:00:26ZengIEEEIEEE Access2169-35362024-01-0112502175024310.1109/ACCESS.2024.338152310478488A Comparative Study of State-of-the-Art Deep Learning Models for Semantic Segmentation of Pores in Scanning Electron Microscope Images of Activated CarbonBishwas Pokharel0https://orcid.org/0009-0001-1351-8322Deep Shankar Pandey1Anjuli Sapkota2Bhimraj Yadav3https://orcid.org/0000-0003-3617-673XVasanta Gurung4Mandira Pradhananga Adhikari5https://orcid.org/0009-0001-1351-8322Lok Nath Regmi6Nanda Bikram Adhikari7https://orcid.org/0009-0001-1351-8322Department of Electronics and Computer Engineering, Pulchowk Campus, Institute of Engineering, Tribhuvan University, Kathmandu, NepalRochester Institute of Technology, Rochester, NY, USADepartment of Civil Engineering, Pulchowk Campus, IOE, Tribhuvan University, Kathmandu, NepalFetchly LLC, Austin, TX, USADepartment of Materials Science and Engineering, College of Engineering, University of North Texas, Denton, TX, USACentral Department of Chemistry, Tribhuvan University, Kirtipur, Kathmandu, NepalDepartment of Electronics and Computer Engineering, Pulchowk Campus, Institute of Engineering, Tribhuvan University, Kathmandu, NepalDepartment of Electronics and Computer Engineering, Pulchowk Campus, Institute of Engineering, Tribhuvan University, Kathmandu, NepalAccurate measurement of the microspores, mesopores, and macropores on the surface of the activated carbon is essential due to its direct influence on the material’s adsorption capacity, surface area, and overall performance in various applications like water purification, air filtration, and gas separation. Traditionally, Scanning Electron Microscopy (SEM) images of activated carbons are collected and manually annotated by a human expert to differentiate and measure different pores on the surface. However, manual analysis of such surfaces is costly, time-consuming, and resource-intensive, as it requires expert supervision. In this paper, we propose an automatic deep-learning-based solution to address this challenge of activated carbon surface segmentation. Our deep-learning approach optimizes pore analysis by reducing time and resources, eliminating human subjectivity, and effectively adapting to diverse pore structures and imaging conditions. We introduce a novel SEM image segmentation dataset for activated carbon, comprising 128 images that capture the variability in pore sizes, structures, and imaging artifacts. Challenges encountered during dataset creation, irregularities in pore structures, and the presence of impurities were addressed to ensure robust model performance. We then evaluate the state-of-the-art deep learning models on the novel semantic segmentation task that shows promising results. Notably, DeepLabV3Plus, DeepLabV3, and FPN emerge as the most promising models based on semantic segmentation test results, with DeepLabV3Plus achieving the highest test Dice coefficient of 68.68%. Finally, we outline the key research challenges and discuss potential research directions to address these challenges.https://ieeexplore.ieee.org/document/10478488/Semantic segmentationSEM imagesactivated carbonground truthAdam optimizerintersection over union |
| spellingShingle | Bishwas Pokharel Deep Shankar Pandey Anjuli Sapkota Bhimraj Yadav Vasanta Gurung Mandira Pradhananga Adhikari Lok Nath Regmi Nanda Bikram Adhikari A Comparative Study of State-of-the-Art Deep Learning Models for Semantic Segmentation of Pores in Scanning Electron Microscope Images of Activated Carbon IEEE Access Semantic segmentation SEM images activated carbon ground truth Adam optimizer intersection over union |
| title | A Comparative Study of State-of-the-Art Deep Learning Models for Semantic Segmentation of Pores in Scanning Electron Microscope Images of Activated Carbon |
| title_full | A Comparative Study of State-of-the-Art Deep Learning Models for Semantic Segmentation of Pores in Scanning Electron Microscope Images of Activated Carbon |
| title_fullStr | A Comparative Study of State-of-the-Art Deep Learning Models for Semantic Segmentation of Pores in Scanning Electron Microscope Images of Activated Carbon |
| title_full_unstemmed | A Comparative Study of State-of-the-Art Deep Learning Models for Semantic Segmentation of Pores in Scanning Electron Microscope Images of Activated Carbon |
| title_short | A Comparative Study of State-of-the-Art Deep Learning Models for Semantic Segmentation of Pores in Scanning Electron Microscope Images of Activated Carbon |
| title_sort | comparative study of state of the art deep learning models for semantic segmentation of pores in scanning electron microscope images of activated carbon |
| topic | Semantic segmentation SEM images activated carbon ground truth Adam optimizer intersection over union |
| url | https://ieeexplore.ieee.org/document/10478488/ |
| work_keys_str_mv | AT bishwaspokharel acomparativestudyofstateoftheartdeeplearningmodelsforsemanticsegmentationofporesinscanningelectronmicroscopeimagesofactivatedcarbon AT deepshankarpandey acomparativestudyofstateoftheartdeeplearningmodelsforsemanticsegmentationofporesinscanningelectronmicroscopeimagesofactivatedcarbon AT anjulisapkota acomparativestudyofstateoftheartdeeplearningmodelsforsemanticsegmentationofporesinscanningelectronmicroscopeimagesofactivatedcarbon AT bhimrajyadav acomparativestudyofstateoftheartdeeplearningmodelsforsemanticsegmentationofporesinscanningelectronmicroscopeimagesofactivatedcarbon AT vasantagurung acomparativestudyofstateoftheartdeeplearningmodelsforsemanticsegmentationofporesinscanningelectronmicroscopeimagesofactivatedcarbon AT mandirapradhanangaadhikari acomparativestudyofstateoftheartdeeplearningmodelsforsemanticsegmentationofporesinscanningelectronmicroscopeimagesofactivatedcarbon AT loknathregmi acomparativestudyofstateoftheartdeeplearningmodelsforsemanticsegmentationofporesinscanningelectronmicroscopeimagesofactivatedcarbon AT nandabikramadhikari acomparativestudyofstateoftheartdeeplearningmodelsforsemanticsegmentationofporesinscanningelectronmicroscopeimagesofactivatedcarbon AT bishwaspokharel comparativestudyofstateoftheartdeeplearningmodelsforsemanticsegmentationofporesinscanningelectronmicroscopeimagesofactivatedcarbon AT deepshankarpandey comparativestudyofstateoftheartdeeplearningmodelsforsemanticsegmentationofporesinscanningelectronmicroscopeimagesofactivatedcarbon AT anjulisapkota comparativestudyofstateoftheartdeeplearningmodelsforsemanticsegmentationofporesinscanningelectronmicroscopeimagesofactivatedcarbon AT bhimrajyadav comparativestudyofstateoftheartdeeplearningmodelsforsemanticsegmentationofporesinscanningelectronmicroscopeimagesofactivatedcarbon AT vasantagurung comparativestudyofstateoftheartdeeplearningmodelsforsemanticsegmentationofporesinscanningelectronmicroscopeimagesofactivatedcarbon AT mandirapradhanangaadhikari comparativestudyofstateoftheartdeeplearningmodelsforsemanticsegmentationofporesinscanningelectronmicroscopeimagesofactivatedcarbon AT loknathregmi comparativestudyofstateoftheartdeeplearningmodelsforsemanticsegmentationofporesinscanningelectronmicroscopeimagesofactivatedcarbon AT nandabikramadhikari comparativestudyofstateoftheartdeeplearningmodelsforsemanticsegmentationofporesinscanningelectronmicroscopeimagesofactivatedcarbon |