Pattern Retrieval in Large Image Databases Using Multiscale Coarse-to-Fine Cascaded Active Learning
Pattern retrieval is a fundamental challenge in machine learning but is often subject to the problem of gathering enough labeled examples of the target pattern, and also to the computational complexity inherent to the training and the evaluation of complex classifier functions on large databases. In...
Uložené v:
| Vydané v: | IEEE journal of selected topics in applied earth observations and remote sensing Ročník 7; číslo 4; s. 1127 - 1141 |
|---|---|
| Hlavní autori: | , , , |
| Médium: | Journal Article |
| Jazyk: | English |
| Vydavateľské údaje: |
Piscataway
IEEE
01.04.2014
The Institute of Electrical and Electronics Engineers, Inc. (IEEE) |
| Predmet: | |
| ISSN: | 1939-1404, 2151-1535 |
| On-line prístup: | Získať plný text |
| Tagy: |
Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
|
| Abstract | Pattern retrieval is a fundamental challenge in machine learning but is often subject to the problem of gathering enough labeled examples of the target pattern, and also to the computational complexity inherent to the training and the evaluation of complex classifier functions on large databases. In this paper, we propose a hierarchical top-down processing scheme for pattern retrieval in high-volume high-resolution optical satellite image repositories. We learn via a multistage active learning process a cascade of classifiers working each at a certain scale on a patch-based representation of images. At each stage of the hierarchy, we seek to eliminate large parts of images considered as nonrelevant, the purpose being to set the focus at the finest scales on more promising and as spatially limited as possible areas. Our scheme is based on the fact that by reducing the size of the analysis window (i.e., the size of the patch), we better capture the properties of the targeted object. The cascaded hierarchy is introduced to compensate for the extra computational burden incurred by diminishing the size of the patch, which causes an explosion of the number of patches to process. Unlike most other retrieval methods, which require large training sets and costly offline training, we propose a cascaded active learning strategy to build a classifier at each level of the hierarchy, and we provide a new Multiple Instance Learning algorithm to propagate automatically the training examples from one level of the hierarchy to the other. Two study cases are performed for validation. The first is a test on a database of 61-cm resolution QuickBird panchromatic images and the second is an example of temporal pattern retrieval from a database of Synthetic Aperture Radar (SAR) image time series. These tests show that our method achieves a reduction in the number of computations of two orders of magnitude, while keeping the same accuracy level as recent state-of-the-art methods. |
|---|---|
| AbstractList | Pattern retrieval is a fundamental challenge in machine learning but is often subject to the problem of gathering enough labeled examples of the target pattern, and also to the computational complexity inherent to the training and the evaluation of complex classifier functions on large databases. In this paper, we propose a hierarchical top-down processing scheme for pattern retrieval in high-volume high-resolution optical satellite image repositories. We learn via a multistage active learning process a cascade of classifiers working each at a certain scale on a patch-based representation of images. At each stage of the hierarchy, we seek to eliminate large parts of images considered as nonrelevant, the purpose being to set the focus at the finest scales on more promising and as spatially limited as possible areas. Our scheme is based on the fact that by reducing the size of the analysis window (i.e., the size of the patch), we better capture the properties of the targeted object. The cascaded hierarchy is introduced to compensate for the extra computational burden incurred by diminishing the size of the patch, which causes an explosion of the number of patches to process. Unlike most other retrieval methods, which require large training sets and costly offline training, we propose a cascaded active learning strategy to build a classifier at each level of the hierarchy, and we provide a new Multiple Instance Learning algorithm to propagate automatically the training examples from one level of the hierarchy to the other. Two study cases are performed for validation. The first is a test on a database of 61-cm resolution QuickBird panchromatic images and the second is an example of temporal pattern retrieval from a database of Synthetic Aperture Radar (SAR) image time series. These tests show that our method achieves a reduction in the number of computations of two orders of magnitude, while keeping the same accuracy level as recent state-of-the-art methods. Pattern retrieval is a fundamental challenge in machine learning but is often subject to the problem of gathering enough labeled examples of the target pattern, and also to the computational complexity inherent to the training and the evaluation of complex classifier functions on large databases. In this paper, we propose a hierarchical top-down processing scheme for pattern retrieval in high-volume high-resolution optical satellite image repositories. We learn via a multistage active learning process a cascade of classifiers working each at a certain scale on a patch-based representation of images. At each stage of the hierarchy we seek to eliminate large parts of images considered as non-relevant, the purpose being to set the focus at the finest scales on more promising and as spatially limited as possible areas. Our scheme is based on the fact that by reducing the size of the analysis window (i.e. the size of thepatch), we better capture the properties of the targeted object. The cascaded hierarchy is introduced to compensate for the extra computational burden incurred by diminishing the size of the patch, which causes an explosion of the number of patches to process. Unlike most other retrieval methods, which require large training sets and costlyoffline training, we propose a cascaded active learning strategy to build a classifier at each level of the hierarchy and we provide a new Multiple Instance Learning algorithm to propagate automatically the training examples from one level of the hierarchy to the other. Two study cases are performed for validation. The first is a test on a database of 61cm resolution QuickBird panchromatic images and the second is an example of temporal pattern retrieval from a database of SAR image time series. These tests show that our method achieves a reduction of the number of computations of two orders of magnitude, while keeping the same accuracy level as recent state of the art methods. Pattern retrieval is a fundamental challenge in machine learning but is often subject to the problem of gathering enough labeled examples of the target pattern, and also to the computational complexity inherent to the training and the evaluation of complex classifier functions on large databases. In this paper, we propose a hierarchical top-down processing scheme for pattern retrieval in high-volume high-resolution optical satellite image repositories. We learn via a multistage active learning process a cascade of classifiers working each at a certain scale on a patch-based representation of images. At each stage of the hierarchy, we seek to eliminate large parts of images considered as nonrelevant, the purpose being to set the focus at the finest scales on more promising and as spatially limited as possible areas. Our scheme is based on the fact that by reducing the size of the analysis window (i.e., the size of the patch), we better capture the properties of the targeted object. The cascaded hierarchy is introduced to compensate for the extra computational burden incurred by diminishing the size of the patch, which causes an explosion of the number of patches to process. Unlike most other retrieval methods, which require large training sets and costly offline training, we propose a cascaded active learning strategy to build a classifier at each level of the hierarchy, and we provide a new Multiple Instance Learning algorithm to propagate automatically the training examples from one level of the hierarchy to the other. Two study cases are performed for validation. The first is a test on a database of 61-cm resolution QuickBird panchromatic images and the second is an example of temporal pattern retrieval from a database of Synthetic Aperture Radar (SAR) image time series. These tests show that our method achieves a reduction in the number of computations of two orders of magnitude, while keeping the same accuracy level as recent state-of-the-art methods. [PUBLICATION ABSTRACT] |
| Author | Cui, Shiyong Ferecatu, Marin Blanchart, Pierre Datcu, Mihai |
| Author_xml | – sequence: 1 givenname: Pierre surname: Blanchart fullname: Blanchart, Pierre email: pierre.blanchart@gmail.com organization: Télécom ParisTech, Paris, France – sequence: 2 givenname: Marin surname: Ferecatu fullname: Ferecatu, Marin email: Marin.Ferecatu@cnam.fr organization: Laboratoire CEDRIC, Conservatoire National des Arts et Métiers (CNAM), Paris, France – sequence: 3 givenname: Shiyong surname: Cui fullname: Cui, Shiyong email: shiyong.cui@dlr.de organization: Photogrammetry and Image Analysis, Remote Sensing Technology Institute (IMF), German Aerospace Center (DLR), Munich, Germany – sequence: 4 givenname: Mihai surname: Datcu fullname: Datcu, Mihai email: mihai.datcu@dlr.de organization: Télécom ParisTech, Paris, France |
| BackLink | https://hal.science/hal-01126389$$DView record in HAL |
| BookMark | eNqFkE1r3DAQhkVJoZu0vyAXQU89eKvRh20dl23TpLi05OMsJvI4VXDkVNIu5N_Xi0MOueQyw7y8z8zwHrOjOEVi7BTEGkDYrz-vrjeXV2spQK-lElIp9Y6tJBiowChzxFZgla1AC_2BHed8L0QtG6tWzP_BUihFfkklBdrjyEPkHaY74hcPONdvWPAWM2V-k0O84792YwnZ40h8O2HKVJWpOgtxHnGWe-r5xpewJ94RpjgjH9n7AcdMn577Cbs5-369Pa-63z8utpuu8kpBqSw1ttE1CGiMJzA4aG_bXim8bQbqG2j7Ggc7SNC6JZBD23thBQhBg2jqVp2wL8vevzi6xxQeMD25CYM733TuoAkAWavW7mH2fl68j2n6t6Nc3P20S3F-z4HR1mgpzMGlFpdPU86Jhpe1INwhebck7w7Ju-fkZ8q-onwoWMIUS8IwvsGeLmwgopdrdaNqYRr1H2jukgU |
| CODEN | IJSTHZ |
| CitedBy_id | crossref_primary_10_1080_22797254_2017_1348914 crossref_primary_10_1109_MGRS_2016_2548504 crossref_primary_10_1109_JSTARS_2018_2795753 crossref_primary_10_1109_MGRS_2024_3403423 crossref_primary_10_1109_JSTARS_2014_2311915 crossref_primary_10_1109_JSTARS_2015_2461556 crossref_primary_10_1109_TGRS_2015_2438400 crossref_primary_10_1109_JSTARS_2015_2420582 crossref_primary_10_3390_ijgi5100174 crossref_primary_10_1007_s12524_019_01049_8 crossref_primary_10_1109_JSTARS_2016_2547843 crossref_primary_10_1109_JSTARS_2016_2549557 crossref_primary_10_1109_LGRS_2016_2590739 crossref_primary_10_1016_j_asoc_2022_109107 crossref_primary_10_1109_JSTARS_2016_2537548 |
| Cites_doi | 10.1109/TGRS.2006.890579 10.1109/LGRS.2013.2246539 10.1109/TPAMI.2006.248 10.1145/1126004.1126005 10.1109/ICPR.2010.799 10.1109/ICCV.2001.937694 10.1109/TPAMI.2009.155 10.1109/TPAMI.2005.188 10.1023/A:1011113216584 10.1109/TGRS.2008.2010404 10.1007/s00530-002-0070-3 10.1016/S0893-6080(01)00104-6 10.1109/TGRS.2007.892007 10.1109/CVPR.2010.5540115 10.1109/34.895972 10.1109/TMM.2005.858383 10.1109/TGRS.2010.2072929 10.1109/TGRS.2010.2083673 10.1109/ICIP.2002.1039124 10.1023/B:VISI.0000013087.49260.fb 10.1007/s00530-007-0094-9 10.1109/TIP.2008.924286 10.1145/500156.500159 10.1109/TIP.2003.815254 10.1145/1348246.1348248 10.7551/mitpress/4175.001.0001 |
| ContentType | Journal Article |
| Copyright | Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) Apr 2014 Distributed under a Creative Commons Attribution 4.0 International License |
| Copyright_xml | – notice: Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) Apr 2014 – notice: Distributed under a Creative Commons Attribution 4.0 International License |
| DBID | 97E RIA RIE AAYXX CITATION 7UA 8FD C1K F1W FR3 H8D H96 KR7 L.G L7M 1XC |
| DOI | 10.1109/JSTARS.2014.2302333 |
| DatabaseName | IEEE Xplore (IEEE) IEEE All-Society Periodicals Package (ASPP) 1998–Present IEEE Xplore CrossRef Water Resources Abstracts Technology Research Database Environmental Sciences and Pollution Management ASFA: Aquatic Sciences and Fisheries Abstracts Engineering Research Database Aerospace Database Aquatic Science & Fisheries Abstracts (ASFA) 2: Ocean Technology, Policy & Non-Living Resources Civil Engineering Abstracts Aquatic Science & Fisheries Abstracts (ASFA) Professional Advanced Technologies Database with Aerospace Hyper Article en Ligne (HAL) |
| DatabaseTitle | CrossRef Aerospace Database Civil Engineering Abstracts Aquatic Science & Fisheries Abstracts (ASFA) Professional Aquatic Science & Fisheries Abstracts (ASFA) 2: Ocean Technology, Policy & Non-Living Resources Technology Research Database ASFA: Aquatic Sciences and Fisheries Abstracts Engineering Research Database Advanced Technologies Database with Aerospace Water Resources Abstracts Environmental Sciences and Pollution Management |
| DatabaseTitleList | Aerospace Database |
| Database_xml | – sequence: 1 dbid: RIE name: IEEE Electronic Library (IEL) url: https://ieeexplore.ieee.org/ sourceTypes: Publisher |
| DeliveryMethod | fulltext_linktorsrc |
| Discipline | Geology Computer Science |
| EISSN | 2151-1535 |
| EndPage | 1141 |
| ExternalDocumentID | oai:HAL:hal-01126389v1 3387985131 10_1109_JSTARS_2014_2302333 6736057 |
| Genre | orig-research |
| GroupedDBID | 0R~ 29I 4.4 5GY 5VS 6IK 97E AAFWJ AAJGR AASAJ AAWTH ABAZT ABVLG ACIWK AENEX AETIX AFPKN AFRAH AGSQL ALMA_UNASSIGNED_HOLDINGS BEFXN BFFAM BGNUA BKEBE BPEOZ DU5 EBS EJD ESBDL GROUPED_DOAJ HZ~ IFIPE IPLJI JAVBF M43 O9- OCL OK1 RIA RIE RNS AAYXX CITATION 7UA 8FD C1K F1W FR3 H8D H96 KR7 L.G L7M RIG 1XC |
| ID | FETCH-LOGICAL-c331t-9e7974610175ce15af4c98d33ab7fed718d6af9f21448e12f8dc090100ef07683 |
| IEDL.DBID | RIE |
| ISICitedReferencesCount | 24 |
| ISICitedReferencesURI | http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000335390000012&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| ISSN | 1939-1404 |
| IngestDate | Sat Nov 29 15:05:45 EST 2025 Fri Jul 25 10:46:25 EDT 2025 Sat Nov 29 07:57:58 EST 2025 Tue Nov 18 21:47:44 EST 2025 Wed Aug 27 02:51:08 EDT 2025 |
| IsPeerReviewed | true |
| IsScholarly | true |
| Issue | 4 |
| Keywords | active learning apprentissage d'instance multiple support vector machines coarse-to-fine testing multiple instance learning apprentissage actif test de gros-de-fine Récupération du modèle Pattern retrieval |
| Language | English |
| License | https://ieeexplore.ieee.org/Xplorehelp/downloads/license-information/IEEE.html Distributed under a Creative Commons Attribution 4.0 International License: http://creativecommons.org/licenses/by/4.0 |
| LinkModel | DirectLink |
| MergedId | FETCHMERGED-LOGICAL-c331t-9e7974610175ce15af4c98d33ab7fed718d6af9f21448e12f8dc090100ef07683 |
| Notes | ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14 |
| PQID | 1549542051 |
| PQPubID | 75722 |
| PageCount | 15 |
| ParticipantIDs | proquest_journals_1549542051 crossref_citationtrail_10_1109_JSTARS_2014_2302333 ieee_primary_6736057 hal_primary_oai_HAL_hal_01126389v1 crossref_primary_10_1109_JSTARS_2014_2302333 |
| PublicationCentury | 2000 |
| PublicationDate | 2014-04-01 |
| PublicationDateYYYYMMDD | 2014-04-01 |
| PublicationDate_xml | – month: 04 year: 2014 text: 2014-04-01 day: 01 |
| PublicationDecade | 2010 |
| PublicationPlace | Piscataway |
| PublicationPlace_xml | – name: Piscataway |
| PublicationTitle | IEEE journal of selected topics in applied earth observations and remote sensing |
| PublicationTitleAbbrev | JSTARS |
| PublicationYear | 2014 |
| Publisher | IEEE The Institute of Electrical and Electronics Engineers, Inc. (IEEE) |
| Publisher_xml | – name: IEEE – name: The Institute of Electrical and Electronics Engineers, Inc. (IEEE) |
| References | ref13 ref12 ref15 gevers (ref34) 2004 ref14 ref31 ref33 ref11 ref32 ref10 ref2 ref1 ref17 ref38 ref16 manjunath (ref28) 2002 abramson (ref9) 2006 ref19 steinwart (ref37) 2008 vapnik (ref36) 1998 andrews (ref21) 2003 liu (ref20) 2006 ref24 ref26 ref25 ref22 sirmacek (ref29) 2010 platt (ref23) 1999 karasulu (ref30) 2010; 4 ref27 schlkopf (ref35) 2001 ref8 ref7 ref4 ref3 ref6 yang (ref18) 2008 sahbi (ref5) 2006; 7 |
| References_xml | – year: 2002 ident: ref28 publication-title: Introduction to MPEG-7 Multimedia Content Description Interface – ident: ref4 doi: 10.1109/TGRS.2006.890579 – ident: ref16 doi: 10.1109/LGRS.2013.2246539 – ident: ref38 doi: 10.1109/TPAMI.2006.248 – ident: ref32 doi: 10.1145/1126004.1126005 – year: 2008 ident: ref37 publication-title: Support Vector Machines – year: 2006 ident: ref20 publication-title: Studies on support vector machines and applications to video object extraction – ident: ref10 doi: 10.1109/ICPR.2010.799 – ident: ref7 doi: 10.1109/ICCV.2001.937694 – year: 2008 ident: ref18 article-title: Review of multi-instance learning and its applications publication-title: Online Publication – volume: 4 start-page: 11 year: 2010 ident: ref30 article-title: Review and evaluation of well-known methods for moving object detection and tracking in videos publication-title: J Aeronaut Space Technol – ident: ref19 doi: 10.1109/TPAMI.2009.155 – ident: ref31 doi: 10.1109/TPAMI.2005.188 – ident: ref12 doi: 10.1023/A:1011113216584 – year: 2004 ident: ref34 publication-title: Emerging Topics in Computer Vision – ident: ref14 doi: 10.1109/TGRS.2008.2010404 – ident: ref25 doi: 10.1007/s00530-002-0070-3 – ident: ref22 doi: 10.1016/S0893-6080(01)00104-6 – year: 1999 ident: ref23 publication-title: Advances in Large Margin Classifiers – ident: ref13 doi: 10.1109/TGRS.2007.892007 – ident: ref8 doi: 10.1109/CVPR.2010.5540115 – ident: ref33 doi: 10.1109/34.895972 – ident: ref3 doi: 10.1109/TMM.2005.858383 – year: 2006 ident: ref9 publication-title: Active learning for visual object detection – year: 1998 ident: ref36 publication-title: The Nature of Statistical Learning Theory – ident: ref15 doi: 10.1109/TGRS.2010.2072929 – start-page: 561 year: 2003 ident: ref21 publication-title: Advances in neural information processing systems – ident: ref17 doi: 10.1109/TGRS.2010.2083673 – ident: ref6 doi: 10.1109/ICIP.2002.1039124 – volume: 7 start-page: 2087 year: 2006 ident: ref5 article-title: A hierarchy of support vector machines for face detection publication-title: J Mach Learn Res – ident: ref11 doi: 10.1023/B:VISI.0000013087.49260.fb – year: 2010 ident: ref29 publication-title: Object Detection in Satellite and Aerial Images Remote Sensing Applications – ident: ref24 doi: 10.1007/s00530-007-0094-9 – ident: ref2 doi: 10.1109/TIP.2008.924286 – ident: ref1 doi: 10.1145/500156.500159 – ident: ref26 doi: 10.1109/TIP.2003.815254 – ident: ref27 doi: 10.1145/1348246.1348248 – year: 2001 ident: ref35 publication-title: Learning With Kernels Support Vector Machines Regularization Optimization and Beyond doi: 10.7551/mitpress/4175.001.0001 |
| SSID | ssj0062793 |
| Score | 2.1396387 |
| Snippet | Pattern retrieval is a fundamental challenge in machine learning but is often subject to the problem of gathering enough labeled examples of the target... |
| SourceID | hal proquest crossref ieee |
| SourceType | Open Access Repository Aggregation Database Enrichment Source Index Database Publisher |
| StartPage | 1127 |
| SubjectTerms | Active learning Algorithms coarse-to-fine testing Computer Science Context Detectors Image databases multiple instance learning (MIL) pattern retrieval Remote sensing Support vector machines support vector machines (SVMs) Training |
| Title | Pattern Retrieval in Large Image Databases Using Multiscale Coarse-to-Fine Cascaded Active Learning |
| URI | https://ieeexplore.ieee.org/document/6736057 https://www.proquest.com/docview/1549542051 https://hal.science/hal-01126389 |
| Volume | 7 |
| WOSCitedRecordID | wos000335390000012&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| hasFullText | 1 |
| inHoldings | 1 |
| isFullTextHit | |
| isPrint | |
| journalDatabaseRights | – providerCode: PRVIEE databaseName: IEEE Electronic Library (IEL) customDbUrl: eissn: 2151-1535 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0062793 issn: 1939-1404 databaseCode: RIE dateStart: 20080101 isFulltext: true titleUrlDefault: https://ieeexplore.ieee.org/ providerName: IEEE |
| link | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV3dS-QwEB_c5Q580Ts_cHVPwnGPG22bpm0el9VVQUT0DvatpPm4O9CuuB_gf-9M2l04lAPpSylJCf0lk5nmN78B-GGUlLLyEddSeZ76SvBKZzFXEqOhVOYUkYViE_nNTTGZqNsNGKxzYZxzgXzmTug2nOXbqVnQr7JT4iChf9GBTp5nTa7WyupmSR4EdtEfUZwkY1qFoThSpzjFh3f3RONKifecCCH-2YU6f4gDGYqrvLHIYZsZb39sgF9gq3Un2bDB_ytsuHoHPl-Ecr0vu2Bug3xmze5C4SycVexvza6J_c2uHtGUsDM917STzVggD7CQkDtD4BwbTTHodXw-5WN0RdlIz4hLb9kwmEjWKrP-3oNf4_Ofo0vellXgRoh4zpXLMYjIaC1K42KpfWpUYYXQVe6dxc3KZtorT2JqhYsTX1gTEYsjcp7O7cQ-dOtp7Q6AZS4uMi0t9ktTofGqLK5jfKO0kY7jHiSrz1yaVnOcSl88lCH2iFTZYFMSNmWLTQ8G605PjeTG_5t_R_zWLUku-3J4XdKziPKj0CNb4kB2Ca11qxaoHvRXcJft2p2VJFon0wSt1eH7vY5gkwbQ8Hf60J0_L9w3-GSWiM_zcZiWr58d3JU |
| linkProvider | IEEE |
| linkToHtml | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV3da9swED_6sbG9tNvasfRrYuyxWm3Lsq3HkC1NWRZK10HfhKyPbbA5o0kL_e97pyiB0TEofjFGMsI_6XRn_e53AO-tklK2IeNGqsDL0AremirnSmI0VMqaIrJYbKKeTJqrK3W-BserXBjvfSSf-Q90G8_y3dTe0K-yE-IgoX-xDptUOStlay3tblXUUWIXPRLFSTQmaQzlmTrBSd6_-EpErpKYz4UQ4q99aP0HsSBjeZUHNjluNMPtxw3xBWwlh5L1FzPgJaz57hU8PY0Fe-92wJ5HAc2OXcTSWTiv2M-OjYn_zc5-ozFhH83c0F42Y5E-wGJK7gyh82wwxbDX8_mUD9EZZQMzIza9Y_1oJFnSZv2-C9-Gny4HI54KK3ArRD7nytcYRlS0GqX1uTShtKpxQpi2Dt7hduUqE1QgObXG50VonM2Ix5H5QCd34jVsdNPOvwFW-bypjHTYryyFwat1uJLxjdJlJs97UCw_s7ZJdZyKX_zSMfrIlF5gowkbnbDpwfGq05-F6Mb_m79D_FYtSTB71B9repZRhhT6ZLc4kB1Ca9UqAdWDgyXcOq3emSbZOlkWaK_2_t3rLTwbXX4Z6_HZ5PM-PKfBLNg8B7Axv77xh_DE3iJW10dxit4DRlHf3g |
| 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%3Ajournal&rft.genre=article&rft.atitle=Pattern+Retrieval+in+Large+Image+Databases+Using+Multiscale+Coarse-to-Fine+Cascaded+Active+Learning&rft.jtitle=IEEE+journal+of+selected+topics+in+applied+earth+observations+and+remote+sensing&rft.au=Blanchart%2C+Pierre&rft.au=Ferecatu%2C+Marin&rft.au=Cui%2C+Shiyong&rft.au=Datcu%2C+Mihai&rft.date=2014-04-01&rft.pub=IEEE&rft.issn=1939-1404&rft.volume=7&rft.issue=4&rft.spage=1127&rft.epage=1141&rft_id=info:doi/10.1109%2FJSTARS.2014.2302333&rft.externalDocID=6736057 |
| thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1939-1404&client=summon |
| thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1939-1404&client=summon |
| thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1939-1404&client=summon |