Image super-resolution reconstruction method based on regularization content mode weight prediction
The invention discloses an image super-resolution reconstruction method based on regularization content mode weight prediction. The method comprises the following steps: S1, carrying out low-resolution feature extraction on an input image through a feature extraction network; s2, performing regulari...
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 | FENG HESEN WEI GANG MA LIHONG |
| description | The invention discloses an image super-resolution reconstruction method based on regularization content mode weight prediction. The method comprises the following steps: S1, carrying out low-resolution feature extraction on an input image through a feature extraction network; s2, performing regularization content mode extraction on the low-resolution features by using a regularization content mode extraction network; s3, performing position mapping on each pixel point on the super-resolution image, and determining position scale information and a regularization content mode of each pixel point; s4, using the convolution kernel weight prediction network to generate a convolution kernel weight for each pixel point on the super-resolution image, wherein the convolution kernel weight is matched with the position scale information and the regularization content mode of the pixel point; and S5, reconstructing the low-resolution feature of the corresponding position by using the convolution kernel weight of each pix |
| format | Patent |
| fullrecord | <record><control><sourceid>epo_EVB</sourceid><recordid>TN_cdi_epo_espacenet_CN114529449A</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>CN114529449A</sourcerecordid><originalsourceid>FETCH-epo_espacenet_CN114529449A3</originalsourceid><addsrcrecordid>eNqNi70KwjAYRbM4iPoOnw_QoVqHjlIUXZzcS0yuaaD5IfmC4NMrwQdwuhzOuUuhrk4aUC4RqUnIYS5sg6cEFXzmVFRFB56CpofM0FS1KbNM9i2r_rYMz-SCBr1gzcQUE7St77VYPOWcsfntSmzPp_twaRDDiBylggePw61tu8Ou77r-uP-n-QBTIz_T</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>patent</recordtype></control><display><type>patent</type><title>Image super-resolution reconstruction method based on regularization content mode weight prediction</title><source>esp@cenet</source><creator>FENG HESEN ; WEI GANG ; MA LIHONG</creator><creatorcontrib>FENG HESEN ; WEI GANG ; MA LIHONG</creatorcontrib><description>The invention discloses an image super-resolution reconstruction method based on regularization content mode weight prediction. The method comprises the following steps: S1, carrying out low-resolution feature extraction on an input image through a feature extraction network; s2, performing regularization content mode extraction on the low-resolution features by using a regularization content mode extraction network; s3, performing position mapping on each pixel point on the super-resolution image, and determining position scale information and a regularization content mode of each pixel point; s4, using the convolution kernel weight prediction network to generate a convolution kernel weight for each pixel point on the super-resolution image, wherein the convolution kernel weight is matched with the position scale information and the regularization content mode of the pixel point; and S5, reconstructing the low-resolution feature of the corresponding position by using the convolution kernel weight of each pix</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>2022</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=20220524&DB=EPODOC&CC=CN&NR=114529449A$$EHTML$$P50$$Gepo$$Hfree_for_read</linktohtml><link.rule.ids>230,308,780,885,25564,76547</link.rule.ids><linktorsrc>$$Uhttps://worldwide.espacenet.com/publicationDetails/biblio?FT=D&date=20220524&DB=EPODOC&CC=CN&NR=114529449A$$EView_record_in_European_Patent_Office$$FView_record_in_$$GEuropean_Patent_Office$$Hfree_for_read</linktorsrc></links><search><creatorcontrib>FENG HESEN</creatorcontrib><creatorcontrib>WEI GANG</creatorcontrib><creatorcontrib>MA LIHONG</creatorcontrib><title>Image super-resolution reconstruction method based on regularization content mode weight prediction</title><description>The invention discloses an image super-resolution reconstruction method based on regularization content mode weight prediction. The method comprises the following steps: S1, carrying out low-resolution feature extraction on an input image through a feature extraction network; s2, performing regularization content mode extraction on the low-resolution features by using a regularization content mode extraction network; s3, performing position mapping on each pixel point on the super-resolution image, and determining position scale information and a regularization content mode of each pixel point; s4, using the convolution kernel weight prediction network to generate a convolution kernel weight for each pixel point on the super-resolution image, wherein the convolution kernel weight is matched with the position scale information and the regularization content mode of the pixel point; and S5, reconstructing the low-resolution feature of the corresponding position by using the convolution kernel weight of each pix</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>2022</creationdate><recordtype>patent</recordtype><sourceid>EVB</sourceid><recordid>eNqNi70KwjAYRbM4iPoOnw_QoVqHjlIUXZzcS0yuaaD5IfmC4NMrwQdwuhzOuUuhrk4aUC4RqUnIYS5sg6cEFXzmVFRFB56CpofM0FS1KbNM9i2r_rYMz-SCBr1gzcQUE7St77VYPOWcsfntSmzPp_twaRDDiBylggePw61tu8Ou77r-uP-n-QBTIz_T</recordid><startdate>20220524</startdate><enddate>20220524</enddate><creator>FENG HESEN</creator><creator>WEI GANG</creator><creator>MA LIHONG</creator><scope>EVB</scope></search><sort><creationdate>20220524</creationdate><title>Image super-resolution reconstruction method based on regularization content mode weight prediction</title><author>FENG HESEN ; WEI GANG ; MA LIHONG</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-epo_espacenet_CN114529449A3</frbrgroupid><rsrctype>patents</rsrctype><prefilter>patents</prefilter><language>chi ; eng</language><creationdate>2022</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>FENG HESEN</creatorcontrib><creatorcontrib>WEI GANG</creatorcontrib><creatorcontrib>MA LIHONG</creatorcontrib><collection>esp@cenet</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>FENG HESEN</au><au>WEI GANG</au><au>MA LIHONG</au><format>patent</format><genre>patent</genre><ristype>GEN</ristype><title>Image super-resolution reconstruction method based on regularization content mode weight prediction</title><date>2022-05-24</date><risdate>2022</risdate><abstract>The invention discloses an image super-resolution reconstruction method based on regularization content mode weight prediction. The method comprises the following steps: S1, carrying out low-resolution feature extraction on an input image through a feature extraction network; s2, performing regularization content mode extraction on the low-resolution features by using a regularization content mode extraction network; s3, performing position mapping on each pixel point on the super-resolution image, and determining position scale information and a regularization content mode of each pixel point; s4, using the convolution kernel weight prediction network to generate a convolution kernel weight for each pixel point on the super-resolution image, wherein the convolution kernel weight is matched with the position scale information and the regularization content mode of the pixel point; and S5, reconstructing the low-resolution feature of the corresponding position by using the convolution kernel weight of each pix</abstract><oa>free_for_read</oa></addata></record> |
| fulltext | fulltext_linktorsrc |
| identifier | |
| ispartof | |
| issn | |
| language | chi ; eng |
| recordid | cdi_epo_espacenet_CN114529449A |
| source | esp@cenet |
| subjects | CALCULATING COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS COMPUTING COUNTING IMAGE DATA PROCESSING OR GENERATION, IN GENERAL PHYSICS |
| title | Image super-resolution reconstruction method based on regularization content mode weight prediction |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-28T18%3A29%3A10IST&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=FENG%20HESEN&rft.date=2022-05-24&rft_id=info:doi/&rft_dat=%3Cepo_EVB%3ECN114529449A%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 |