Steel surface defect detection method based on improved YOLOv5
The invention discloses a steel surface defect detection method based on improved YOLOv5, and the method comprises the steps: collecting a steel surface defect data set, and carrying out the preprocessing of the data set; secondly, an original YOLOv5 model is improved, and an ECA attention mechanism...
Gespeichert in:
| Hauptverfasser: | , , |
|---|---|
| Format: | Patent |
| Sprache: | chi ; eng |
| Schlagworte: | |
| Online-Zugang: | Volltext bestellen |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| container_end_page | |
|---|---|
| container_issue | |
| container_start_page | |
| container_title | |
| container_volume | |
| creator | MAO HAOJIE JIA MINGZHENG GONG YONGWANG |
| description | The invention discloses a steel surface defect detection method based on improved YOLOv5, and the method comprises the steps: collecting a steel surface defect data set, and carrying out the preprocessing of the data set; secondly, an original YOLOv5 model is improved, and an ECA attention mechanism is introduced into a backbone network and a neck network of an original YOLOv5 network; feature fusion layers in all C3 modules in the original YOLOv5 model are replaced with an SPP layer and a cross-stage residual connection layer in a C2F module, and meanwhile, the number of original output channels is set as the number of output channels of the C2F module, so that the small target detection precision of the model is improved; adjusting and optimizing a spatial pyramid pooling structure; thirdly, training parameters are set, and the improved YOLOv5 model is trained; and finally, inputting to-be-detected steel surface defect data into the trained model, and outputting position and category information of the to-b |
| format | Patent |
| fullrecord | <record><control><sourceid>epo_EVB</sourceid><recordid>TN_cdi_epo_espacenet_CN117541553A</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>CN117541553A</sourcerecordid><originalsourceid>FETCH-epo_espacenet_CN117541553A3</originalsourceid><addsrcrecordid>eNrjZLALLklNzVEoLi1KS0xOVUhJTUtNLgFSJUAqMz9PITe1JCM_RSEpsTg1RQHIz8wtKMovA7Ij_X38y0x5GFjTEnOKU3mhNDeDoptriLOHbmpBfnxqcQHQzLzUknhnP0NDc1MTQ1NTY0djYtQAABn-L-Y</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>patent</recordtype></control><display><type>patent</type><title>Steel surface defect detection method based on improved YOLOv5</title><source>esp@cenet</source><creator>MAO HAOJIE ; JIA MINGZHENG ; GONG YONGWANG</creator><creatorcontrib>MAO HAOJIE ; JIA MINGZHENG ; GONG YONGWANG</creatorcontrib><description>The invention discloses a steel surface defect detection method based on improved YOLOv5, and the method comprises the steps: collecting a steel surface defect data set, and carrying out the preprocessing of the data set; secondly, an original YOLOv5 model is improved, and an ECA attention mechanism is introduced into a backbone network and a neck network of an original YOLOv5 network; feature fusion layers in all C3 modules in the original YOLOv5 model are replaced with an SPP layer and a cross-stage residual connection layer in a C2F module, and meanwhile, the number of original output channels is set as the number of output channels of the C2F module, so that the small target detection precision of the model is improved; adjusting and optimizing a spatial pyramid pooling structure; thirdly, training parameters are set, and the improved YOLOv5 model is trained; and finally, inputting to-be-detected steel surface defect data into the trained model, and outputting position and category information of the to-b</description><language>chi ; eng</language><subject>CALCULATING ; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS ; COMPUTING ; COUNTING ; IMAGE DATA PROCESSING OR GENERATION, IN GENERAL ; PHYSICS</subject><creationdate>2024</creationdate><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://worldwide.espacenet.com/publicationDetails/biblio?FT=D&date=20240209&DB=EPODOC&CC=CN&NR=117541553A$$EHTML$$P50$$Gepo$$Hfree_for_read</linktohtml><link.rule.ids>230,308,780,885,25564,76547</link.rule.ids><linktorsrc>$$Uhttps://worldwide.espacenet.com/publicationDetails/biblio?FT=D&date=20240209&DB=EPODOC&CC=CN&NR=117541553A$$EView_record_in_European_Patent_Office$$FView_record_in_$$GEuropean_Patent_Office$$Hfree_for_read</linktorsrc></links><search><creatorcontrib>MAO HAOJIE</creatorcontrib><creatorcontrib>JIA MINGZHENG</creatorcontrib><creatorcontrib>GONG YONGWANG</creatorcontrib><title>Steel surface defect detection method based on improved YOLOv5</title><description>The invention discloses a steel surface defect detection method based on improved YOLOv5, and the method comprises the steps: collecting a steel surface defect data set, and carrying out the preprocessing of the data set; secondly, an original YOLOv5 model is improved, and an ECA attention mechanism is introduced into a backbone network and a neck network of an original YOLOv5 network; feature fusion layers in all C3 modules in the original YOLOv5 model are replaced with an SPP layer and a cross-stage residual connection layer in a C2F module, and meanwhile, the number of original output channels is set as the number of output channels of the C2F module, so that the small target detection precision of the model is improved; adjusting and optimizing a spatial pyramid pooling structure; thirdly, training parameters are set, and the improved YOLOv5 model is trained; and finally, inputting to-be-detected steel surface defect data into the trained model, and outputting position and category information of the to-b</description><subject>CALCULATING</subject><subject>COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS</subject><subject>COMPUTING</subject><subject>COUNTING</subject><subject>IMAGE DATA PROCESSING OR GENERATION, IN GENERAL</subject><subject>PHYSICS</subject><fulltext>true</fulltext><rsrctype>patent</rsrctype><creationdate>2024</creationdate><recordtype>patent</recordtype><sourceid>EVB</sourceid><recordid>eNrjZLALLklNzVEoLi1KS0xOVUhJTUtNLgFSJUAqMz9PITe1JCM_RSEpsTg1RQHIz8wtKMovA7Ij_X38y0x5GFjTEnOKU3mhNDeDoptriLOHbmpBfnxqcQHQzLzUknhnP0NDc1MTQ1NTY0djYtQAABn-L-Y</recordid><startdate>20240209</startdate><enddate>20240209</enddate><creator>MAO HAOJIE</creator><creator>JIA MINGZHENG</creator><creator>GONG YONGWANG</creator><scope>EVB</scope></search><sort><creationdate>20240209</creationdate><title>Steel surface defect detection method based on improved YOLOv5</title><author>MAO HAOJIE ; JIA MINGZHENG ; GONG YONGWANG</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-epo_espacenet_CN117541553A3</frbrgroupid><rsrctype>patents</rsrctype><prefilter>patents</prefilter><language>chi ; eng</language><creationdate>2024</creationdate><topic>CALCULATING</topic><topic>COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS</topic><topic>COMPUTING</topic><topic>COUNTING</topic><topic>IMAGE DATA PROCESSING OR GENERATION, IN GENERAL</topic><topic>PHYSICS</topic><toplevel>online_resources</toplevel><creatorcontrib>MAO HAOJIE</creatorcontrib><creatorcontrib>JIA MINGZHENG</creatorcontrib><creatorcontrib>GONG YONGWANG</creatorcontrib><collection>esp@cenet</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>MAO HAOJIE</au><au>JIA MINGZHENG</au><au>GONG YONGWANG</au><format>patent</format><genre>patent</genre><ristype>GEN</ristype><title>Steel surface defect detection method based on improved YOLOv5</title><date>2024-02-09</date><risdate>2024</risdate><abstract>The invention discloses a steel surface defect detection method based on improved YOLOv5, and the method comprises the steps: collecting a steel surface defect data set, and carrying out the preprocessing of the data set; secondly, an original YOLOv5 model is improved, and an ECA attention mechanism is introduced into a backbone network and a neck network of an original YOLOv5 network; feature fusion layers in all C3 modules in the original YOLOv5 model are replaced with an SPP layer and a cross-stage residual connection layer in a C2F module, and meanwhile, the number of original output channels is set as the number of output channels of the C2F module, so that the small target detection precision of the model is improved; adjusting and optimizing a spatial pyramid pooling structure; thirdly, training parameters are set, and the improved YOLOv5 model is trained; and finally, inputting to-be-detected steel surface defect data into the trained model, and outputting position and category information of the to-b</abstract><oa>free_for_read</oa></addata></record> |
| fulltext | fulltext_linktorsrc |
| identifier | |
| ispartof | |
| issn | |
| language | chi ; eng |
| recordid | cdi_epo_espacenet_CN117541553A |
| source | esp@cenet |
| subjects | CALCULATING COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS COMPUTING COUNTING IMAGE DATA PROCESSING OR GENERATION, IN GENERAL PHYSICS |
| title | Steel surface defect detection method based on improved YOLOv5 |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-22T02%3A42%3A27IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-epo_EVB&rft_val_fmt=info:ofi/fmt:kev:mtx:patent&rft.genre=patent&rft.au=MAO%20HAOJIE&rft.date=2024-02-09&rft_id=info:doi/&rft_dat=%3Cepo_EVB%3ECN117541553A%3C/epo_EVB%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true |