IMAGE HARMONIZATION FOR DEEP LEARNING MODEL OPTIMIZATION
Techniques are described for optimizing deep learning model performance using image harmonization as a pre-processing step. According to an embodiment, a method comprises decomposing, by a system operatively coupled to a processor, an input image into sub-images. The method further comprises harmoni...
Saved in:
Main Authors | , , , , , , , , |
---|---|
Format | Patent |
Language | English French German |
Published |
08.03.2023
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Abstract | Techniques are described for optimizing deep learning model performance using image harmonization as a pre-processing step. According to an embodiment, a method comprises decomposing, by a system operatively coupled to a processor, an input image into sub-images. The method further comprises harmonizing the sub-images with corresponding reference sub-images of at least one reference image based on two or more different statistical values respectively calculated for the sub-images and the corresponding reference-sub images, resulting in transformation of the sub-images into modified sub-images images. In some implementations, the modified sub-images can be combined into a harmonized image having a more similar appearance to the at least one reference image relative to the input image. In other implementations, harmonized images and/or modified sub-images generated using these techniques can be used as ground-truth training samples for training one or more deep learning model to transform input images with appearance variations into harmonized images. |
---|---|
AbstractList | Techniques are described for optimizing deep learning model performance using image harmonization as a pre-processing step. According to an embodiment, a method comprises decomposing, by a system operatively coupled to a processor, an input image into sub-images. The method further comprises harmonizing the sub-images with corresponding reference sub-images of at least one reference image based on two or more different statistical values respectively calculated for the sub-images and the corresponding reference-sub images, resulting in transformation of the sub-images into modified sub-images images. In some implementations, the modified sub-images can be combined into a harmonized image having a more similar appearance to the at least one reference image relative to the input image. In other implementations, harmonized images and/or modified sub-images generated using these techniques can be used as ground-truth training samples for training one or more deep learning model to transform input images with appearance variations into harmonized images. |
Author | TAN, Tao TOROK, Levente Imre AVINASH, Gopal B RAO, Gireesha Chinthamani GHOSE, Soumya TEGZES, Pal RUSKO, Laszlo YOUNIS, Khaled FERENCZI, Lehel |
Author_xml | – fullname: TOROK, Levente Imre – fullname: TEGZES, Pal – fullname: YOUNIS, Khaled – fullname: TAN, Tao – fullname: FERENCZI, Lehel – fullname: AVINASH, Gopal B – fullname: GHOSE, Soumya – fullname: RUSKO, Laszlo – fullname: RAO, Gireesha Chinthamani |
BookMark | eNrjYmDJy89L5WSw8PR1dHdV8HAM8vX384xyDPH091Nw8w9ScHF1DVDwcXUM8vP0c1fw9Xdx9VHwDwjx9IUq4mFgTUvMKU7lhdLcDApuriHOHrqpBfnxqcUFicmpeakl8a4BJoYmxuYWBo6GxkQoAQAf1Cnu |
ContentType | Patent |
DBID | EVB |
DatabaseName | esp@cenet |
DatabaseTitleList | |
Database_xml | – sequence: 1 dbid: EVB name: esp@cenet url: http://worldwide.espacenet.com/singleLineSearch?locale=en_EP sourceTypes: Open Access Repository |
DeliveryMethod | fulltext_linktorsrc |
Discipline | Medicine Chemistry Sciences Physics |
DocumentTitleAlternate | BILDHARMONISIERUNG ZUR OPTIMIERUNG EINES TIEFENLERNMODELLS HARMONISATION D'IMAGE POUR UNE OPTIMISATION DE MODÈLE D'APPRENTISSAGE PROFOND |
ExternalDocumentID | EP4143780A1 |
GroupedDBID | EVB |
ID | FETCH-epo_espacenet_EP4143780A13 |
IEDL.DBID | EVB |
IngestDate | Fri Aug 09 05:00:40 EDT 2024 |
IsOpenAccess | true |
IsPeerReviewed | false |
IsScholarly | false |
Language | English French German |
LinkModel | DirectLink |
MergedId | FETCHMERGED-epo_espacenet_EP4143780A13 |
Notes | Application Number: EP20210719418 |
OpenAccessLink | https://worldwide.espacenet.com/publicationDetails/biblio?FT=D&date=20230308&DB=EPODOC&CC=EP&NR=4143780A1 |
ParticipantIDs | epo_espacenet_EP4143780A1 |
PublicationCentury | 2000 |
PublicationDate | 20230308 |
PublicationDateYYYYMMDD | 2023-03-08 |
PublicationDate_xml | – month: 03 year: 2023 text: 20230308 day: 08 |
PublicationDecade | 2020 |
PublicationYear | 2023 |
RelatedCompanies | GE Precision Healthcare LLC |
RelatedCompanies_xml | – name: GE Precision Healthcare LLC |
Score | 3.4625576 |
Snippet | Techniques are described for optimizing deep learning model performance using image harmonization as a pre-processing step. According to an embodiment, a... |
SourceID | epo |
SourceType | Open Access Repository |
SubjectTerms | CALCULATING COMPUTING COUNTING IMAGE DATA PROCESSING OR GENERATION, IN GENERAL PHYSICS |
Title | IMAGE HARMONIZATION FOR DEEP LEARNING MODEL OPTIMIZATION |
URI | https://worldwide.espacenet.com/publicationDetails/biblio?FT=D&date=20230308&DB=EPODOC&locale=&CC=EP&NR=4143780A1 |
hasFullText | 1 |
inHoldings | 1 |
isFullTextHit | |
isPrint | |
link | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwfV1La8MwDBale962bqPdCx9GbmF51z2EkSZO29E8CNkou5Q8oZe0LBn7-1O8tttluxnbGFsgffpkWQZ4KBJFzo2yFPVMoqKWG7mIFCwVEWxpqqmUpjym6_nG9EV7XuiLDqx2b2F4ndBPXhwRNSpDfW-4vd78BLEcnltZP6Yr7Fo_ubHpCFt2jP60KlHBGZssDJzAFmwbW4IfmRr6BUMqWUiUDlovui2zz17H7aOUzW9Ecc_gMMTFquYcOkXVgxN79_FaD4697X13D454gmZWY-dWCesLoDPPmjAytSK0h7M3HmQiSOaIw1hI5syK_Jk_IV7gsDkJwhjN0_ekSyAui-2piJtZ7g--ZOF-2-oVdKt1VfSBZKVSGogx-sjINaQHozLREYcTWRkmSS6XAxj8ucz1P2M3cNpKkCdY0VvoNu8fxR0ibpPec1l9AbWkfzU |
link.rule.ids | 230,309,786,891,25594,76906 |
linkProvider | European Patent Office |
linkToHtml | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwfV1LT8MwDLam8Rg3KCDGMwfUW0W7tV12mFDXZmuhL1UFTVymPqVduokV8fdxwza4wC1KoiixZH_-HMcBuC-SnpLrZSlpmUwlNddzCSlYKiHY0lTtU5rymK7n6_aL-jTTZi1YbN_C8Dqhn7w4ImpUhvpec3u9-gliWTy3cv2QLrBr-TiJR5a4YcfoT_dlKlrjEQsDKzBF08SW6EcjFf2CAZUNJEp7g6Y4b-M5vY6bRymr34gyOYb9EBer6hNoFZUAHXP78ZoAh97mvluAA56gma2xc6OE61OgjmdMGbGNCO2h88aDTATJHLEYC4nLjMh3_CnxAou5JAhjNE_fk86ATFhs2hJuZr47-JyFu233z6FdLaviAkhW9kodMUYb6rmK9GBYJhricKL0BkmSK2UXun8uc_nP2B107Nhz567jP1_BUSNNnmxFr6Fdv38UN4i-dXrL5fYF0PuCIg |
openUrl | ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Apatent&rft.title=IMAGE+HARMONIZATION+FOR+DEEP+LEARNING+MODEL+OPTIMIZATION&rft.inventor=TOROK%2C+Levente+Imre&rft.inventor=TEGZES%2C+Pal&rft.inventor=YOUNIS%2C+Khaled&rft.inventor=TAN%2C+Tao&rft.inventor=FERENCZI%2C+Lehel&rft.inventor=AVINASH%2C+Gopal+B&rft.inventor=GHOSE%2C+Soumya&rft.inventor=RUSKO%2C+Laszlo&rft.inventor=RAO%2C+Gireesha+Chinthamani&rft.date=2023-03-08&rft.externalDBID=A1&rft.externalDocID=EP4143780A1 |