Automatic optical inspection platform for real-time surface defects detection on plane optical components based on semantic segmentation
The tendency to increase the accuracy and quality of optical parts inspection can be observed all over the world. The imperfection of manufacturing techniques can cause different defects on the optical component surface, making surface defects inspection a crucial part of the manufacturing of optica...
Gespeichert in:
| Veröffentlicht in: | Applied optics (2004) 2021-07, Vol.60 (19), p.5496-5506 |
|---|---|
| Hauptverfasser: | , , , |
| Format: | Artikel |
| Sprache: | eng |
| Schlagworte: | |
| Online-Zugang: | Volltext |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| container_end_page | 5506 |
|---|---|
| container_issue | 19 |
| container_start_page | 5496 |
| container_title | Applied optics (2004) |
| container_volume | 60 |
| creator | Karangwa, Jules Kong, Linghua Yi, Dingrong Zheng, Jishi |
| description | The tendency to increase the accuracy and quality of optical parts inspection can be observed all over the world. The imperfection of manufacturing techniques can cause different defects on the optical component surface, making surface defects inspection a crucial part of the manufacturing of optical components. Currently, the inspection of lenses, filters, mirrors, and other optical components is performed by human inspectors. However, human-based inspections are time-consuming, subjective, and incompatible with a highly efficient high-quality digital workflow. Moreover, they cannot meet the complex criteria of ISO 10110-7 for the quality pass and fail optical element samples. To meet the high demand for high-quality products, intelligent visual inspection systems are being used in many manufacturing processes. Automated surface imperfection detection based on machine learning has become a fascinating and promising area of research, with a great direct impact on different visual inspection applications. In this paper, an optical inspection platform combining parallel deep learning-based image-processing approaches with a high-resolution optomechanical module was developed to detect surface defects on optical plane components. The system involves the mechanical modules, the illumination and imaging modules, and the machine vision algorithm. Dark-field images were acquired using a 2448 x 2048-pixel line-scanning CMOS camera with 3.45 mu m per-pixel resolution. Convolutional neural networks and semantic segmentation were used for a machine vision algorithm to detect and classify defects on captured images of optical bandpass filters. The experimental results on different bandpass filter samples have shown the best performance compared to traditional methods by reaching an impressive detection speed of 0.07 s per image and an overall detection pixel accuracy of 0.923. (C) 2021 Optical Society of America |
| doi_str_mv | 10.1364/AO.424547 |
| format | Article |
| fullrecord | <record><control><sourceid>proquest_webof</sourceid><recordid>TN_cdi_proquest_miscellaneous_2552056178</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2552295001</sourcerecordid><originalsourceid>FETCH-LOGICAL-c220t-d8cf8f754777403652ec6bfae4f05c5fa6d1f34c3e0a57e2bc278664ba95a66d3</originalsourceid><addsrcrecordid>eNqN0c1q3DAQAGARWrKbn0PfwNBLQ3Cif8vHZWmSQmAvCeRmZHlUHGzJlWRK36CPHTkOe-ipF41A34xGGoS-EHxDmOS3u8MNp1zw6gRtKRGiZESKT2ibt3VJqHrZoLMYXzFmgtfVKdowTiVTTG7R392c_KhTbwo_5VUPRe_iBCb13hXToJP1YSzyUgTQQ5n6EYo4B6sNFB3YDGOO6SNhzXFwLGb8OHkHLqtWR-gWEWHUbrkxws8xH-kl9QJ9tnqIcPkRz9Hz3fen_UP5eLj_sd89loZSnMpOGatsld9aVRwzKSgY2VoN3GJhhNWyI5ZxwwBrUQFtDa2UlLzVtdBSduwcfVvrTsH_miGmZuyjgWHp2s-xoUJQLCSpVKZf_6Gvfg4ud_euaC0wJlldrcoEH2MA20yhH3X40xDcLONpdodmHU-216v9Da230fTgDBw9xlhKVSuZ62JSZ63-X-_79R_3fnaJvQE0q6PT</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2552295001</pqid></control><display><type>article</type><title>Automatic optical inspection platform for real-time surface defects detection on plane optical components based on semantic segmentation</title><source>Web of Science - Science Citation Index Expanded - 2021<img src="https://exlibris-pub.s3.amazonaws.com/fromwos-v2.jpg" /></source><source>Alma/SFX Local Collection</source><source>Optica Publishing Group Journals</source><creator>Karangwa, Jules ; Kong, Linghua ; Yi, Dingrong ; Zheng, Jishi</creator><creatorcontrib>Karangwa, Jules ; Kong, Linghua ; Yi, Dingrong ; Zheng, Jishi</creatorcontrib><description>The tendency to increase the accuracy and quality of optical parts inspection can be observed all over the world. The imperfection of manufacturing techniques can cause different defects on the optical component surface, making surface defects inspection a crucial part of the manufacturing of optical components. Currently, the inspection of lenses, filters, mirrors, and other optical components is performed by human inspectors. However, human-based inspections are time-consuming, subjective, and incompatible with a highly efficient high-quality digital workflow. Moreover, they cannot meet the complex criteria of ISO 10110-7 for the quality pass and fail optical element samples. To meet the high demand for high-quality products, intelligent visual inspection systems are being used in many manufacturing processes. Automated surface imperfection detection based on machine learning has become a fascinating and promising area of research, with a great direct impact on different visual inspection applications. In this paper, an optical inspection platform combining parallel deep learning-based image-processing approaches with a high-resolution optomechanical module was developed to detect surface defects on optical plane components. The system involves the mechanical modules, the illumination and imaging modules, and the machine vision algorithm. Dark-field images were acquired using a 2448 x 2048-pixel line-scanning CMOS camera with 3.45 mu m per-pixel resolution. Convolutional neural networks and semantic segmentation were used for a machine vision algorithm to detect and classify defects on captured images of optical bandpass filters. The experimental results on different bandpass filter samples have shown the best performance compared to traditional methods by reaching an impressive detection speed of 0.07 s per image and an overall detection pixel accuracy of 0.923. (C) 2021 Optical Society of America</description><identifier>ISSN: 1559-128X</identifier><identifier>EISSN: 2155-3165</identifier><identifier>EISSN: 1539-4522</identifier><identifier>DOI: 10.1364/AO.424547</identifier><identifier>PMID: 34263836</identifier><language>eng</language><publisher>WASHINGTON: Optica Publishing Group</publisher><subject>Algorithms ; Artificial neural networks ; Bandpass filters ; CMOS ; Deep learning ; Human performance ; Image acquisition ; Image classification ; Image filters ; Image processing ; Image segmentation ; Inspection ; Machine learning ; Machine vision ; Manufacturing ; Modules ; Optical components ; Optics ; Physical Sciences ; Pixels ; Science & Technology ; Semantic segmentation ; Semantics ; Surface defects ; Vision systems ; Workflow</subject><ispartof>Applied optics (2004), 2021-07, Vol.60 (19), p.5496-5506</ispartof><rights>Copyright Optical Society of America Jul 1, 2021</rights><lds50>peer_reviewed</lds50><woscitedreferencessubscribed>true</woscitedreferencessubscribed><woscitedreferencescount>21</woscitedreferencescount><woscitedreferencesoriginalsourcerecordid>wos000668986500019</woscitedreferencesoriginalsourcerecordid><citedby>FETCH-LOGICAL-c220t-d8cf8f754777403652ec6bfae4f05c5fa6d1f34c3e0a57e2bc278664ba95a66d3</citedby><cites>FETCH-LOGICAL-c220t-d8cf8f754777403652ec6bfae4f05c5fa6d1f34c3e0a57e2bc278664ba95a66d3</cites><orcidid>0000-0002-6993-6857 ; 0000-0002-7139-6431</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>315,781,785,3259,27929,27930,39263</link.rule.ids></links><search><creatorcontrib>Karangwa, Jules</creatorcontrib><creatorcontrib>Kong, Linghua</creatorcontrib><creatorcontrib>Yi, Dingrong</creatorcontrib><creatorcontrib>Zheng, Jishi</creatorcontrib><title>Automatic optical inspection platform for real-time surface defects detection on plane optical components based on semantic segmentation</title><title>Applied optics (2004)</title><addtitle>APPL OPTICS</addtitle><description>The tendency to increase the accuracy and quality of optical parts inspection can be observed all over the world. The imperfection of manufacturing techniques can cause different defects on the optical component surface, making surface defects inspection a crucial part of the manufacturing of optical components. Currently, the inspection of lenses, filters, mirrors, and other optical components is performed by human inspectors. However, human-based inspections are time-consuming, subjective, and incompatible with a highly efficient high-quality digital workflow. Moreover, they cannot meet the complex criteria of ISO 10110-7 for the quality pass and fail optical element samples. To meet the high demand for high-quality products, intelligent visual inspection systems are being used in many manufacturing processes. Automated surface imperfection detection based on machine learning has become a fascinating and promising area of research, with a great direct impact on different visual inspection applications. In this paper, an optical inspection platform combining parallel deep learning-based image-processing approaches with a high-resolution optomechanical module was developed to detect surface defects on optical plane components. The system involves the mechanical modules, the illumination and imaging modules, and the machine vision algorithm. Dark-field images were acquired using a 2448 x 2048-pixel line-scanning CMOS camera with 3.45 mu m per-pixel resolution. Convolutional neural networks and semantic segmentation were used for a machine vision algorithm to detect and classify defects on captured images of optical bandpass filters. The experimental results on different bandpass filter samples have shown the best performance compared to traditional methods by reaching an impressive detection speed of 0.07 s per image and an overall detection pixel accuracy of 0.923. (C) 2021 Optical Society of America</description><subject>Algorithms</subject><subject>Artificial neural networks</subject><subject>Bandpass filters</subject><subject>CMOS</subject><subject>Deep learning</subject><subject>Human performance</subject><subject>Image acquisition</subject><subject>Image classification</subject><subject>Image filters</subject><subject>Image processing</subject><subject>Image segmentation</subject><subject>Inspection</subject><subject>Machine learning</subject><subject>Machine vision</subject><subject>Manufacturing</subject><subject>Modules</subject><subject>Optical components</subject><subject>Optics</subject><subject>Physical Sciences</subject><subject>Pixels</subject><subject>Science & Technology</subject><subject>Semantic segmentation</subject><subject>Semantics</subject><subject>Surface defects</subject><subject>Vision systems</subject><subject>Workflow</subject><issn>1559-128X</issn><issn>2155-3165</issn><issn>1539-4522</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2021</creationdate><recordtype>article</recordtype><sourceid>HGBXW</sourceid><recordid>eNqN0c1q3DAQAGARWrKbn0PfwNBLQ3Cif8vHZWmSQmAvCeRmZHlUHGzJlWRK36CPHTkOe-ipF41A34xGGoS-EHxDmOS3u8MNp1zw6gRtKRGiZESKT2ibt3VJqHrZoLMYXzFmgtfVKdowTiVTTG7R392c_KhTbwo_5VUPRe_iBCb13hXToJP1YSzyUgTQQ5n6EYo4B6sNFB3YDGOO6SNhzXFwLGb8OHkHLqtWR-gWEWHUbrkxws8xH-kl9QJ9tnqIcPkRz9Hz3fen_UP5eLj_sd89loZSnMpOGatsld9aVRwzKSgY2VoN3GJhhNWyI5ZxwwBrUQFtDa2UlLzVtdBSduwcfVvrTsH_miGmZuyjgWHp2s-xoUJQLCSpVKZf_6Gvfg4ud_euaC0wJlldrcoEH2MA20yhH3X40xDcLONpdodmHU-216v9Da230fTgDBw9xlhKVSuZ62JSZ63-X-_79R_3fnaJvQE0q6PT</recordid><startdate>20210701</startdate><enddate>20210701</enddate><creator>Karangwa, Jules</creator><creator>Kong, Linghua</creator><creator>Yi, Dingrong</creator><creator>Zheng, Jishi</creator><general>Optica Publishing Group</general><general>Optical Society of America</general><scope>BLEPL</scope><scope>DTL</scope><scope>HGBXW</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>7SP</scope><scope>7U5</scope><scope>8FD</scope><scope>H8D</scope><scope>L7M</scope><scope>7X8</scope><orcidid>https://orcid.org/0000-0002-6993-6857</orcidid><orcidid>https://orcid.org/0000-0002-7139-6431</orcidid></search><sort><creationdate>20210701</creationdate><title>Automatic optical inspection platform for real-time surface defects detection on plane optical components based on semantic segmentation</title><author>Karangwa, Jules ; Kong, Linghua ; Yi, Dingrong ; Zheng, Jishi</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c220t-d8cf8f754777403652ec6bfae4f05c5fa6d1f34c3e0a57e2bc278664ba95a66d3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2021</creationdate><topic>Algorithms</topic><topic>Artificial neural networks</topic><topic>Bandpass filters</topic><topic>CMOS</topic><topic>Deep learning</topic><topic>Human performance</topic><topic>Image acquisition</topic><topic>Image classification</topic><topic>Image filters</topic><topic>Image processing</topic><topic>Image segmentation</topic><topic>Inspection</topic><topic>Machine learning</topic><topic>Machine vision</topic><topic>Manufacturing</topic><topic>Modules</topic><topic>Optical components</topic><topic>Optics</topic><topic>Physical Sciences</topic><topic>Pixels</topic><topic>Science & Technology</topic><topic>Semantic segmentation</topic><topic>Semantics</topic><topic>Surface defects</topic><topic>Vision systems</topic><topic>Workflow</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Karangwa, Jules</creatorcontrib><creatorcontrib>Kong, Linghua</creatorcontrib><creatorcontrib>Yi, Dingrong</creatorcontrib><creatorcontrib>Zheng, Jishi</creatorcontrib><collection>Web of Science Core Collection</collection><collection>Science Citation Index Expanded</collection><collection>Web of Science - Science Citation Index Expanded - 2021</collection><collection>CrossRef</collection><collection>Electronics & Communications Abstracts</collection><collection>Solid State and Superconductivity Abstracts</collection><collection>Technology Research Database</collection><collection>Aerospace Database</collection><collection>Advanced Technologies Database with Aerospace</collection><collection>MEDLINE - Academic</collection><jtitle>Applied optics (2004)</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Karangwa, Jules</au><au>Kong, Linghua</au><au>Yi, Dingrong</au><au>Zheng, Jishi</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Automatic optical inspection platform for real-time surface defects detection on plane optical components based on semantic segmentation</atitle><jtitle>Applied optics (2004)</jtitle><stitle>APPL OPTICS</stitle><date>2021-07-01</date><risdate>2021</risdate><volume>60</volume><issue>19</issue><spage>5496</spage><epage>5506</epage><pages>5496-5506</pages><issn>1559-128X</issn><eissn>2155-3165</eissn><eissn>1539-4522</eissn><abstract>The tendency to increase the accuracy and quality of optical parts inspection can be observed all over the world. The imperfection of manufacturing techniques can cause different defects on the optical component surface, making surface defects inspection a crucial part of the manufacturing of optical components. Currently, the inspection of lenses, filters, mirrors, and other optical components is performed by human inspectors. However, human-based inspections are time-consuming, subjective, and incompatible with a highly efficient high-quality digital workflow. Moreover, they cannot meet the complex criteria of ISO 10110-7 for the quality pass and fail optical element samples. To meet the high demand for high-quality products, intelligent visual inspection systems are being used in many manufacturing processes. Automated surface imperfection detection based on machine learning has become a fascinating and promising area of research, with a great direct impact on different visual inspection applications. In this paper, an optical inspection platform combining parallel deep learning-based image-processing approaches with a high-resolution optomechanical module was developed to detect surface defects on optical plane components. The system involves the mechanical modules, the illumination and imaging modules, and the machine vision algorithm. Dark-field images were acquired using a 2448 x 2048-pixel line-scanning CMOS camera with 3.45 mu m per-pixel resolution. Convolutional neural networks and semantic segmentation were used for a machine vision algorithm to detect and classify defects on captured images of optical bandpass filters. The experimental results on different bandpass filter samples have shown the best performance compared to traditional methods by reaching an impressive detection speed of 0.07 s per image and an overall detection pixel accuracy of 0.923. (C) 2021 Optical Society of America</abstract><cop>WASHINGTON</cop><pub>Optica Publishing Group</pub><pmid>34263836</pmid><doi>10.1364/AO.424547</doi><tpages>11</tpages><orcidid>https://orcid.org/0000-0002-6993-6857</orcidid><orcidid>https://orcid.org/0000-0002-7139-6431</orcidid></addata></record> |
| fulltext | fulltext |
| identifier | ISSN: 1559-128X |
| ispartof | Applied optics (2004), 2021-07, Vol.60 (19), p.5496-5506 |
| issn | 1559-128X 2155-3165 1539-4522 |
| language | eng |
| recordid | cdi_proquest_miscellaneous_2552056178 |
| source | Web of Science - Science Citation Index Expanded - 2021<img src="https://exlibris-pub.s3.amazonaws.com/fromwos-v2.jpg" />; Alma/SFX Local Collection; Optica Publishing Group Journals |
| subjects | Algorithms Artificial neural networks Bandpass filters CMOS Deep learning Human performance Image acquisition Image classification Image filters Image processing Image segmentation Inspection Machine learning Machine vision Manufacturing Modules Optical components Optics Physical Sciences Pixels Science & Technology Semantic segmentation Semantics Surface defects Vision systems Workflow |
| title | Automatic optical inspection platform for real-time surface defects detection on plane optical components based on semantic segmentation |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-13T14%3A46%3A52IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_webof&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Automatic%20optical%20inspection%20platform%20for%20real-time%20surface%20defects%20detection%20on%20plane%20optical%20components%20based%20on%20semantic%20segmentation&rft.jtitle=Applied%20optics%20(2004)&rft.au=Karangwa,%20Jules&rft.date=2021-07-01&rft.volume=60&rft.issue=19&rft.spage=5496&rft.epage=5506&rft.pages=5496-5506&rft.issn=1559-128X&rft.eissn=2155-3165&rft_id=info:doi/10.1364/AO.424547&rft_dat=%3Cproquest_webof%3E2552295001%3C/proquest_webof%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2552295001&rft_id=info:pmid/34263836&rfr_iscdi=true |