{"id":"https://openalex.org/W7225627386","doi":"https://doi.org/10.3390/w18212791","title":"Distribution-Aware Evaluation of Deep Learning Detectors for Floating-Object Detection on Water Surfaces: Small Objects, Class Imbalance, and Computational Efficiency","display_name":"Distribution-Aware Evaluation of Deep Learning Detectors for Floating-Object Detection on Water Surfaces: Small Objects, Class Imbalance, and Computational Efficiency","publication_year":2026,"publication_date":"2026-11-09","ids":{"openalex":"https://openalex.org/W7225627386","doi":"https://doi.org/10.3390/w18212791"},"language":"en","primary_location":{"id":"doi:10.3390/w18212791","doi":"https://doi.org/10.3390/w18212791","is_oa":true,"landing_page_url":"https://doi.org/10.3390/w18212791","pdf_url":null,"source":{"id":"https://openalex.org/S134216166","display_name":"Water","issn_l":"2073-4441","issn":["2073-4441"],"is_oa":true,"is_in_doaj":false,"is_core":true,"listed_in":["anvur-scientific","cwts-core","norway-1","poland-100","russia-white-list-1","vabb-shw"],"host_organization":"https://openalex.org/P4310310987","host_organization_name":"Multidisciplinary Digital Publishing Institute","host_organization_lineage":["https://openalex.org/P4310310987"],"host_organization_lineage_names":["Multidisciplinary Digital Publishing Institute"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Water","raw_type":"journal-article"},"volume":null,"type":"article","is_traditional":true,"indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://doi.org/10.3390/w18212791","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5161344740","display_name":"Jianwen Mao","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I129604602","display_name":"The University of Sydney","ror":"https://ror.org/0384j8v12","country_code":"AU","type":"education","lineage":["https://openalex.org/I129604602"]}],"countries":["AU"],"is_corresponding":true,"raw_author_name":"Jianwen Mao","raw_affiliation_strings":["School of Computer Science, Faculty of Engineering, The University of Sydney, Sydney, NSW 2006, Australia"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Computer Science, Faculty of Engineering, The University of Sydney, Sydney, NSW 2006, Australia","institution_ids":["https://openalex.org/I129604602"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5074199759","display_name":"Shibiao Fang","orcid":"https://orcid.org/0000-0002-5893-0580","observed_orcids":["https://orcid.org/0000-0002-5893-0580"]},"institutions":[{"id":"https://openalex.org/I109935558","display_name":"Ningbo University","ror":"https://ror.org/03et85d35","country_code":"CN","type":"education","lineage":["https://openalex.org/I109935558"]},{"id":"https://openalex.org/I180726961","display_name":"Shenzhen University","ror":"https://ror.org/01vy4gh70","country_code":"CN","type":"education","lineage":["https://openalex.org/I180726961"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Shibiao Fang","raw_affiliation_strings":["College of Computer Science and Software Engineering, Shenzhen University, Shenzhen 518060, China","Faculty of Maritime and Transportation, Ningbo University, Ningbo 315832, China"],"raw_orcid":"https://orcid.org/0000-0002-5893-0580","affiliations":[{"raw_affiliation_string":"College of Computer Science and Software Engineering, Shenzhen University, Shenzhen 518060, China","institution_ids":["https://openalex.org/I180726961"]},{"raw_affiliation_string":"Faculty of Maritime and Transportation, Ningbo University, Ningbo 315832, China","institution_ids":["https://openalex.org/I109935558"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5059262581","display_name":"Wenrong Tu","orcid":"https://orcid.org/0000-0003-2653-5903","observed_orcids":["https://orcid.org/0000-0003-2653-5903"]},"institutions":[{"id":"https://openalex.org/I4210108865","display_name":"Zhejiang Institute of Communications","ror":"https://ror.org/01q17sd51","country_code":"CN","type":"education","lineage":["https://openalex.org/I4210108865"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Wenrong Tu","raw_affiliation_strings":["Road and Bridge School, Zhejiang Institute of Communications, Hangzhou 311112, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Road and Bridge School, Zhejiang Institute of Communications, Hangzhou 311112, China","institution_ids":["https://openalex.org/I4210108865"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":4,"corresponding_author_ids":["https://openalex.org/A5074199759","https://openalex.org/A5161344740"],"corresponding_institution_ids":["https://openalex.org/I109935558","https://openalex.org/I129604602","https://openalex.org/I180726961"],"apc_list":{"value":2600,"currency":"CHF","value_usd":3137},"apc_paid":{"value":2600,"currency":"CHF","value_usd":3137},"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.0,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"18","issue":"21","first_page":"2791","last_page":"2791"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9961000084877014,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9961000084877014,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T12697","display_name":"Water Quality Monitoring Technologies","score":0.0012000000569969416,"subfield":{"id":"https://openalex.org/subfields/2312","display_name":"Water Science and Technology"},"field":{"id":"https://openalex.org/fields/23","display_name":"Environmental Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10689","display_name":"Remote-Sensing Image Classification","score":0.000699999975040555,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/object-detection","display_name":"object detection","score":0.5210000276565552},{"id":"https://openalex.org/keywords/small-object-detection","display_name":"small object detection","score":0.7559999823570251},{"id":"https://openalex.org/keywords/class-imbalance","display_name":"class imbalance","score":0.996999979019165},{"id":"https://openalex.org/keywords/distribution-shift","display_name":"distribution shift","score":0.7459999918937683},{"id":"https://openalex.org/keywords/retinanet","display_name":"RetinaNet","score":0.828000009059906},{"id":"https://openalex.org/keywords/faster-r-cnn","display_name":"Faster R-CNN","score":0.9959999918937683},{"id":"https://openalex.org/keywords/computational-efficiency","display_name":"computational efficiency","score":0.8169999718666077},{"id":"https://openalex.org/keywords/inference-throughput","display_name":"inference throughput","score":0.7979999780654907},{"id":"https://openalex.org/keywords/class-balanced-sampling","display_name":"class-balanced sampling","score":0.824999988079071},{"id":"https://openalex.org/keywords/water-pollution-monitoring","display_name":"water pollution monitoring","score":0.4659999907016754}],"concepts":[{"id":"https://openalex.org/C189430467","wikidata":"https://www.wikidata.org/wiki/Q7293293","display_name":"Ranking (information retrieval)","level":2,"score":0.6597999930381775},{"id":"https://openalex.org/C140779682","wikidata":"https://www.wikidata.org/wiki/Q210868","display_name":"Sampling (signal processing)","level":3,"score":0.5570999979972839},{"id":"https://openalex.org/C94915269","wikidata":"https://www.wikidata.org/wiki/Q1834857","display_name":"Detector","level":2,"score":0.5264000296592712},{"id":"https://openalex.org/C2777212361","wikidata":"https://www.wikidata.org/wiki/Q5127848","display_name":"Class (philosophy)","level":2,"score":0.5065000057220459},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.4970000088214874},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.49619999527931213},{"id":"https://openalex.org/C169903167","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Test set","level":2,"score":0.4788999855518341},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.47040000557899475},{"id":"https://openalex.org/C4679612","wikidata":"https://www.wikidata.org/wiki/Q866298","display_name":"Aggregate (composite)","level":2,"score":0.4684999883174896},{"id":"https://openalex.org/C2776151529","wikidata":"https://www.wikidata.org/wiki/Q3045304","display_name":"Object detection","level":3,"score":0.45320001244544983},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.40299999713897705},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.37790000438690186},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.37299999594688416},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.37229999899864197},{"id":"https://openalex.org/C2781238097","wikidata":"https://www.wikidata.org/wiki/Q175026","display_name":"Object (grammar)","level":2,"score":0.3361999988555908},{"id":"https://openalex.org/C169258074","wikidata":"https://www.wikidata.org/wiki/Q245748","display_name":"Random forest","level":2,"score":0.32170000672340393},{"id":"https://openalex.org/C27181475","wikidata":"https://www.wikidata.org/wiki/Q541014","display_name":"Cross-validation","level":2,"score":0.3005000054836273},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3003999888896942},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.30000001192092896},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.29910001158714294},{"id":"https://openalex.org/C149629883","wikidata":"https://www.wikidata.org/wiki/Q660926","display_name":"Fraction (chemistry)","level":2,"score":0.27149999141693115},{"id":"https://openalex.org/C2777267654","wikidata":"https://www.wikidata.org/wiki/Q3519023","display_name":"Test (biology)","level":2,"score":0.2685000002384186},{"id":"https://openalex.org/C82685317","wikidata":"https://www.wikidata.org/wiki/Q19829510","display_name":"Pollutant","level":2,"score":0.26489999890327454},{"id":"https://openalex.org/C2985946229","wikidata":"https://www.wikidata.org/wiki/Q49908","display_name":"Data sampling","level":2,"score":0.2581999897956848},{"id":"https://openalex.org/C58489278","wikidata":"https://www.wikidata.org/wiki/Q1172284","display_name":"Data set","level":2,"score":0.25589999556541443},{"id":"https://openalex.org/C87007009","wikidata":"https://www.wikidata.org/wiki/Q210832","display_name":"Statistical hypothesis testing","level":2,"score":0.2551000118255615},{"id":"https://openalex.org/C3019060180","wikidata":"https://www.wikidata.org/wiki/Q184199","display_name":"Automated method","level":2,"score":0.25459998846054077},{"id":"https://openalex.org/C44249647","wikidata":"https://www.wikidata.org/wiki/Q208498","display_name":"Confidence interval","level":2,"score":0.2526000142097473}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.3390/w18212791","doi":"https://doi.org/10.3390/w18212791","is_oa":true,"landing_page_url":"https://doi.org/10.3390/w18212791","pdf_url":null,"source":{"id":"https://openalex.org/S134216166","display_name":"Water","issn_l":"2073-4441","issn":["2073-4441"],"is_oa":true,"is_in_doaj":false,"is_core":true,"listed_in":["anvur-scientific","cwts-core","norway-1","poland-100","russia-white-list-1","vabb-shw"],"host_organization":"https://openalex.org/P4310310987","host_organization_name":"Multidisciplinary Digital Publishing Institute","host_organization_lineage":["https://openalex.org/P4310310987"],"host_organization_lineage_names":["Multidisciplinary Digital Publishing Institute"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Water","raw_type":"journal-article"}],"best_oa_location":{"id":"doi:10.3390/w18212791","doi":"https://doi.org/10.3390/w18212791","is_oa":true,"landing_page_url":"https://doi.org/10.3390/w18212791","pdf_url":null,"source":{"id":"https://openalex.org/S134216166","display_name":"Water","issn_l":"2073-4441","issn":["2073-4441"],"is_oa":true,"is_in_doaj":false,"is_core":true,"listed_in":["anvur-scientific","cwts-core","norway-1","poland-100","russia-white-list-1","vabb-shw"],"host_organization":"https://openalex.org/P4310310987","host_organization_name":"Multidisciplinary Digital Publishing Institute","host_organization_lineage":["https://openalex.org/P4310310987"],"host_organization_lineage_names":["Multidisciplinary Digital Publishing Institute"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Water","raw_type":"journal-article"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/6","display_name":"Clean water and sanitation","score":0.53329998254776}],"study_designs":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":22,"referenced_works":["https://openalex.org/W1861492603","https://openalex.org/W2031342017","https://openalex.org/W2031489346","https://openalex.org/W2037496050","https://openalex.org/W2193145675","https://openalex.org/W2565639579","https://openalex.org/W2604316091","https://openalex.org/W2604319603","https://openalex.org/W2936645470","https://openalex.org/W2963351448","https://openalex.org/W3013211776","https://openalex.org/W3030520226","https://openalex.org/W3043995050","https://openalex.org/W3122799380","https://openalex.org/W3204885083","https://openalex.org/W4285803975","https://openalex.org/W4386076325","https://openalex.org/W4389393236","https://openalex.org/W4390874554","https://openalex.org/W4402754006","https://openalex.org/W4410579470","https://openalex.org/W4413144769"],"related_works":["https://openalex.org/W4318476537","https://openalex.org/W7117987953","https://openalex.org/W7162841185","https://openalex.org/W4392801944","https://openalex.org/W7153802255","https://openalex.org/W7216388352","https://openalex.org/W4417363089","https://openalex.org/W7133224318","https://openalex.org/W4386322317","https://openalex.org/W7119503992"],"abstract_inverted_index":{"Reliable":[0],"computer-vision-based":[1],"monitoring":[2,63],"of":[3,40,51,106,116,126,131,233,240],"floating":[4],"pollutants":[5],"requires":[6],"detectors":[7,270],"that":[8,294],"can":[9],"identify":[10],"small":[11,110,117,173],"objects,":[12],"accommodate":[13],"imbalanced":[14],"pollutant":[15,74],"categories,":[16,75],"and":[17,37,60,76,92,129,144,218,226,242,311],"operate":[18],"under":[19,314],"practical":[20],"computational":[21,93,312],"constraints.":[22],"However,":[23],"conventional":[24],"aggregate":[25,302],"detection":[26,83],"metrics":[27],"may":[28],"not":[29],"fully":[30],"characterize":[31],"model":[32],"performance":[33,90,168],"when":[34],"the":[35,135,140,150,154,162,190,195,222,237,243,260,265,269,279,315],"class":[36,99],"object-scale":[38,87],"compositions":[39],"dataset":[41,69,96],"partitions":[42],"differ":[43],"substantially.":[44],"This":[45],"study":[46],"presents":[47],"a":[48,68],"distribution-aware":[49],"evaluation":[50],"five":[52],"representative":[53],"object":[54,306],"detectors\u2014SSD300,":[55],"RetinaNet,":[56],"Faster":[57],"R-CNN,":[58],"YOLO26n,":[59],"RT-DETR-L\u2014for":[61],"floating-pollutant":[62,295],"on":[64,189,194],"water":[65],"surfaces.":[66],"Using":[67],"comprising":[70],"4274":[71],"images,":[72],"8":[73],"8606":[77],"annotated":[78],"instances,":[79,108],"we":[80],"evaluated":[81,271],"overall":[82],"accuracy,":[84,303],"class-wise":[85],"performance,":[86,305],"sensitivity,":[88],"validation-to-test":[89],"changes,":[91],"efficiency.":[94],"The":[95,114],"exhibited":[97],"pronounced":[98],"imbalance,":[100],"with":[101,213],"\u2018bottle\u2019":[102],"accounting":[103,123],"for":[104,124,172,175],"79.70%":[105],"all":[107,179],"while":[109,198,221],"objects":[111,118,177],"constituted":[112],"46.40%.":[113],"proportion":[115],"differed":[119],"markedly":[120],"across":[121,178],"partitions,":[122],"0.29%":[125],"validation":[127,196],"instances":[128],"66.19%":[130],"test":[132,136,191],"instances.":[133],"On":[134],"set,":[137,197],"RT-DETR-L":[138,159],"achieved":[139,161,264,278],"highest":[141,151,163,266,280],"mAP@0.50":[142,185],"(75.23%)":[143],"mAP@0.50:0.95":[145,188,202],"(45.73%),":[146],"whereas":[147,276],"YOLO26n":[148,263],"attained":[149],"macro-F1":[152,217],"at":[153,272,283],"predefined":[155],"confidence":[156],"threshold":[157],"(69.64%).":[158],"also":[160],"small-object":[164,227],"AP":[165,228],"(32.56%),":[166],"although":[167],"remained":[169,203],"substantially":[170],"lower":[171,187,230,285],"than":[174,193,231,254],"large":[176],"detectors.":[180],"All":[181],"models":[182],"showed":[183],"higher":[184,214],"but":[186],"set":[192],"their":[199],"ranking":[200],"by":[201],"unchanged.":[204],"Across":[205],"three":[206],"random":[207],"seeds,":[208],"class-balanced":[209],"sampling":[210],"was":[211],"associated":[212],"observed":[215,223,244],"mean":[216,224],"rare-class":[219],"recall,":[220],"mAP":[225],"were":[229],"those":[232],"standard":[234],"training.":[235],"Given":[236],"limited":[238],"number":[239],"runs":[241],"between-seed":[245],"variability,":[246],"these":[247],"differences":[248],"should":[249],"be":[250],"interpreted":[251],"descriptively":[252],"rather":[253],"as":[255],"statistically":[256],"established":[257],"effects.":[258],"Under":[259],"tested":[261,316],"configurations,":[262],"throughput":[267,282],"among":[268],"640":[273],"\u00d7":[274,287],"640,":[275],"SSD300":[277],"raw":[281],"its":[284],"300":[286,288],"input":[289],"resolution.":[290],"These":[291],"findings":[292],"demonstrate":[293],"detector":[296],"assessment":[297],"benefits":[298],"from":[299],"jointly":[300],"considering":[301],"category-level":[304],"scale,":[307],"partition-level":[308],"distribution":[309],"differences,":[310],"efficiency":[313],"configuration.":[317]},"counts_by_year":[],"updated_date":"2026-11-11T12:16:26.678958","created_date":"2026-11-10T00:00:00"}