{"id":"https://openalex.org/W7118184130","doi":"https://doi.org/10.1109/jiot.2026.3650951","title":"Depthwise-Attentive Hierarchical Cross-Modal Knowledge Distillation Network for Rail Surface Defect Detection","display_name":"Depthwise-Attentive Hierarchical Cross-Modal Knowledge Distillation Network for Rail Surface Defect Detection","publication_year":2026,"publication_date":"2026-01-05","ids":{"openalex":"https://openalex.org/W7118184130","doi":"https://doi.org/10.1109/jiot.2026.3650951"},"language":null,"primary_location":{"id":"doi:10.1109/jiot.2026.3650951","is_oa":false,"landing_page_url":"https://doi.org/10.1109/jiot.2026.3650951","pdf_url":null,"source":{"id":"https://openalex.org/S2480266640","display_name":"IEEE Internet of Things Journal","issn_l":"2327-4662","issn":["2327-4662","2372-2541"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Internet of Things Journal","raw_type":"journal-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/A5121976605","display_name":"Xu Zhou","orcid":null},"institutions":[{"id":"https://openalex.org/I55022517","display_name":"Heilongjiang University","ror":"https://ror.org/04zyhq975","country_code":"CN","type":"education","lineage":["https://openalex.org/I55022517"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xu Zhou","raw_affiliation_strings":["School of Computer and Big Data, Heilongjiang University, Harbin, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Computer and Big Data, Heilongjiang University, Harbin, China","institution_ids":["https://openalex.org/I55022517"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5121946767","display_name":"Xin Guan","orcid":null},"institutions":[{"id":"https://openalex.org/I55022517","display_name":"Heilongjiang University","ror":"https://ror.org/04zyhq975","country_code":"CN","type":"education","lineage":["https://openalex.org/I55022517"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xin Guan","raw_affiliation_strings":["School of Computer and Big Data, Heilongjiang University, Harbin, China"],"raw_orcid":"https://orcid.org/0000-0002-9129-327X","affiliations":[{"raw_affiliation_string":"School of Computer and Big Data, Heilongjiang University, Harbin, China","institution_ids":["https://openalex.org/I55022517"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5121939160","display_name":"Yu Peng","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yu Peng","raw_affiliation_strings":["State Grid Heilongjiang Electric Power Company Ltd., Harbin, China"],"raw_orcid":"https://orcid.org/0009-0006-3911-4488","affiliations":[{"raw_affiliation_string":"State Grid Heilongjiang Electric Power Company Ltd., Harbin, China","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5054390762","display_name":"Zhaogong Zhang","orcid":"https://orcid.org/0000-0002-1195-936X"},"institutions":[{"id":"https://openalex.org/I55022517","display_name":"Heilongjiang University","ror":"https://ror.org/04zyhq975","country_code":"CN","type":"education","lineage":["https://openalex.org/I55022517"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhaogong Zhang","raw_affiliation_strings":["School of Computer and Big Data, Heilongjiang University, Harbin, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Computer and Big Data, Heilongjiang University, Harbin, China","institution_ids":["https://openalex.org/I55022517"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5121943417","display_name":"Xiongjie Zhou","orcid":null},"institutions":[{"id":"https://openalex.org/I55022517","display_name":"Heilongjiang University","ror":"https://ror.org/04zyhq975","country_code":"CN","type":"education","lineage":["https://openalex.org/I55022517"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xiongjie Zhou","raw_affiliation_strings":["School of Computer and Big Data, Heilongjiang University, Harbin, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Computer and Big Data, Heilongjiang University, Harbin, China","institution_ids":["https://openalex.org/I55022517"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5008473103","display_name":"Hongyang Chen","orcid":"https://orcid.org/0000-0002-7626-0162"},"institutions":[{"id":"https://openalex.org/I4210123185","display_name":"Zhejiang Lab","ror":"https://ror.org/02m2h7991","country_code":"CN","type":"facility","lineage":["https://openalex.org/I4210123185"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Hongyang Chen","raw_affiliation_strings":["Zhejiang Laboratory, Hangzhou, Zhejiang, China"],"raw_orcid":"https://orcid.org/0000-0002-7626-0162","affiliations":[{"raw_affiliation_string":"Zhejiang Laboratory, Hangzhou, Zhejiang, China","institution_ids":["https://openalex.org/I4210123185"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5121465370","display_name":"Tomoaki Ohtsuki","orcid":null},"institutions":[{"id":"https://openalex.org/I203951103","display_name":"Keio University","ror":"https://ror.org/02kn6nx58","country_code":"JP","type":"education","lineage":["https://openalex.org/I203951103"]}],"countries":["JP"],"is_corresponding":false,"raw_author_name":"Tomoaki Ohtsuki","raw_affiliation_strings":["Department of Information and Computer Science, Keio University, Yokohama, Japan"],"raw_orcid":"https://orcid.org/0000-0003-3961-1426","affiliations":[{"raw_affiliation_string":"Department of Information and Computer Science, Keio University, Yokohama, Japan","institution_ids":["https://openalex.org/I203951103"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5121964510","display_name":"Zhu Han","orcid":null},"institutions":[{"id":"https://openalex.org/I44461941","display_name":"University of Houston","ror":"https://ror.org/048sx0r50","country_code":"US","type":"education","lineage":["https://openalex.org/I44461941"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Zhu Han","raw_affiliation_strings":["Department of Electrical and Computer Engineering, University of Houston, Houston, TX, USA"],"raw_orcid":"https://orcid.org/0000-0002-6606-5822","affiliations":[{"raw_affiliation_string":"Department of Electrical and Computer Engineering, University of Houston, Houston, TX, USA","institution_ids":["https://openalex.org/I44461941"]}]}],"institutions":[],"countries_distinct_count":3,"institutions_distinct_count":4,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.03449044,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"13","issue":"6","first_page":"10915","last_page":"10928"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.41929998993873596,"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":0.41929998993873596,"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/T11606","display_name":"Infrastructure Maintenance and Monitoring","score":0.28349998593330383,"subfield":{"id":"https://openalex.org/subfields/2205","display_name":"Civil and Structural 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/T10842","display_name":"Railway Engineering and Dynamics","score":0.16349999606609344,"subfield":{"id":"https://openalex.org/subfields/2210","display_name":"Mechanical Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.5631999969482422},{"id":"https://openalex.org/keywords/block","display_name":"Block (permutation group theory)","score":0.5026000142097473},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.4300000071525574},{"id":"https://openalex.org/keywords/flops","display_name":"FLOPS","score":0.4287000000476837},{"id":"https://openalex.org/keywords/limiting","display_name":"Limiting","score":0.397599995136261},{"id":"https://openalex.org/keywords/distillation","display_name":"Distillation","score":0.3944000005722046},{"id":"https://openalex.org/keywords/computational-complexity-theory","display_name":"Computational complexity theory","score":0.3937000036239624},{"id":"https://openalex.org/keywords/software-deployment","display_name":"Software deployment","score":0.3926999866962433}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8222000002861023},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.5631999969482422},{"id":"https://openalex.org/C2777210771","wikidata":"https://www.wikidata.org/wiki/Q4927124","display_name":"Block (permutation group theory)","level":2,"score":0.5026000142097473},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.46630001068115234},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.4300000071525574},{"id":"https://openalex.org/C3826847","wikidata":"https://www.wikidata.org/wiki/Q188768","display_name":"FLOPS","level":2,"score":0.4287000000476837},{"id":"https://openalex.org/C188198153","wikidata":"https://www.wikidata.org/wiki/Q1613840","display_name":"Limiting","level":2,"score":0.397599995136261},{"id":"https://openalex.org/C204030448","wikidata":"https://www.wikidata.org/wiki/Q101017","display_name":"Distillation","level":2,"score":0.3944000005722046},{"id":"https://openalex.org/C179799912","wikidata":"https://www.wikidata.org/wiki/Q205084","display_name":"Computational complexity theory","level":2,"score":0.3937000036239624},{"id":"https://openalex.org/C105339364","wikidata":"https://www.wikidata.org/wiki/Q2297740","display_name":"Software deployment","level":2,"score":0.3926999866962433},{"id":"https://openalex.org/C186967261","wikidata":"https://www.wikidata.org/wiki/Q5082128","display_name":"Mobile device","level":2,"score":0.3799999952316284},{"id":"https://openalex.org/C82990744","wikidata":"https://www.wikidata.org/wiki/Q166194","display_name":"RGB color model","level":2,"score":0.36399999260902405},{"id":"https://openalex.org/C52622490","wikidata":"https://www.wikidata.org/wiki/Q1026626","display_name":"Feature extraction","level":2,"score":0.33980000019073486},{"id":"https://openalex.org/C77618280","wikidata":"https://www.wikidata.org/wiki/Q1155772","display_name":"Scheme (mathematics)","level":2,"score":0.32820001244544983},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.32409998774528503},{"id":"https://openalex.org/C151201525","wikidata":"https://www.wikidata.org/wiki/Q177239","display_name":"Limit (mathematics)","level":2,"score":0.3172999918460846},{"id":"https://openalex.org/C41608201","wikidata":"https://www.wikidata.org/wiki/Q980509","display_name":"Embedding","level":2,"score":0.30970001220703125},{"id":"https://openalex.org/C111335779","wikidata":"https://www.wikidata.org/wiki/Q3454686","display_name":"Reduction (mathematics)","level":2,"score":0.30169999599456787},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.30059999227523804},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.28999999165534973},{"id":"https://openalex.org/C175154964","wikidata":"https://www.wikidata.org/wiki/Q380077","display_name":"Task analysis","level":3,"score":0.2822999954223633},{"id":"https://openalex.org/C67186912","wikidata":"https://www.wikidata.org/wiki/Q367664","display_name":"Data modeling","level":2,"score":0.2623000144958496},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.25859999656677246},{"id":"https://openalex.org/C84685590","wikidata":"https://www.wikidata.org/wiki/Q1540472","display_name":"Knowledge engineering","level":2,"score":0.2529999911785126},{"id":"https://openalex.org/C83665646","wikidata":"https://www.wikidata.org/wiki/Q42139305","display_name":"Feature vector","level":2,"score":0.25209999084472656}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/jiot.2026.3650951","is_oa":false,"landing_page_url":"https://doi.org/10.1109/jiot.2026.3650951","pdf_url":null,"source":{"id":"https://openalex.org/S2480266640","display_name":"IEEE Internet of Things Journal","issn_l":"2327-4662","issn":["2327-4662","2372-2541"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Internet of Things Journal","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G544269053","display_name":null,"funder_award_id":"SGHL0000JYJS2201819","funder_id":"https://openalex.org/F4320326707","funder_display_name":"State Grid Corporation of China"}],"funders":[{"id":"https://openalex.org/F4320326707","display_name":"State Grid Corporation of China","ror":"https://ror.org/05twwhs70"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":39,"referenced_works":["https://openalex.org/W20683899","https://openalex.org/W1976409045","https://openalex.org/W1993713494","https://openalex.org/W2620175222","https://openalex.org/W3097053213","https://openalex.org/W3106583357","https://openalex.org/W3109574507","https://openalex.org/W3114152269","https://openalex.org/W3122006940","https://openalex.org/W3135874576","https://openalex.org/W3210144980","https://openalex.org/W3214526888","https://openalex.org/W4206670089","https://openalex.org/W4214696292","https://openalex.org/W4220849367","https://openalex.org/W4281391985","https://openalex.org/W4283809560","https://openalex.org/W4285120968","https://openalex.org/W4295424676","https://openalex.org/W4312226907","https://openalex.org/W4313160444","https://openalex.org/W4315606099","https://openalex.org/W4316038422","https://openalex.org/W4378977233","https://openalex.org/W4381936735","https://openalex.org/W4382450424","https://openalex.org/W4386275800","https://openalex.org/W4387469824","https://openalex.org/W4389664527","https://openalex.org/W4389950586","https://openalex.org/W4390431338","https://openalex.org/W4392745066","https://openalex.org/W4394951205","https://openalex.org/W4399376243","https://openalex.org/W4405974332","https://openalex.org/W4409233645","https://openalex.org/W4411171776","https://openalex.org/W4413136983","https://openalex.org/W4414908645"],"related_works":[],"abstract_inverted_index":{"Accurate":[0],"detection":[1,39],"of":[2,179],"surface":[3],"defects":[4],"on":[5,19,160,214,229],"railway":[6,12],"tracks":[7],"is":[8,114,181],"critical":[9],"for":[10,60,109,227],"safe":[11],"operation.":[13],"Most":[14],"existing":[15],"models":[16,48],"rely":[17],"solely":[18],"Red\u2013Green\u2013Blue":[20,43],"(RGB)":[21],"images,":[22],"limiting":[23],"their":[24],"ability":[25],"to":[26,97,127,140,154,175,186,197],"capture":[27,98],"structural":[28,152],"information.":[29],"Incorporating":[30],"depth":[31,104],"features":[32],"provides":[33,221],"richer":[34],"spatial":[35],"cues,":[36],"significantly":[37],"improving":[38],"accuracy.":[40,157],"However,":[41],"current":[42],"and":[44,54,88,143,151,189,210,224],"Depth":[45],"(RGB-D)":[46],"dual-stream":[47],"suffer":[49],"from":[50,183,194],"high":[51],"computational":[52,191],"complexity":[53],"hardware":[55],"dependencies,":[56],"making":[57],"them":[58],"impractical":[59],"real-world":[61,233],"deployment.":[62],"To":[63],"address":[64],"these":[65],"limitations,":[66],"we":[67],"propose":[68],"DAHNet,":[69],"an":[70],"asymmetric":[71],"knowledge":[72,145],"distillation":[73,134],"model":[74,122,169,202,220],"with":[75,102,136],"a":[76,90,103,118,132,222],"teacher\u2013student":[77],"architecture.":[78],"DAHNet-T":[79],"serves":[80],"as":[81],"the":[82,115,161,177,190,219],"teacher":[83],"network,":[84,117],"taking":[85],"RGB-D":[86],"inputs":[87],"integrating":[89],"cross-modal":[91,111],"attention":[92],"feature":[93,105],"enhancement":[94],"(CAFE)":[95],"module":[96],"contextual":[99],"information,":[100],"along":[101],"interaction":[106],"block":[107],"(DFIB)":[108],"efficient":[110],"fusion.":[112],"DAHNet-S":[113],"student":[116],"lightweight":[119,223],"single-stream":[120],"RGB":[121],"employing":[123],"depthwise":[124],"separable":[125],"convolutions":[126],"reduce":[128],"computation.":[129],"We":[130],"introduce":[131],"multi-level":[133],"strategy":[135],"dynamic":[137],"temperature":[138],"scaling":[139],"balance":[141],"coarse-grained":[142],"fine-grained":[144],"transfer,":[146],"while":[147],"incorporating":[148],"contrastive":[149],"learning":[150],"loss":[153],"improve":[155],"pixel-level":[156],"Extensive":[158],"experiments":[159],"NEU":[162],"RSDDS-AUG":[163],"dataset":[164],"demonstrate":[165],"that":[166],"our":[167],"distilled":[168],"DAHNet-KD":[170],"outperforms":[171],"state-of-the-art":[172],"methods.":[173],"Compared":[174],"DAHNet-T,":[176],"number":[178],"parameters":[180],"reduced":[182],"87.72":[184],"MParams":[185],"13.97":[187],"MParams,":[188],"cost":[192],"decreases":[193],"19.79":[195],"GFLOPs":[196],"5.41":[198],"GFLOPs.":[199],"The":[200],"proposed":[201],"achieves":[203],"superior":[204],"performance":[205],"across":[206],"various":[207],"evaluation":[208],"metrics":[209],"also":[211],"generalizes":[212],"well":[213],"other":[215],"public":[216],"datasets.":[217],"Therefore,":[218],"high-accuracy":[225],"solution":[226],"deployment":[228],"mobile":[230],"devices":[231],"in":[232],"industrial":[234],"scenarios.":[235]},"counts_by_year":[],"updated_date":"2026-03-27T05:58:40.876381","created_date":"2026-01-05T00:00:00"}