Comparing accuracy assessments to infer superiority of image classification methods

The z-test based on the Kappa statistic is commonly used to infer superiority of one map production method over another. Typically the same reference data set is used to calculate and next compare the Kappa's of the two maps. This data structure easily leads to dependence between the two error-...

Full description

Saved in:

Bibliographic Details
Published in:	International journal of remote sensing Vol. 27; no. 1; pp. 223 - 232
Main Authors:	de Leeuw, J., Jia, H., Yang, L., Liu, X., Schmidt, K., Skidmore, A. K.
Format:	Journal Article
Language:	English
Published:	Abingdon Taylor & Francis 10.01.2006 Taylor and Francis
Subjects:	Animal, plant and microbial ecology Applied geophysics Biological and medical sciences Earth sciences Earth, ocean, space Exact sciences and technology Fundamental and applied biological sciences. Psychology General aspects. Techniques Internal geophysics system Teledetection and vegetation maps Netherlands Western Europe brackish-water environment maps Europe accuracy vegetation Classifier classification evaluation Randomization salt marshes imagery Alternative method statistical analysis Performance Data structure errors
ISSN:	0143-1161, 1366-5901
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Abstract	The z-test based on the Kappa statistic is commonly used to infer superiority of one map production method over another. Typically the same reference data set is used to calculate and next compare the Kappa's of the two maps. This data structure easily leads to dependence between the two error-matrices. This may result in overly large variance estimates and too conservative inference about the difference in accuracy between the two methods. Tests considering the dependency between the error matrices would be more sensitive in such case. In this article we compare the performance of two such tests, a randomization and McNemar's test, with the traditional z-test. We compared 16 alternative methods to classify salt marsh vegetation in The Netherlands. The error matrices were positively associated in all 120 possible comparisons of pairs of classification methods. This suggests that dependency between pairs of error matrices used in classifier comparison is a common phenomenon. Both the randomization and McNemar test gave lower p values and rejected the null hypothesis of equal performance more frequently than the z-test. We therefore recommend considering their use.
AbstractList	The z-test based on the Kappa statistic is commonly used to infer superiority of one map production method over another. Typically the same reference data set is used to calculate and next compare the Kappa's of the two maps. This data structure easily leads to dependence between the two error-matrices. This may result in overly large variance estimates and too conservative inference about the difference in accuracy between the two methods. Tests considering the dependency between the error matrices would be more sensitive in such case. In this article we compare the performance of two such tests, a randomization and McNemar's test, with the traditional z-test. We compared 16 alternative methods to classify salt marsh vegetation in The Netherlands. The error matrices were positively associated in all 120 possible comparisons of pairs of classification methods. This suggests that dependency between pairs of error matrices used in classifier comparison is a common phenomenon. Both the randomization and McNemar test gave lower p values and rejected the null hypothesis of equal performance more frequently than the z-test. We therefore recommend considering their use The z-test based on the Kappa statistic is commonly used to infer superiority of one map production method over another. Typically the same reference data set is used to calculate and next compare the Kappa's of the two maps. This data structure easily leads to dependence between the two error-matrices. This may result in overly large variance estimates and too conservative inference about the difference in accuracy between the two methods. Tests considering the dependency between the error matrices would be more sensitive in such case. In this article we compare the performance of two such tests, a randomization and McNemar's test, with the traditional z-test. We compared 16 alternative methods to classify salt marsh vegetation in The Netherlands. The error matrices were positively associated in all 120 possible comparisons of pairs of classification methods. This suggests that dependency between pairs of error matrices used in classifier comparison is a common phenomenon. Both the randomization and McNemar test gave lower p values and rejected the null hypothesis of equal performance more frequently than the z-test. We therefore recommend considering their use.
Author	Schmidt, K. Jia, H. Liu, X. de Leeuw, J. Yang, L. Skidmore, A. K.
Author_xml	– sequence: 1 givenname: J. surname: de Leeuw fullname: de Leeuw, J. email: leeuw@itc.nl organization: Department of Natural Resources , International Institute for Geo-Information Science and Earth Observation – sequence: 2 givenname: H. surname: Jia fullname: Jia, H. organization: Department of Environmental Science and Engineering , Tsinghua University – sequence: 3 givenname: L. surname: Yang fullname: Yang, L. organization: Department of Civil and Environmental Engineering , University of Cincinnati – sequence: 4 givenname: X. surname: Liu fullname: Liu, X. organization: Department of Environmental Science and Engineering , Tsinghua University – sequence: 5 givenname: K. surname: Schmidt fullname: Schmidt, K. organization: Department of Natural Resources , International Institute for Geo-Information Science and Earth Observation – sequence: 6 givenname: A. K. surname: Skidmore fullname: Skidmore, A. K. organization: Department of Natural Resources , International Institute for Geo-Information Science and Earth Observation
BackLink	http://pascal-francis.inist.fr/vibad/index.php?action=getRecordDetail&idt=17474293$$DView record in Pascal Francis
BookMark	eNqFkUtvEzEUhS1UJNKUH8DOG9gN-DkTIzYo4iVVYtF2bd1x7GLksQfbo5B_j0MCCypgdRf3fOfcxyW6iClahJ5R8pKSDXlFqOCU9kQSwgY59OwRWlHe951UhF6g1bHfNQF9gi5L-UoI6ZtshW62aZoh-3iPwZglgzlgKMWWMtlYC64J--hsxmWZbfYp-3rAyWE_wb3FJjStd95A9SniydYvaVeu0GMHodin57pGd-_f3W4_dtefP3zavr3ujGSidmZjpXPjxggpleFy3FElnLKCkpExJhXlXAgxmBHEMIxgqd0xpQSjdtNWU3yNXp98922W2FawUUfIxhedwOvgxwz5oPdL1jEcy7yMRXPJSLNeoxcneM7p22JL1ZMvxoYA0aalaKZoy6akCZ-fhVAMBJchHhPm3E7Q3OkgBsEUb7rhpDM5lZKt08bXn3epGXzQlOjjp_SDTzWS_kH-Nv8Hc05r30l5gn3KYacrHELKv0Z8QOn6vTbyzX9J_vfgH1wtvag
CODEN	IJSEDK
CitedBy_id	crossref_primary_10_1016_j_jag_2010_11_007 crossref_primary_10_3390_rs11171966 crossref_primary_10_1080_01431160701408436 crossref_primary_10_1109_JSTARS_2013_2266354 crossref_primary_10_1371_journal_pone_0098690 crossref_primary_10_3390_rs11010043 crossref_primary_10_1016_j_cageo_2011_08_019 crossref_primary_10_1007_s43621_024_00735_z crossref_primary_10_3390_fire4010014 crossref_primary_10_3390_rs10101643 crossref_primary_10_1016_j_isprsjprs_2017_05_002 crossref_primary_10_3390_rs70911664 crossref_primary_10_1016_j_rse_2014_08_008 crossref_primary_10_1080_10106049_2018_1474274 crossref_primary_10_3390_agronomy9070373 crossref_primary_10_1016_j_rse_2010_08_019 crossref_primary_10_1007_s12518_019_00258_7 crossref_primary_10_1007_s10336_008_0319_5 crossref_primary_10_1016_j_jag_2018_03_002 crossref_primary_10_1080_01431161_2019_1608391 crossref_primary_10_1016_j_envsoft_2018_01_023 crossref_primary_10_1080_07038992_2015_1065706 crossref_primary_10_1007_s12665_025_12377_z crossref_primary_10_7717_peerj_4540 crossref_primary_10_1080_01431161_2014_903435 crossref_primary_10_1088_1755_1315_789_1_012052 crossref_primary_10_3390_rs11111279 crossref_primary_10_1109_JSTARS_2014_2347203 crossref_primary_10_1016_j_rsase_2020_100436 crossref_primary_10_1016_j_envpol_2019_113360 crossref_primary_10_1080_10106049_2014_883439 crossref_primary_10_1016_j_atmosenv_2017_01_004 crossref_primary_10_1109_TGRS_2020_2999558 crossref_primary_10_3390_rs1030330 crossref_primary_10_1016_j_rse_2013_08_029 crossref_primary_10_1080_01431161_2024_2379515 crossref_primary_10_1109_JSTARS_2021_3083517 crossref_primary_10_1016_j_jas_2018_12_003 crossref_primary_10_1080_10106049_2020_1756460 crossref_primary_10_1007_s12517_015_2267_8 crossref_primary_10_1016_j_apgeog_2018_10_004 crossref_primary_10_1016_j_apgeog_2017_04_005 crossref_primary_10_1016_j_rsase_2020_100302 crossref_primary_10_1177_0309133307081294 crossref_primary_10_1016_j_isprsjprs_2017_04_016 crossref_primary_10_1007_s12145_021_00685_4 crossref_primary_10_1016_j_jag_2018_06_011 crossref_primary_10_1145_3686990 crossref_primary_10_3390_rs9090885 crossref_primary_10_1080_19475705_2024_2383309 crossref_primary_10_1016_j_rse_2017_03_001 crossref_primary_10_5721_EuJRS20164924 crossref_primary_10_1016_j_jag_2014_08_001 crossref_primary_10_1007_s12145_025_01819_8 crossref_primary_10_3390_rs12030529 crossref_primary_10_1590_1807_1929_agriambi_v22n5p366_370 crossref_primary_10_1016_j_jag_2018_06_018 crossref_primary_10_1080_01431161_2023_2292015 crossref_primary_10_1016_j_jag_2012_02_004 crossref_primary_10_3390_rs13071383 crossref_primary_10_1016_j_pce_2017_02_015 crossref_primary_10_3390_ijgi8090398 crossref_primary_10_1080_01431161_2013_774099 crossref_primary_10_1016_j_compag_2015_09_020 crossref_primary_10_3390_rs12061010 crossref_primary_10_1007_s11042_022_12263_x crossref_primary_10_1016_j_rse_2017_07_010 crossref_primary_10_1007_s10346_013_0391_7 crossref_primary_10_1080_01431161_2020_1754493 crossref_primary_10_3390_rs13091666 crossref_primary_10_1080_2150704X_2015_1109158 crossref_primary_10_1016_j_rsase_2018_01_002 crossref_primary_10_3390_ijgi3010297 crossref_primary_10_3390_rs13193909 crossref_primary_10_7717_peerj_15027 crossref_primary_10_1080_01431161_2013_793861 crossref_primary_10_1016_j_jag_2017_09_004 crossref_primary_10_1109_TGRS_2011_2162589 crossref_primary_10_3390_su14095700 crossref_primary_10_3390_f5061304 crossref_primary_10_1016_j_rse_2020_111939 crossref_primary_10_1016_j_cageo_2013_07_002 crossref_primary_10_1109_TGRS_2008_2001035 crossref_primary_10_1007_s10661_007_9981_y crossref_primary_10_3390_rs3112364 crossref_primary_10_1007_s11042_021_10824_0 crossref_primary_10_1007_s12524_019_01069_4 crossref_primary_10_3390_w12123366 crossref_primary_10_1080_10106049_2019_1581266 crossref_primary_10_1080_10106049_2019_1629645 crossref_primary_10_1007_s11273_025_10052_5 crossref_primary_10_1016_j_rse_2012_05_019 crossref_primary_10_1109_JSTARS_2015_2461136 crossref_primary_10_1109_JSTARS_2016_2574360 crossref_primary_10_1016_j_apgeog_2010_11_006 crossref_primary_10_3390_rs13132445 crossref_primary_10_1016_j_envsoft_2024_105956 crossref_primary_10_1016_j_isprsjprs_2013_11_013 crossref_primary_10_1016_j_jag_2011_06_008 crossref_primary_10_3390_f16040676 crossref_primary_10_3390_rs16122167 crossref_primary_10_1080_01431160903571791 crossref_primary_10_1080_01431160903130937 crossref_primary_10_1007_s00330_017_5171_7 crossref_primary_10_1016_j_enggeo_2015_04_004 crossref_primary_10_1016_j_isprsjprs_2018_02_002 crossref_primary_10_1088_1755_1315_767_1_012041 crossref_primary_10_1109_JSTARS_2025_3600991 crossref_primary_10_1016_j_rse_2015_12_026 crossref_primary_10_1016_j_atech_2023_100347 crossref_primary_10_1007_s12518_022_00423_5 crossref_primary_10_1080_01431161_2024_2424510 crossref_primary_10_1007_s13280_015_0649_5 crossref_primary_10_3390_rs14051226 crossref_primary_10_1007_s00024_023_03418_4 crossref_primary_10_1080_10106049_2015_1132483 crossref_primary_10_3390_rs10111704
Cites_doi	10.1037/h0028106 10.1177/001316446002000104 10.1016/0034-4257(91)90048-B 10.1016/0034-4257(93)90013-N 10.1007/0-306-47647-9_12 10.14358/PERS.70.6.703 10.1007/BF02295996
ContentType	Journal Article
Copyright	Copyright Taylor & Francis Group, LLC 2006 2006 INIST-CNRS Wageningen University & Research
Copyright_xml	– notice: Copyright Taylor & Francis Group, LLC 2006 – notice: 2006 INIST-CNRS – notice: Wageningen University & Research
DBID	AAYXX CITATION IQODW 8FD FR3 H8D KR7 L7M QVL
DOI	10.1080/01431160500275762
DatabaseName	CrossRef Pascal-Francis Technology Research Database Engineering Research Database Aerospace Database Civil Engineering Abstracts Advanced Technologies Database with Aerospace NARCIS:Publications
DatabaseTitle	CrossRef Aerospace Database Civil Engineering Abstracts Engineering Research Database Technology Research Database Advanced Technologies Database with Aerospace
DatabaseTitleList	Aerospace Database
DeliveryMethod	fulltext_linktorsrc
Discipline	Geography
EISSN	1366-5901
EndPage	232
ExternalDocumentID	oai_library_wur_nl_wurpubs_352091 17474293 10_1080_01431160500275762 127559
GeographicLocations	Netherlands Western Europe
GroupedDBID	-~X .7F .DC .QJ 0BK 0R~ 29J 30N 4.4 5GY 5VS 6TJ AAENE AAGDL AAHBH AAHIA AAJMT AALDU AAMIU AAPUL AAQRR ABCCY ABDPE ABFIM ABHAV ABJNI ABLIJ ABLJU ABPAQ ABPEM ABRLO ABUFD ABXUL ABXYU ACGEJ ACGFS ACIWK ACTIO ACTTO ADCVX ADGTB ADXPE AEISY AENEX AEOZL AEPSL AEXLP AEYOC AFBWG AFION AFKVX AFRVT AGDLA AGMYJ AGVKY AGWUF AHDZW AIJEM AIYEW AJWEG AKBVH AKOOK ALMA_UNASSIGNED_HOLDINGS ALQZU ALRRR AQRUH AQTUD AVBZW AWYRJ BLEHA BWMZZ CAG CCCUG CE4 COF CS3 CYRSC DAOYK DGEBU DKSSO DU5 EBS EJD E~A E~B F5P H13 H~9 IPNFZ J.P KYCEM M4Z OPCYK P2P RIG RNANH ROSJB RTWRZ S-T SNACF TAJZE TASJS TBQAZ TDBHL TEN TFL TFT TFW TN5 TNC TQWBC TTHFI TUROJ TWF UPT UT5 UU3 ZGOLN ZY4 ~02 ~S~ AAYXX CITATION 07I 1TA 4B5 ABFMO ADXEU ADYSH AEHZU AEZBV AGBLW AI. AIDBO AKHJE AKMBP ALXIB BGSSV C0- C5H DEXXA FETWF HF~ IFELN IQODW L8C LJTGL TAP UB6 VH1 VOH 8FD FR3 H8D KR7 L7M 02 08R 7F AAAVI ABFLS ABJVF ABPTK ABQHQ ADCOW AEGYZ AFOLD AHDLD AIRXU FUNRP FVPDL G8K NUSFT QJ QVL S TEJ UAO UNR V1K X
ID	FETCH-LOGICAL-c524t-c8e5ffb8c4559c35bd194f9e410b222591334447cba477bae1ed299421e813693
IEDL.DBID	TFW
ISICitedReferencesCount	136
ISICitedReferencesURI	http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000235364700014&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN	0143-1161
IngestDate	Tue Jan 05 18:10:52 EST 2021 Thu Sep 04 16:08:13 EDT 2025 Mon Jul 21 09:15:01 EDT 2025 Sat Nov 29 03:33:27 EST 2025 Tue Nov 18 22:25:24 EST 2025 Mon May 13 12:08:30 EDT 2019 Mon Oct 20 23:43:29 EDT 2025
IsPeerReviewed	true
IsScholarly	true
Issue	1
Keywords	brackish-water environment maps Europe accuracy vegetation Classifier classification evaluation Randomization salt marshes imagery Alternative method statistical analysis Performance Data structure errors
Language	English
License	CC BY 4.0
LinkModel	DirectLink
MergedId	FETCHMERGED-LOGICAL-c524t-c8e5ffb8c4559c35bd194f9e410b222591334447cba477bae1ed299421e813693
Notes	ObjectType-Article-2 SourceType-Scholarly Journals-1 ObjectType-Feature-1 content type line 23
PQID	29144710
PQPubID	23500
PageCount	10
ParticipantIDs	pascalfrancis_primary_17474293 proquest_miscellaneous_29144710 wageningen_narcis_oai_library_wur_nl_wurpubs_352091 informaworld_taylorfrancis_310_1080_01431160500275762 crossref_citationtrail_10_1080_01431160500275762 crossref_primary_10_1080_01431160500275762
ProviderPackageCode	QVL
PublicationCentury	2000
PublicationDate	2006-01-10
PublicationDateYYYYMMDD	2006-01-10
PublicationDate_xml	– month: 01 year: 2006 text: 2006-01-10 day: 10
PublicationDecade	2000
PublicationPlace	Abingdon
PublicationPlace_xml	– name: Abingdon
PublicationTitle	International journal of remote sensing
PublicationYear	2006
Publisher	Taylor & Francis Taylor and Francis
Publisher_xml	– name: Taylor & Francis – name: Taylor and Francis
References	ref12 Rosenfield G. H. (ref11) 1981; 47 ref10 ref2 Janssen L. L. F. (ref6) 1994; 60 ref7 Congalton R. G. (ref3) 1983; 49 Manly B. F. J. (ref9) 1996 ref4 ref5 Ma Z. (ref8) 1995; 61 Skidmore A. K. (ref13) 1999 Zarr J. H. (ref14) 1996 Agresti A. (ref1) 1990
References_xml	– ident: ref5 doi: 10.1037/h0028106 – ident: ref2 doi: 10.1177/001316446002000104 – volume: 60 start-page: 419 year: 1994 ident: ref6 publication-title: Photogrammetric Engineering and Remote Sensing – ident: ref4 doi: 10.1016/0034-4257(91)90048-B – ident: ref7 doi: 10.1016/0034-4257(93)90013-N – volume-title: Biostatistical Analysis year: 1996 ident: ref14 – volume: 49 start-page: 1671 year: 1983 ident: ref3 publication-title: Photogrammetric Engineering and Remote Sensing – volume: 47 start-page: 1685 year: 1981 ident: ref11 publication-title: Photogrammetric Engineering and Remote Sensing – volume: 61 start-page: 435 year: 1995 ident: ref8 publication-title: Photogrammetric Engineering and Remote Sensing – volume-title: Categorical data analysis year: 1990 ident: ref1 – volume-title: Randomization, Bootstrap and Monte Carlo Tests in Biology year: 1996 ident: ref9 – start-page: 197 volume-title: Spatial Statistics for Remote Sensing year: 1999 ident: ref13 doi: 10.1007/0-306-47647-9_12 – ident: ref12 doi: 10.14358/PERS.70.6.703 – ident: ref10 doi: 10.1007/BF02295996
SSID	ssj0006757
Score	2.218967
Snippet	The z-test based on the Kappa statistic is commonly used to infer superiority of one map production method over another. Typically the same reference data set...
SourceID	wageningen proquest pascalfrancis crossref informaworld
SourceType	Open Access Repository Aggregation Database Index Database Enrichment Source Publisher
StartPage	223
SubjectTerms	Animal, plant and microbial ecology Applied geophysics Biological and medical sciences Earth sciences Earth, ocean, space Exact sciences and technology Fundamental and applied biological sciences. Psychology General aspects. Techniques Internal geophysics system Teledetection and vegetation maps
Title	Comparing accuracy assessments to infer superiority of image classification methods
URI	https://www.tandfonline.com/doi/abs/10.1080/01431160500275762 https://www.proquest.com/docview/29144710 http://www.narcis.nl/publication/RecordID/oai:library.wur.nl:wurpubs%2F352091
Volume	27
WOSCitedRecordID	wos000235364700014&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: PRVAWR databaseName: Taylor & Francis Journals Complete customDbUrl: eissn: 1366-5901 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0006757 issn: 0143-1161 databaseCode: TFW dateStart: 19800101 isFulltext: true titleUrlDefault: https://www.tandfonline.com providerName: Taylor & Francis
link	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwpV1NT9wwEB1VCKlcCrRUDRTqQ0-VIuLYWSfHCrHqCSGVqtwsx2u3SDRBcVLKv2cmHwsL1R7aU6QoH45nxn7jjN8D-Oikz23mfZz5UsTSpVlszMLHDrG_MUYlZT6ITaizs_zysjgfa3PCWFZJObQfiCL6sZqC25Rhqog7Jko6zhGHU06FeJlGYET15N8X8-_LcRih8LBZmkg4EdhM_zT_9oSVWWmFs5SKJU3A_vKD0MUKEt26xaCv-l1Qj2al-fZ_fs8OvBrhKPs8-M8uvHDVa3g5KqP_vHsDX08GocLqBzPWdo2xd8ws2TwDa2tG9VwNCx1xJtekhcdqz65-YZuZJWxOxUi9_dkgVx324Nv89OLkSzwKMcQ2S2Ub29yhOcvcSsw_rMjKBS-kL5zkSUn5YoGJrpRS2dJIpUrjuFvgNCdT7nIuZoV4CxtVXbl3wOTM2EKKlPZNSZelRnCDp3KrnJrNlIogmQyh7chSTmIZ15pPZKZPeyuCT8tbbgaKjnUXJ4-tq9t-XWS07fPLdfunjSBbc4tY86qjFc95aBymcAgDRAQfJlfSGNT0p8ZUru6CTgvsHcR-EYgHD9MVqUsFTYTg4xKfvu0aXV3TAaMXG0NFTXz_Hxt8AFvTAhNP3sNG23TuEDbt7_YqNEd9cN0DZHciUQ
linkProvider	Taylor & Francis
linkToHtml	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwpV1LT9wwEB4hWgkufVcNpeBDT5WixrGzTo4VYgUqrJC6VblZjtdukWhS5QHl33cmj4Vt0R7KKVKUh2PP2N9Mxt8H8N5Jn9rE-zDxuQili5PQmIUPHWJ_Y4yK8rQXm1CzWXp-np0NCbd6KKukGNr3RBHdXE3OTcnosSTuI3HScY5AnIIqBMw4BT9KcJ2lkr759NtyJkYw3G-XJhpOhDbjX837HrGyLq2wllK5pKmxx3wvdbGCRbev0e2Lbh_UnXVp-vShX_QMngyIlH3qTeg5bLjiBWwN4ug_bl7Cl4Neq7D4zoy1bWXsDTNLQs-aNSWjkq6K1S3RJpckh8dKzy5-YqOZJXhO9UidCbBesbp-BV-nh_ODo3DQYghtEssmtKnDEc1TKzEEsSLJFzyTPnOSRzmFjBnGulJKZXMjlcqN426BK52MuUu5mGTiNWwWZeHeAJMTYzMpYto6JV0SG8ENnkqtcmoyUSqAaBwJbQeictLLuNR85DP9u7cC-LC85VfP0rHu4uju8OqmS40Mg_vv5br53QSQrLlFrHnV3orp3DYOozhEAiKA_dGWNPo1_awxhSvbWscZ9g7CvwDErYnpggSmak2c4EOWT1-3lS4u6YAOjI2huia-858N3oeto_npiT45nn1-C9tjvolHu7DZVK17B4_tVXNRV3udp_0Bjewmcg
linkToPdf	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwpV1LT9wwEB4hilouhT4QaSn40FOlqHHsrJNjBaxagVZIUJWb5XjtFokmKI9S_n1n8ljYttpDe4oUxYljz9jf2OPvA3jrpE9t4n2Y-FyE0sVJaMzchw6xvzFGRXnai02o2Sy9vMzOhtycekirpBja90QR3VhNzn0z92NG3HuipOMccTjFVIiXcQR-1BFjoTlfTL8sBmLEwv1paWLhRGQzbmr-7RVL09ISaSllS5oaG8z3ShdLUHTzFr2-6I5BPZiWplv_-UPb8HTAo-xDb0DPYM0Vz-HJII3-7e4FnB_2SoXFV2asbStj75hZ0HnWrCkZJXRVrG6JNLkkMTxWenb1HevMLIFzykbqDID1etX1S_g8Pb44_BgOSgyhTWLZhDZ12J95aiUGIFYk-Zxn0mdO8iingDHDSFdKqWxupFK5cdzNcZ6TMXcpF5NM7MB6URZuF5icGJtJEdPBKemS2Ahu8FZqlVOTiVIBRGNHaDvQlJNaxrXmI5vp760VwLtFkZueo2PVw9HD3tVNtzAy9O2fj-vmZxNAsqKIWPGp_SXLua8cxnCIA0QAB6MpafRq2qoxhSvbWscZtg6CvwDEvYXpguSlak2M4MMan75tK11c0wXdFytDWU381T9W-AAenx1N9emn2clr2BwXm3i0B-tN1bo3sGF_NFd1td_52S_KnyUW
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=Comparing+accuracy+assessments+to+infer+superiority+of+image+classification+methods&rft.jtitle=International+journal+of+remote+sensing&rft.au=de+Leeuw%2C+J.&rft.au=Jia%2C+H.&rft.au=Yang%2C+L.&rft.au=Liu%2C+X.&rft.date=2006-01-10&rft.issn=0143-1161&rft.eissn=1366-5901&rft.volume=27&rft.issue=1&rft.spage=223&rft.epage=232&rft_id=info:doi/10.1080%2F01431160500275762&rft.externalDBID=n%2Fa&rft.externalDocID=10_1080_01431160500275762
thumbnail_l	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0143-1161&client=summon
thumbnail_m	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0143-1161&client=summon
thumbnail_s	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0143-1161&client=summon