Tungsten concentrate flotation image working condition identification method
A tungsten concentrate flotation froth image working condition recognition method comprises the steps that a froth data set is divided into five working condition states, after image preprocessing and image enhancement are conducted, a tungsten concentrate froth image data set is trained through an...
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 | HEI HONGWEI GUO AIBING LIU JUN XUE MINGXIANG ZHANG GONG WANG HANG LIU HONGLIANG NIE QI YAO SHAOXIN |
| description | A tungsten concentrate flotation froth image working condition recognition method comprises the steps that a froth data set is divided into five working condition states, after image preprocessing and image enhancement are conducted, a tungsten concentrate froth image data set is trained through an improved resnet50 network, and according to the degree of accuracy of a training model, the working condition of a froth image is recognized. And selecting the working condition recognition model with the highest accuracy as an offline model, and performing recognition and classification on the input foam image. In an improved renet50 structure, a convolution block is used for changing the number of feature map channels, and a channel attention module is added to a residual block after convolution is completed to achieve deep extraction of foam features. The method enables the training model to have a high recognition rate for the foam image, improves the discrimination capability for the working condition state, a |
| format | Patent |
| fullrecord | <record><control><sourceid>epo_EVB</sourceid><recordid>TN_cdi_epo_espacenet_CN117975096A</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>CN117975096A</sourcerecordid><originalsourceid>FETCH-epo_espacenet_CN117975096A3</originalsourceid><addsrcrecordid>eNrjZPAJKc1LLy5JzVNIzs9LTs0rKUosSVVIy8kvSSzJzM9TyMxNTE9VKM8vys7MSwepScmEiKcA1WamZSZDlOWmlmTkp_AwsKYl5hSn8kJpbgZFN9cQZw_d1IL8-NTigkSgBakl8c5-hobmluamBpZmjsbEqAEAK4Y3AQ</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>patent</recordtype></control><display><type>patent</type><title>Tungsten concentrate flotation image working condition identification method</title><source>esp@cenet</source><creator>HEI HONGWEI ; GUO AIBING ; LIU JUN ; XUE MINGXIANG ; ZHANG GONG ; WANG HANG ; LIU HONGLIANG ; NIE QI ; YAO SHAOXIN</creator><creatorcontrib>HEI HONGWEI ; GUO AIBING ; LIU JUN ; XUE MINGXIANG ; ZHANG GONG ; WANG HANG ; LIU HONGLIANG ; NIE QI ; YAO SHAOXIN</creatorcontrib><description>A tungsten concentrate flotation froth image working condition recognition method comprises the steps that a froth data set is divided into five working condition states, after image preprocessing and image enhancement are conducted, a tungsten concentrate froth image data set is trained through an improved resnet50 network, and according to the degree of accuracy of a training model, the working condition of a froth image is recognized. And selecting the working condition recognition model with the highest accuracy as an offline model, and performing recognition and classification on the input foam image. In an improved renet50 structure, a convolution block is used for changing the number of feature map channels, and a channel attention module is added to a residual block after convolution is completed to achieve deep extraction of foam features. The method enables the training model to have a high recognition rate for the foam image, improves the discrimination capability for the working condition state, a</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=20240503&DB=EPODOC&CC=CN&NR=117975096A$$EHTML$$P50$$Gepo$$Hfree_for_read</linktohtml><link.rule.ids>230,309,782,887,25573,76557</link.rule.ids><linktorsrc>$$Uhttps://worldwide.espacenet.com/publicationDetails/biblio?FT=D&date=20240503&DB=EPODOC&CC=CN&NR=117975096A$$EView_record_in_European_Patent_Office$$FView_record_in_$$GEuropean_Patent_Office$$Hfree_for_read</linktorsrc></links><search><creatorcontrib>HEI HONGWEI</creatorcontrib><creatorcontrib>GUO AIBING</creatorcontrib><creatorcontrib>LIU JUN</creatorcontrib><creatorcontrib>XUE MINGXIANG</creatorcontrib><creatorcontrib>ZHANG GONG</creatorcontrib><creatorcontrib>WANG HANG</creatorcontrib><creatorcontrib>LIU HONGLIANG</creatorcontrib><creatorcontrib>NIE QI</creatorcontrib><creatorcontrib>YAO SHAOXIN</creatorcontrib><title>Tungsten concentrate flotation image working condition identification method</title><description>A tungsten concentrate flotation froth image working condition recognition method comprises the steps that a froth data set is divided into five working condition states, after image preprocessing and image enhancement are conducted, a tungsten concentrate froth image data set is trained through an improved resnet50 network, and according to the degree of accuracy of a training model, the working condition of a froth image is recognized. And selecting the working condition recognition model with the highest accuracy as an offline model, and performing recognition and classification on the input foam image. In an improved renet50 structure, a convolution block is used for changing the number of feature map channels, and a channel attention module is added to a residual block after convolution is completed to achieve deep extraction of foam features. The method enables the training model to have a high recognition rate for the foam image, improves the discrimination capability for the working condition state, a</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>eNrjZPAJKc1LLy5JzVNIzs9LTs0rKUosSVVIy8kvSSzJzM9TyMxNTE9VKM8vys7MSwepScmEiKcA1WamZSZDlOWmlmTkp_AwsKYl5hSn8kJpbgZFN9cQZw_d1IL8-NTigkSgBakl8c5-hobmluamBpZmjsbEqAEAK4Y3AQ</recordid><startdate>20240503</startdate><enddate>20240503</enddate><creator>HEI HONGWEI</creator><creator>GUO AIBING</creator><creator>LIU JUN</creator><creator>XUE MINGXIANG</creator><creator>ZHANG GONG</creator><creator>WANG HANG</creator><creator>LIU HONGLIANG</creator><creator>NIE QI</creator><creator>YAO SHAOXIN</creator><scope>EVB</scope></search><sort><creationdate>20240503</creationdate><title>Tungsten concentrate flotation image working condition identification method</title><author>HEI HONGWEI ; GUO AIBING ; LIU JUN ; XUE MINGXIANG ; ZHANG GONG ; WANG HANG ; LIU HONGLIANG ; NIE QI ; YAO SHAOXIN</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-epo_espacenet_CN117975096A3</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>HEI HONGWEI</creatorcontrib><creatorcontrib>GUO AIBING</creatorcontrib><creatorcontrib>LIU JUN</creatorcontrib><creatorcontrib>XUE MINGXIANG</creatorcontrib><creatorcontrib>ZHANG GONG</creatorcontrib><creatorcontrib>WANG HANG</creatorcontrib><creatorcontrib>LIU HONGLIANG</creatorcontrib><creatorcontrib>NIE QI</creatorcontrib><creatorcontrib>YAO SHAOXIN</creatorcontrib><collection>esp@cenet</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>HEI HONGWEI</au><au>GUO AIBING</au><au>LIU JUN</au><au>XUE MINGXIANG</au><au>ZHANG GONG</au><au>WANG HANG</au><au>LIU HONGLIANG</au><au>NIE QI</au><au>YAO SHAOXIN</au><format>patent</format><genre>patent</genre><ristype>GEN</ristype><title>Tungsten concentrate flotation image working condition identification method</title><date>2024-05-03</date><risdate>2024</risdate><abstract>A tungsten concentrate flotation froth image working condition recognition method comprises the steps that a froth data set is divided into five working condition states, after image preprocessing and image enhancement are conducted, a tungsten concentrate froth image data set is trained through an improved resnet50 network, and according to the degree of accuracy of a training model, the working condition of a froth image is recognized. And selecting the working condition recognition model with the highest accuracy as an offline model, and performing recognition and classification on the input foam image. In an improved renet50 structure, a convolution block is used for changing the number of feature map channels, and a channel attention module is added to a residual block after convolution is completed to achieve deep extraction of foam features. The method enables the training model to have a high recognition rate for the foam image, improves the discrimination capability for the working condition state, a</abstract><oa>free_for_read</oa></addata></record> |
| fulltext | fulltext_linktorsrc |
| identifier | |
| ispartof | |
| issn | |
| language | chi ; eng |
| recordid | cdi_epo_espacenet_CN117975096A |
| source | esp@cenet |
| subjects | CALCULATING COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS COMPUTING COUNTING IMAGE DATA PROCESSING OR GENERATION, IN GENERAL PHYSICS |
| title | Tungsten concentrate flotation image working condition identification method |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-02T19%3A48%3A33IST&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=HEI%20HONGWEI&rft.date=2024-05-03&rft_id=info:doi/&rft_dat=%3Cepo_EVB%3ECN117975096A%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 |