Industrial instrument detection method based on fusion of domain adaptation and unsupervised technology
The invention relates to an industrial instrument detection method based on domain adaptation and unsupervised technology fusion. The method comprises the following steps: step 1, constructing a source domain data set composed of generated images and a target domain data set composed of real shot im...
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 | TIAN MENG HAO CHUNXIAO LIU WEI ZHANG ZHAO HAN BIN SONG GUANGJU ZHANG JUN ZENG MING LIU CHAO MA YUE ZHONG SHUTONG ZHANG LINA YU HANSHEN |
| description | The invention relates to an industrial instrument detection method based on domain adaptation and unsupervised technology fusion. The method comprises the following steps: step 1, constructing a source domain data set composed of generated images and a target domain data set composed of real shot images; 2, building a target detection network; 3, obtaining the confidence coefficient of bounding box regression and the overall output confidence coefficient; 4, training the target detection network in the step 2; step 5, simultaneously utilizing the source domain data and the target domain data to carry out domain adaptation training on the target detection network obtained in the step 4, and selecting the target domain sample and the source domain sample with the highest confidence coefficient to carry out splicing to obtain a spliced image and a pseudo label thereof; step 6, calculating the consistency loss between the spliced prediction result in the step 5 and the pseudo label; 7, calculating the total loss |
| format | Patent |
| fullrecord | <record><control><sourceid>epo_EVB</sourceid><recordid>TN_cdi_epo_espacenet_CN117911323A</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>CN117911323A</sourcerecordid><originalsourceid>FETCH-epo_espacenet_CN117911323A3</originalsourceid><addsrcrecordid>eNqNjDEKwkAQRdNYiHqH8QAWawqxlKBoY2UfxswkWdjMLNlZwdubiAew-v8_Hn9ZdDehnGz0GMDLVPLAYkBs3JhXgYGtV4InJiaYdpvTjLUF0gG9ABJGw6-LQpAl5cjjy8_-dNKLBu3e62LRYki8-eWq2F7Oj-q646g1p4gNC1td3Z07HJ0r9-Wp_Mf5AEuiQPw</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>patent</recordtype></control><display><type>patent</type><title>Industrial instrument detection method based on fusion of domain adaptation and unsupervised technology</title><source>esp@cenet</source><creator>TIAN MENG ; HAO CHUNXIAO ; LIU WEI ; ZHANG ZHAO ; HAN BIN ; SONG GUANGJU ; ZHANG JUN ; ZENG MING ; LIU CHAO ; MA YUE ; ZHONG SHUTONG ; ZHANG LINA ; YU HANSHEN</creator><creatorcontrib>TIAN MENG ; HAO CHUNXIAO ; LIU WEI ; ZHANG ZHAO ; HAN BIN ; SONG GUANGJU ; ZHANG JUN ; ZENG MING ; LIU CHAO ; MA YUE ; ZHONG SHUTONG ; ZHANG LINA ; YU HANSHEN</creatorcontrib><description>The invention relates to an industrial instrument detection method based on domain adaptation and unsupervised technology fusion. The method comprises the following steps: step 1, constructing a source domain data set composed of generated images and a target domain data set composed of real shot images; 2, building a target detection network; 3, obtaining the confidence coefficient of bounding box regression and the overall output confidence coefficient; 4, training the target detection network in the step 2; step 5, simultaneously utilizing the source domain data and the target domain data to carry out domain adaptation training on the target detection network obtained in the step 4, and selecting the target domain sample and the source domain sample with the highest confidence coefficient to carry out splicing to obtain a spliced image and a pseudo label thereof; step 6, calculating the consistency loss between the spliced prediction result in the step 5 and the pseudo label; 7, calculating the total loss</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=20240419&DB=EPODOC&CC=CN&NR=117911323A$$EHTML$$P50$$Gepo$$Hfree_for_read</linktohtml><link.rule.ids>230,308,776,881,25544,76293</link.rule.ids><linktorsrc>$$Uhttps://worldwide.espacenet.com/publicationDetails/biblio?FT=D&date=20240419&DB=EPODOC&CC=CN&NR=117911323A$$EView_record_in_European_Patent_Office$$FView_record_in_$$GEuropean_Patent_Office$$Hfree_for_read</linktorsrc></links><search><creatorcontrib>TIAN MENG</creatorcontrib><creatorcontrib>HAO CHUNXIAO</creatorcontrib><creatorcontrib>LIU WEI</creatorcontrib><creatorcontrib>ZHANG ZHAO</creatorcontrib><creatorcontrib>HAN BIN</creatorcontrib><creatorcontrib>SONG GUANGJU</creatorcontrib><creatorcontrib>ZHANG JUN</creatorcontrib><creatorcontrib>ZENG MING</creatorcontrib><creatorcontrib>LIU CHAO</creatorcontrib><creatorcontrib>MA YUE</creatorcontrib><creatorcontrib>ZHONG SHUTONG</creatorcontrib><creatorcontrib>ZHANG LINA</creatorcontrib><creatorcontrib>YU HANSHEN</creatorcontrib><title>Industrial instrument detection method based on fusion of domain adaptation and unsupervised technology</title><description>The invention relates to an industrial instrument detection method based on domain adaptation and unsupervised technology fusion. The method comprises the following steps: step 1, constructing a source domain data set composed of generated images and a target domain data set composed of real shot images; 2, building a target detection network; 3, obtaining the confidence coefficient of bounding box regression and the overall output confidence coefficient; 4, training the target detection network in the step 2; step 5, simultaneously utilizing the source domain data and the target domain data to carry out domain adaptation training on the target detection network obtained in the step 4, and selecting the target domain sample and the source domain sample with the highest confidence coefficient to carry out splicing to obtain a spliced image and a pseudo label thereof; step 6, calculating the consistency loss between the spliced prediction result in the step 5 and the pseudo label; 7, calculating the total loss</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>eNqNjDEKwkAQRdNYiHqH8QAWawqxlKBoY2UfxswkWdjMLNlZwdubiAew-v8_Hn9ZdDehnGz0GMDLVPLAYkBs3JhXgYGtV4InJiaYdpvTjLUF0gG9ABJGw6-LQpAl5cjjy8_-dNKLBu3e62LRYki8-eWq2F7Oj-q646g1p4gNC1td3Z07HJ0r9-Wp_Mf5AEuiQPw</recordid><startdate>20240419</startdate><enddate>20240419</enddate><creator>TIAN MENG</creator><creator>HAO CHUNXIAO</creator><creator>LIU WEI</creator><creator>ZHANG ZHAO</creator><creator>HAN BIN</creator><creator>SONG GUANGJU</creator><creator>ZHANG JUN</creator><creator>ZENG MING</creator><creator>LIU CHAO</creator><creator>MA YUE</creator><creator>ZHONG SHUTONG</creator><creator>ZHANG LINA</creator><creator>YU HANSHEN</creator><scope>EVB</scope></search><sort><creationdate>20240419</creationdate><title>Industrial instrument detection method based on fusion of domain adaptation and unsupervised technology</title><author>TIAN MENG ; HAO CHUNXIAO ; LIU WEI ; ZHANG ZHAO ; HAN BIN ; SONG GUANGJU ; ZHANG JUN ; ZENG MING ; LIU CHAO ; MA YUE ; ZHONG SHUTONG ; ZHANG LINA ; YU HANSHEN</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-epo_espacenet_CN117911323A3</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>TIAN MENG</creatorcontrib><creatorcontrib>HAO CHUNXIAO</creatorcontrib><creatorcontrib>LIU WEI</creatorcontrib><creatorcontrib>ZHANG ZHAO</creatorcontrib><creatorcontrib>HAN BIN</creatorcontrib><creatorcontrib>SONG GUANGJU</creatorcontrib><creatorcontrib>ZHANG JUN</creatorcontrib><creatorcontrib>ZENG MING</creatorcontrib><creatorcontrib>LIU CHAO</creatorcontrib><creatorcontrib>MA YUE</creatorcontrib><creatorcontrib>ZHONG SHUTONG</creatorcontrib><creatorcontrib>ZHANG LINA</creatorcontrib><creatorcontrib>YU HANSHEN</creatorcontrib><collection>esp@cenet</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>TIAN MENG</au><au>HAO CHUNXIAO</au><au>LIU WEI</au><au>ZHANG ZHAO</au><au>HAN BIN</au><au>SONG GUANGJU</au><au>ZHANG JUN</au><au>ZENG MING</au><au>LIU CHAO</au><au>MA YUE</au><au>ZHONG SHUTONG</au><au>ZHANG LINA</au><au>YU HANSHEN</au><format>patent</format><genre>patent</genre><ristype>GEN</ristype><title>Industrial instrument detection method based on fusion of domain adaptation and unsupervised technology</title><date>2024-04-19</date><risdate>2024</risdate><abstract>The invention relates to an industrial instrument detection method based on domain adaptation and unsupervised technology fusion. The method comprises the following steps: step 1, constructing a source domain data set composed of generated images and a target domain data set composed of real shot images; 2, building a target detection network; 3, obtaining the confidence coefficient of bounding box regression and the overall output confidence coefficient; 4, training the target detection network in the step 2; step 5, simultaneously utilizing the source domain data and the target domain data to carry out domain adaptation training on the target detection network obtained in the step 4, and selecting the target domain sample and the source domain sample with the highest confidence coefficient to carry out splicing to obtain a spliced image and a pseudo label thereof; step 6, calculating the consistency loss between the spliced prediction result in the step 5 and the pseudo label; 7, calculating the total loss</abstract><oa>free_for_read</oa></addata></record> |
| fulltext | fulltext_linktorsrc |
| identifier | |
| ispartof | |
| issn | |
| language | chi ; eng |
| recordid | cdi_epo_espacenet_CN117911323A |
| source | esp@cenet |
| subjects | CALCULATING COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS COMPUTING COUNTING IMAGE DATA PROCESSING OR GENERATION, IN GENERAL PHYSICS |
| title | Industrial instrument detection method based on fusion of domain adaptation and unsupervised technology |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-27T19%3A50%3A06IST&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=TIAN%20MENG&rft.date=2024-04-19&rft_id=info:doi/&rft_dat=%3Cepo_EVB%3ECN117911323A%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 |