{"id":"https://openalex.org/W4387802894","doi":"https://doi.org/10.1109/igarss52108.2023.10281628","title":"MSAU-Net: Road Extraction Based on Multi-Headed Self-Attention Mechanism and U-Net with High Resolution Remote Sensing Images","display_name":"MSAU-Net: Road Extraction Based on Multi-Headed Self-Attention Mechanism and U-Net with High Resolution Remote Sensing Images","publication_year":2023,"publication_date":"2023-07-16","ids":{"openalex":"https://openalex.org/W4387802894","doi":"https://doi.org/10.1109/igarss52108.2023.10281628"},"language":"en","primary_location":{"id":"doi:10.1109/igarss52108.2023.10281628","is_oa":false,"landing_page_url":"https://doi.org/10.1109/igarss52108.2023.10281628","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IGARSS 2023 - 2023 IEEE International Geoscience and Remote Sensing Symposium","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5100648591","display_name":"Hang Yu","orcid":"https://orcid.org/0000-0002-8869-6166"},"institutions":[{"id":"https://openalex.org/I149594827","display_name":"Xidian University","ror":"https://ror.org/05s92vm98","country_code":"CN","type":"education","lineage":["https://openalex.org/I149594827"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Hang Yu","raw_affiliation_strings":["Xidian University,School of Aerospace Science and Technology,Xi’an,China,710126"],"affiliations":[{"raw_affiliation_string":"Xidian University,School of Aerospace Science and Technology,Xi’an,China,710126","institution_ids":["https://openalex.org/I149594827"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5016667520","display_name":"Yuru Guo","orcid":"https://orcid.org/0009-0003-5252-7939"},"institutions":[{"id":"https://openalex.org/I149594827","display_name":"Xidian University","ror":"https://ror.org/05s92vm98","country_code":"CN","type":"education","lineage":["https://openalex.org/I149594827"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yuru Guo","raw_affiliation_strings":["Xidian University,School of Aerospace Science and Technology,Xi’an,China,710126"],"affiliations":[{"raw_affiliation_string":"Xidian University,School of Aerospace Science and Technology,Xi’an,China,710126","institution_ids":["https://openalex.org/I149594827"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5068610174","display_name":"Zhiheng Liu","orcid":"https://orcid.org/0000-0002-2794-3557"},"institutions":[{"id":"https://openalex.org/I149594827","display_name":"Xidian University","ror":"https://ror.org/05s92vm98","country_code":"CN","type":"education","lineage":["https://openalex.org/I149594827"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhiheng Liu","raw_affiliation_strings":["Xidian University,School of Aerospace Science and Technology,Xi’an,China,710126"],"affiliations":[{"raw_affiliation_string":"Xidian University,School of Aerospace Science and Technology,Xi’an,China,710126","institution_ids":["https://openalex.org/I149594827"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5049798236","display_name":"Suiping Zhou","orcid":"https://orcid.org/0000-0003-0914-066X"},"institutions":[{"id":"https://openalex.org/I149594827","display_name":"Xidian University","ror":"https://ror.org/05s92vm98","country_code":"CN","type":"education","lineage":["https://openalex.org/I149594827"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Suiping Zhou","raw_affiliation_strings":["Xidian University,School of Aerospace Science and Technology,Xi’an,China,710126"],"affiliations":[{"raw_affiliation_string":"Xidian University,School of Aerospace Science and Technology,Xi’an,China,710126","institution_ids":["https://openalex.org/I149594827"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5049255145","display_name":"Chenyang Li","orcid":"https://orcid.org/0000-0002-5371-5574"},"institutions":[{"id":"https://openalex.org/I149594827","display_name":"Xidian University","ror":"https://ror.org/05s92vm98","country_code":"CN","type":"education","lineage":["https://openalex.org/I149594827"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Chenyang Li","raw_affiliation_strings":["Xidian University,School of Aerospace Science and Technology,Xi’an,China,710126"],"affiliations":[{"raw_affiliation_string":"Xidian University,School of Aerospace Science and Technology,Xi’an,China,710126","institution_ids":["https://openalex.org/I149594827"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100385495","display_name":"Wenjie Zhang","orcid":"https://orcid.org/0000-0001-5949-0268"},"institutions":[{"id":"https://openalex.org/I149594827","display_name":"Xidian University","ror":"https://ror.org/05s92vm98","country_code":"CN","type":"education","lineage":["https://openalex.org/I149594827"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Wenjie Zhang","raw_affiliation_strings":["Xidian University,School of Aerospace Science and Technology,Xi’an,China,710126"],"affiliations":[{"raw_affiliation_string":"Xidian University,School of Aerospace Science and Technology,Xi’an,China,710126","institution_ids":["https://openalex.org/I149594827"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5115076809","display_name":"Wenjuan Qi","orcid":"https://orcid.org/0009-0001-4268-1612"},"institutions":[{"id":"https://openalex.org/I149594827","display_name":"Xidian University","ror":"https://ror.org/05s92vm98","country_code":"CN","type":"education","lineage":["https://openalex.org/I149594827"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Wenjuan Qi","raw_affiliation_strings":["Xidian University,School of Aerospace Science and Technology,Xi’an,China,710126"],"affiliations":[{"raw_affiliation_string":"Xidian University,School of Aerospace Science and Technology,Xi’an,China,710126","institution_ids":["https://openalex.org/I149594827"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":7,"corresponding_author_ids":["https://openalex.org/A5100648591"],"corresponding_institution_ids":["https://openalex.org/I149594827"],"apc_list":null,"apc_paid":null,"fwci":0.6111,"has_fulltext":false,"cited_by_count":3,"citation_normalized_percentile":{"value":0.67321644,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":90,"max":96},"biblio":{"volume":null,"issue":null,"first_page":"6898","last_page":"6900"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T13282","display_name":"Automated Road and Building Extraction","score":0.9998999834060669,"subfield":{"id":"https://openalex.org/subfields/2212","display_name":"Ocean Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T13282","display_name":"Automated Road and Building Extraction","score":0.9998999834060669,"subfield":{"id":"https://openalex.org/subfields/2212","display_name":"Ocean Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11164","display_name":"Remote Sensing and LiDAR Applications","score":0.995199978351593,"subfield":{"id":"https://openalex.org/subfields/2305","display_name":"Environmental Engineering"},"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/T10331","display_name":"Video Surveillance and Tracking Methods","score":0.9937999844551086,"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/computer-science","display_name":"Computer science","score":0.7619220018386841},{"id":"https://openalex.org/keywords/upsampling","display_name":"Upsampling","score":0.7128223180770874},{"id":"https://openalex.org/keywords/block","display_name":"Block (permutation group theory)","score":0.6236678957939148},{"id":"https://openalex.org/keywords/canny-edge-detector","display_name":"Canny edge detector","score":0.6182934641838074},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.6003448963165283},{"id":"https://openalex.org/keywords/computer-vision","display_name":"Computer vision","score":0.5254230499267578},{"id":"https://openalex.org/keywords/enhanced-data-rates-for-gsm-evolution","display_name":"Enhanced Data Rates for GSM Evolution","score":0.5050826668739319},{"id":"https://openalex.org/keywords/net","display_name":"Net (polyhedron)","score":0.45264697074890137},{"id":"https://openalex.org/keywords/feature-extraction","display_name":"Feature extraction","score":0.4454447329044342},{"id":"https://openalex.org/keywords/edge-detection","display_name":"Edge detection","score":0.39841240644454956},{"id":"https://openalex.org/keywords/remote-sensing","display_name":"Remote sensing","score":0.34658366441726685},{"id":"https://openalex.org/keywords/image","display_name":"Image (mathematics)","score":0.31792134046554565},{"id":"https://openalex.org/keywords/image-processing","display_name":"Image processing","score":0.3057307004928589},{"id":"https://openalex.org/keywords/geography","display_name":"Geography","score":0.11761745810508728},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.09229746460914612}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7619220018386841},{"id":"https://openalex.org/C110384440","wikidata":"https://www.wikidata.org/wiki/Q1143270","display_name":"Upsampling","level":3,"score":0.7128223180770874},{"id":"https://openalex.org/C2777210771","wikidata":"https://www.wikidata.org/wiki/Q4927124","display_name":"Block (permutation group theory)","level":2,"score":0.6236678957939148},{"id":"https://openalex.org/C14705441","wikidata":"https://www.wikidata.org/wiki/Q597183","display_name":"Canny edge detector","level":5,"score":0.6182934641838074},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6003448963165283},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.5254230499267578},{"id":"https://openalex.org/C162307627","wikidata":"https://www.wikidata.org/wiki/Q204833","display_name":"Enhanced Data Rates for GSM Evolution","level":2,"score":0.5050826668739319},{"id":"https://openalex.org/C14166107","wikidata":"https://www.wikidata.org/wiki/Q253829","display_name":"Net (polyhedron)","level":2,"score":0.45264697074890137},{"id":"https://openalex.org/C52622490","wikidata":"https://www.wikidata.org/wiki/Q1026626","display_name":"Feature extraction","level":2,"score":0.4454447329044342},{"id":"https://openalex.org/C193536780","wikidata":"https://www.wikidata.org/wiki/Q1513153","display_name":"Edge detection","level":4,"score":0.39841240644454956},{"id":"https://openalex.org/C62649853","wikidata":"https://www.wikidata.org/wiki/Q199687","display_name":"Remote sensing","level":1,"score":0.34658366441726685},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.31792134046554565},{"id":"https://openalex.org/C9417928","wikidata":"https://www.wikidata.org/wiki/Q1070689","display_name":"Image processing","level":3,"score":0.3057307004928589},{"id":"https://openalex.org/C205649164","wikidata":"https://www.wikidata.org/wiki/Q1071","display_name":"Geography","level":0,"score":0.11761745810508728},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.09229746460914612},{"id":"https://openalex.org/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/igarss52108.2023.10281628","is_oa":false,"landing_page_url":"https://doi.org/10.1109/igarss52108.2023.10281628","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IGARSS 2023 - 2023 IEEE International Geoscience and Remote Sensing Symposium","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"display_name":"Sustainable cities and communities","score":0.8199999928474426,"id":"https://metadata.un.org/sdg/11"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":18,"referenced_works":["https://openalex.org/W1901129140","https://openalex.org/W2395611524","https://openalex.org/W2505004417","https://openalex.org/W2623331213","https://openalex.org/W2891954420","https://openalex.org/W2898504016","https://openalex.org/W2907750714","https://openalex.org/W2942520561","https://openalex.org/W2962731543","https://openalex.org/W2987603566","https://openalex.org/W3047362324","https://openalex.org/W3096627121","https://openalex.org/W3150573203","https://openalex.org/W4206830255","https://openalex.org/W4221145217","https://openalex.org/W4251530854","https://openalex.org/W6762000186","https://openalex.org/W6762440033"],"related_works":["https://openalex.org/W3169126738","https://openalex.org/W1994279415","https://openalex.org/W2558559991","https://openalex.org/W1986338341","https://openalex.org/W2545065926","https://openalex.org/W2980082320","https://openalex.org/W2387510934","https://openalex.org/W2015446345","https://openalex.org/W3004771112","https://openalex.org/W3132644099"],"abstract_inverted_index":{"ObjectWith":[0],"social":[1],"and":[2,6,24,73,105,129,136,147,161,193,205,236,328,334,358,414,436,443,455,484,515,520,530,539,549,569,581,589,599,609,617,627,642,702,709,726],"economic":[3],"development,":[4],"more":[5,7,148,291,316,640],"areas":[8],"are":[9,34,278,526,545,605,623],"evolving":[10],"toward":[11],"modern":[12],"cities":[13],"[1].":[14],"Road":[15,662,685],"information":[16,49,119,235,247,315,389,419],"is":[17,89,120,170,213,224,242,508,639],"critical":[18],"for":[19,43,63,134,396,407,465,665,688],"urban":[20],"planning,":[21],"traffic":[22],"navigation":[23],"GIS":[25],"data":[26,31],"updates":[27],"[2].":[28],"Traditional":[29],"manual":[30],"collection":[32],"methods":[33],"inefficient.":[35],"The":[36,115,168,209,219,517,537,596],"advancement":[37],"of":[38,47,80,92,152,211,221,325,343,351,411,432,441,462,502,522,541,583,601,619,649,671,721,723,731],"remote":[39,53,65,85,408],"sensing":[40,54,66,86,409],"technology":[41],"allows":[42],"the":[44,95,100,106,123,132,143,174,183,189,222,228,233,238,249,258,272,276,281,284,299,321,326,330,339,363,379,397,429,439,446,459,463,498,503,523,542,560,571,592,602,620,658,680,696,714,719,724,729],"automatic":[45],"extraction":[46,61,79,323,341,500,651,663,673,686,705],"road":[48,60,116,144,322,340,466,499,650,672,704],"from":[50,280],"high":[51,83],"resolution":[52,84],"images":[55,67,410],"[3]-[7].MethodThis":[56],"paper":[57,383,469,507],"proposes":[58],"a":[59,159,392,423],"model":[62,133,725],"high-resolution":[64],"based":[68],"on":[69,591,657,679,706],"multi-headed":[70,164,273],"self-retaining":[71],"mechanism":[72],"U-Net":[74,101,354],"(MSAU-Net)":[75],"to":[76,131,141,178,187,201,227,244,267,283,298,349,376,390,535,554,614,632,728],"achieve":[77,700],"accurate":[78,641,701],"roads":[81,336,435,682,690,711],"in":[82,112,199,248,371,381,452,495,505,557,566],"images.":[87],"MSAU-Net":[88,154],"made":[90],"up":[91],"three":[93,261],"components:":[94],"Canny":[96,124,175],"edge":[97,117,125,176,430],"detection":[98,126],"operator,":[99],"base":[102],"network":[103,155,346,464],"architecture,":[104],"multi-head":[107],"self-attention":[108,165,229,274],"module,":[109,230,275],"as":[110,490],"shown":[111,494,556],"Fig.":[113,372,496,558],"1.":[114],"feature":[118,256,418],"extracted":[121],"by":[122,528,547,607,625],"operator":[127,177],"[8]":[128],"inputted":[130],"training":[135],"learning":[137],"via":[138],"convolutional":[139],"operation":[140],"make":[142],"edges":[145],"smoother":[146],"accurate.Fig.":[149],"1.The":[150],"architecture":[151,270],"MSAU-Net.The":[153],"includes":[156],"an":[157,162,575],"encoder,":[158],"decoder,":[160],"intermediate":[163],"module":[166,186],"M-SABlock.":[167],"image":[169],"first":[171],"passed":[172],"through":[173,182,195,271,420],"detect":[179],"edges,":[180],"then":[181,225],"encoding":[184,250],"convolution":[185],"extract":[188,245,313,428],"image's":[190],"underlying":[191],"features,":[192],"finally":[194],"two":[196],"downsampling":[197,217,223,266],"blocks":[198],"sequence":[200],"reduce":[202,438,718],"space":[203],"size":[204],"obtain":[206],"high-level":[207],"features.":[208],"number":[210,720],"channels":[212],"doubled":[214],"after":[215],"each":[216],"block.":[218],"result":[220],"fed":[226],"which":[231],"summarizes":[232],"global":[234,292,309,314,387,393],"produces":[237],"encoder":[239,282],"output.":[240],"Convolution":[241],"used":[243],"local":[246],"module.":[251],"To":[252],"recover":[253],"its":[254],"corresponding":[255],"images,":[257],"decoder":[259,300],"employs":[260],"upsampling":[262,268],"blocks.":[263],"By":[264],"implementing":[265],"skip":[269],"features":[277,288,305],"copied":[279],"decoder.":[285,398],"Because":[286],"lower-level":[287],"already":[289],"contain":[290],"information,":[293,310],"they":[294],"can":[295,312,319,368,384,401,426,699],"proceed":[296],"directly":[297],"side;":[301],"however,":[302],"because":[303],"higher-level":[304],"have":[306],"relatively":[307],"less":[308],"M-SABlock":[311],"effectively.":[317],"This":[318],"improve":[320],"accuracy":[324],"model.Result":[327],"DiscussionOn":[329],"CHN6-CUG":[331,659,666,707],"dataset":[332,337,708],"[9]":[333],"Massachusetts":[335,681,689,710],"[10],":[338],"performance":[342],"this":[344,382,468,506],"paper's":[345],"was":[347],"compared":[348,375,534,553,631],"that":[350,636,695],"FCN":[352],"[11],":[353],"[12],":[355],"ResU-Net":[356,514],"[13],":[357],"SAU-Net":[359],"[14]":[360],"networks":[361],"using":[362],"same":[364],"experimental":[365,692],"settings.":[366],"As":[367,422,493],"be":[369],"seen":[370],"2(a,c),":[373],"when":[374,533,552,612,630],"other":[377,450,564],"algorithms,":[378],"algorithm":[380,448,504,525,544,562,604,622,638,698],"effectively":[385],"combine":[386],"contextual":[388],"provide":[391],"information-guided":[394],"flow":[395],"Simultaneously,":[399],"it":[400,425],"dynamically":[402],"select":[403],"suitable":[404],"sensory":[405],"fields":[406],"different":[412,654,676],"sizes":[413],"better":[415,427,510],"integrate":[416],"multi-scale":[417],"self-learning.":[421],"result,":[424],"contour":[431],"blurred":[433],"complex":[434],"significantly":[437,509],"phenomenon":[440],"over-":[442],"under-extraction.":[444],"Meanwhile,":[445],"proposed":[447,524,543,561,603,621,697],"outperforms":[449,563],"algorithms":[451,565,656,678],"both":[453],"overall":[454],"detail":[456],"performance.To":[457],"realize":[458],"quantitative":[460],"analysis":[461],"extraction,":[467],"uses":[470],"Hausdorff":[471],"distance":[472],"(HD)":[473],"[15],":[474],"Precision":[475],"(PRE),":[476],"Dice":[477],"similarity":[478],"coefficient":[479],"(DSC)":[480],"[16],":[481],"Sensitivity":[482],"(SEN)":[483],"Mean":[485],"intersection":[486],"over":[487],"union":[488],"(MIOU)":[489],"evaluation":[491],"metrics.":[492],"2(b),":[497],"effect":[501],"than":[511],"FCN,":[512],"U-Net,":[513],"SAU-Net.":[516],"average":[518,576,597],"MIOU":[519,582,616],"DSC":[521,600],"improved":[527,546,606,624],"6.96%":[529],"7.16%,":[531],"respectively,":[532,551,611,629],"ResU-Net.":[536],"PRE":[538,598],"SEN":[540,580,618],"9.08%":[548],"6.56%,":[550],"SAU-Net.As":[555],"2(d),":[559],"these":[567],"metrics":[568,664,687],"achieves":[570],"best":[572],"results":[573,652,674,693],"with":[574,653,675],"HD,":[577],"PRE,":[578],"DSC,":[579],"2.209":[584],"cm,":[585],"90.12%,":[586],"91.26%,":[587],"89.38%,":[588],"88.74%":[590],"test":[593],"set,":[594],"respectively.":[595],"7.78%":[608],"8.14%,":[610],"campared":[613],"FCN.The":[615],"5.58%":[626],"7.41%,":[628],"U-Net.":[633],"It":[634],"indicates":[635],"our":[637],"efficient.Fig.":[643],"2.Experiments":[644],"result.":[645],"(a)":[646],"Visual":[647,669],"comparisons":[648,670],"comparing":[655,677],"dataset,":[660,683],"(b)":[661],"dataset.":[667,712],"(c)":[668],"(d)":[684],"dataset.ConclusionThe":[691],"show":[694],"effective":[703],"In":[713],"future,":[715],"we":[716],"will":[717],"parameters":[722],"devote":[727],"research":[730],"lightweight":[732],"model.":[733]},"counts_by_year":[{"year":2025,"cited_by_count":2},{"year":2024,"cited_by_count":1}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}