Deep Multitask Learning for Railway Track Inspection
Railroad tracks need to be periodically inspected and monitored to ensure safe transportation. Automated track inspection using computer vision and pattern recognition methods has recently shown the potential to improve safety by allowing for more frequent inspections while reducing human errors. Ac...
Saved in:
| Published in | IEEE transactions on intelligent transportation systems Vol. 18; no. 1; pp. 153 - 164 |
|---|---|
| Main Authors | , , |
| Format | Journal Article |
| Language | English |
| Published |
New York
IEEE
01.01.2017
The Institute of Electrical and Electronics Engineers, Inc. (IEEE) |
| Subjects | |
| Online Access | Get full text |
Cover
Loading…
| Abstract | Railroad tracks need to be periodically inspected and monitored to ensure safe transportation. Automated track inspection using computer vision and pattern recognition methods has recently shown the potential to improve safety by allowing for more frequent inspections while reducing human errors. Achieving full automation is still very challenging due to the number of different possible failure modes, as well as the broad range of image variations that can potentially trigger false alarms. In addition, the number of defective components is very small, so not many training examples are available for the machine to learn a robust anomaly detector. In this paper, we show that detection performance can be improved by combining multiple detectors within a multitask learning framework. We show that this approach results in improved accuracy for detecting defects on railway ties and fasteners. |
|---|---|
| AbstractList | Railroad tracks need to be periodically inspected and monitored to ensure safe transportation. Automated track inspection using computer vision and pattern recognition methods has recently shown the potential to improve safety by allowing for more frequent inspections while reducing human errors. Achieving full automation is still very challenging due to the number of different possible failure modes, as well as the broad range of image variations that can potentially trigger false alarms. In addition, the number of defective components is very small, so not many training examples are available for the machine to learn a robust anomaly detector. In this paper, we show that detection performance can be improved by combining multiple detectors within a multitask learning framework. We show that this approach results in improved accuracy for detecting defects on railway ties and fasteners. |
| Author | Patel, Vishal M. Gibert, Xavier Chellappa, Rama |
| Author_xml | – sequence: 1 givenname: Xavier surname: Gibert fullname: Gibert, Xavier email: xgibert@google.com organization: Univ. of Maryland, Mountain View, CA, USA – sequence: 2 givenname: Vishal M. surname: Patel fullname: Patel, Vishal M. email: vishal.m.patel@rutgers.edu organization: Dept. of Electr. Eng., Rutgers Univ., Piscataway, NJ, USA – sequence: 3 givenname: Rama surname: Chellappa fullname: Chellappa, Rama email: rama@umiacs.umd.edu organization: Dept. of Electr. Eng., Univ. of Maryland Inst. for Adv. Comput. Studies, College Park, MD, USA |
| BookMark | eNo9kE1Lw0AQhhepYFv9AeIl4Dl1djf7kaPUr0JE0HhetptZSVuTuJsi_fcmtHia4eV5Z-CZkUnTNkjINYUFpZDflavyY8GAygUTUiuhz8iUCqFTGLLJuLMszUHABZnFuBnSTFA6JdkDYpe87nd93du4TQq0oambr8S3IXm39e7XHpIyWLdNVk3s0PV121ySc293Ea9Oc04-nx7L5UtavD2vlvdF6nhG-9TloBkD7yurgbOcU1S-UpWTvModq1AAMs69VBL4mmMFSjOveM7Wzopc8Tm5Pd7tQvuzx9ibTbsPzfDSUC2EZKAEHyh6pFxoYwzoTRfqbxsOhoIZ5ZhRjhnlmJOcoXNz7NSI-M8rAZJSxf8AG-ZgiQ |
| CODEN | ITISFG |
| CitedBy_id | crossref_primary_10_1039_D0NA00388C crossref_primary_10_1109_TIM_2023_3243673 crossref_primary_10_1049_ipr2_13020 crossref_primary_10_1007_s13349_023_00695_4 crossref_primary_10_1016_j_jmst_2020_04_033 crossref_primary_10_1016_j_autcon_2023_104854 crossref_primary_10_1016_j_ijtst_2018_11_001 crossref_primary_10_1109_JSEN_2017_2761858 crossref_primary_10_1007_s11042_021_10809_z crossref_primary_10_1016_j_compind_2020_103303 crossref_primary_10_3390_s23187776 crossref_primary_10_1109_TIM_2018_2871353 crossref_primary_10_1109_TASE_2019_2900170 crossref_primary_10_1088_1361_6501_ac31ca crossref_primary_10_1109_TIE_2023_3250766 crossref_primary_10_1109_JIOT_2022_3162295 crossref_primary_10_1109_TII_2024_3370330 crossref_primary_10_1088_1757_899X_750_1_012219 crossref_primary_10_1016_j_icte_2017_11_010 crossref_primary_10_1016_j_measurement_2021_110240 crossref_primary_10_1109_JSEN_2020_3009194 crossref_primary_10_1109_ACCESS_2020_2984264 crossref_primary_10_3390_s22197275 crossref_primary_10_1080_19439962_2019_1572681 crossref_primary_10_3390_s23146457 crossref_primary_10_1155_2023_1832501 crossref_primary_10_3390_s22176419 crossref_primary_10_1109_MITS_2019_2903540 crossref_primary_10_1016_j_measurement_2019_107332 crossref_primary_10_1049_hve2_12210 crossref_primary_10_1007_s00521_021_06280_4 crossref_primary_10_1109_ACCESS_2024_3355034 crossref_primary_10_31590_ejosat_1029905 crossref_primary_10_1007_s10489_021_02981_4 crossref_primary_10_3390_s19214738 crossref_primary_10_1109_TIM_2022_3162615 crossref_primary_10_2339_politeknik_693223 crossref_primary_10_1038_s41598_022_14971_8 crossref_primary_10_1109_TITS_2021_3131637 crossref_primary_10_1111_mice_12625 crossref_primary_10_1177_1729881418773943 crossref_primary_10_1109_JSEN_2019_2911015 crossref_primary_10_3390_app13106047 crossref_primary_10_1007_s40747_023_01219_9 crossref_primary_10_1016_j_culher_2019_02_002 crossref_primary_10_1016_j_eswa_2022_117946 crossref_primary_10_1109_TITS_2023_3334290 crossref_primary_10_3390_app122312044 crossref_primary_10_1109_TITS_2020_3020287 crossref_primary_10_3390_s22218399 crossref_primary_10_3390_sym14061100 crossref_primary_10_1109_ACCESS_2019_2962072 crossref_primary_10_1155_2018_4832972 crossref_primary_10_1109_TIM_2024_3417537 crossref_primary_10_1109_TITS_2023_3331901 crossref_primary_10_1142_S0218001422520048 crossref_primary_10_1007_s44163_024_00127_2 crossref_primary_10_1111_mice_13004 crossref_primary_10_1016_j_measurement_2021_110613 crossref_primary_10_1007_s12599_022_00745_z crossref_primary_10_1016_j_measurement_2021_109742 crossref_primary_10_1109_TIM_2018_2868490 crossref_primary_10_1016_j_neucom_2019_01_036 crossref_primary_10_1109_JSEN_2021_3133280 crossref_primary_10_1016_j_ifacol_2018_07_017 crossref_primary_10_1016_j_trc_2018_02_019 crossref_primary_10_1109_JSEN_2023_3339883 crossref_primary_10_1016_j_isprsjprs_2023_04_011 crossref_primary_10_1109_JSEN_2021_3062021 crossref_primary_10_1016_j_trc_2018_05_007 crossref_primary_10_1109_ACCESS_2020_3007603 crossref_primary_10_1109_TNNLS_2021_3105284 crossref_primary_10_1016_j_compind_2023_103909 crossref_primary_10_1007_s11440_021_01266_x crossref_primary_10_3390_s21144710 crossref_primary_10_1002_inf2_12028 crossref_primary_10_1109_TIM_2024_3370748 crossref_primary_10_3390_app14093573 crossref_primary_10_1109_TCCN_2021_3118456 crossref_primary_10_1109_ACCESS_2020_2997946 crossref_primary_10_1109_JIOT_2021_3057835 crossref_primary_10_1109_TIM_2020_3006324 crossref_primary_10_1155_2021_8956164 crossref_primary_10_1111_mice_13109 crossref_primary_10_1016_j_neucom_2018_10_107 crossref_primary_10_1109_ACCESS_2019_2955753 crossref_primary_10_1109_ACCESS_2019_2950985 crossref_primary_10_3390_s20051367 crossref_primary_10_7717_peerj_cs_1727 crossref_primary_10_1109_TIM_2017_2775345 crossref_primary_10_1109_TITS_2017_2749981 crossref_primary_10_3390_s20144042 crossref_primary_10_1109_TITS_2022_3214121 crossref_primary_10_1088_1757_899X_1059_1_012062 crossref_primary_10_1364_PRJ_388253 crossref_primary_10_1016_j_engfailanal_2020_104906 crossref_primary_10_4236_jtts_2024_142013 crossref_primary_10_1109_MIE_2019_2952165 crossref_primary_10_1007_s10044_019_00781_8 crossref_primary_10_1016_j_measen_2024_101157 crossref_primary_10_1109_TII_2017_2774242 crossref_primary_10_1109_TIM_2020_2990541 crossref_primary_10_1088_1742_6596_1684_1_012030 crossref_primary_10_1109_TITS_2021_3136215 crossref_primary_10_1016_j_ijfatigue_2023_107864 crossref_primary_10_4018_IJSI_2019100103 crossref_primary_10_1109_TITS_2020_3034239 crossref_primary_10_1177_1475921720921772 crossref_primary_10_1007_s42417_023_01001_8 crossref_primary_10_1109_JIOT_2021_3126875 crossref_primary_10_1109_TIE_2020_3003622 crossref_primary_10_17671_gazibtd_762853 crossref_primary_10_1109_ACCESS_2023_3309814 crossref_primary_10_3390_app11125320 crossref_primary_10_1111_mice_12351 crossref_primary_10_3390_s21124065 crossref_primary_10_1177_13694332231175395 crossref_primary_10_1016_j_ndteint_2018_04_007 crossref_primary_10_1109_ACCESS_2020_3009654 crossref_primary_10_1109_TIM_2019_2909940 crossref_primary_10_35234_fumbd_894102 crossref_primary_10_1088_1742_6596_1834_1_012004 crossref_primary_10_1007_s11042_024_19565_2 crossref_primary_10_1109_TSMC_2018_2827937 crossref_primary_10_1109_TITS_2021_3058635 crossref_primary_10_1016_j_engstruct_2023_117085 crossref_primary_10_1109_TITS_2023_3305300 crossref_primary_10_1061_AJRUA6_0001024 crossref_primary_10_1111_mice_12362 crossref_primary_10_1109_JSEN_2019_2955378 crossref_primary_10_1109_TIM_2022_3144224 crossref_primary_10_1109_TITS_2020_3024216 crossref_primary_10_1109_JSEN_2023_3297205 crossref_primary_10_3390_ai5020024 crossref_primary_10_3934_ElectrEng_2020_1_19 crossref_primary_10_1016_j_engappai_2019_01_008 crossref_primary_10_1109_TITS_2022_3207490 crossref_primary_10_3390_s22249970 crossref_primary_10_48175_IJARSCT_18238 crossref_primary_10_1016_j_compbiomed_2023_107450 crossref_primary_10_1109_TIM_2018_2803830 crossref_primary_10_1088_1757_899X_397_1_012119 crossref_primary_10_1111_mice_12451 crossref_primary_10_3390_metrology4020016 crossref_primary_10_1109_TIM_2021_3117357 crossref_primary_10_1016_j_trc_2022_103679 crossref_primary_10_3390_electronics10111271 crossref_primary_10_1155_2021_6687146 crossref_primary_10_1016_j_engappai_2020_103678 crossref_primary_10_1109_ACCESS_2020_3022405 crossref_primary_10_1109_TIM_2019_2920721 crossref_primary_10_1016_j_measurement_2021_109454 crossref_primary_10_3390_buildings13081976 crossref_primary_10_3390_s21248335 crossref_primary_10_1109_TITS_2021_3095167 crossref_primary_10_1016_j_ssci_2022_105755 crossref_primary_10_1109_TIM_2019_2958580 crossref_primary_10_1109_TIM_2020_2973843 crossref_primary_10_48175_IJARSCT_8664 crossref_primary_10_1109_TIM_2021_3138498 crossref_primary_10_1051_e3sconf_202340902014 crossref_primary_10_3390_electronics13132564 crossref_primary_10_1002_stc_2504 crossref_primary_10_1007_s11801_022_2013_x crossref_primary_10_1016_j_chemosphere_2018_10_084 crossref_primary_10_1177_0954409718796385 crossref_primary_10_4236_ajor_2022_126015 crossref_primary_10_1587_transinf_2020EDP7097 crossref_primary_10_3390_constrmater4010005 crossref_primary_10_1007_s11042_020_09520_2 crossref_primary_10_14489_vkit_2022_02_pp_008_018 crossref_primary_10_1016_j_ijleo_2022_168607 crossref_primary_10_1016_j_eswa_2024_124105 crossref_primary_10_1109_TII_2018_2842821 crossref_primary_10_1016_j_trpro_2023_11_354 crossref_primary_10_1177_0959651820953679 crossref_primary_10_1016_j_conbuildmat_2023_131472 crossref_primary_10_3390_s20174945 crossref_primary_10_1016_j_jmsy_2022_06_011 crossref_primary_10_1080_19439962_2019_1611682 crossref_primary_10_3390_ma13214837 crossref_primary_10_1038_s41529_018_0058_x crossref_primary_10_1109_TIM_2020_3039301 crossref_primary_10_3390_app8091575 crossref_primary_10_3390_coatings13071140 crossref_primary_10_1109_ACCESS_2019_2921912 crossref_primary_10_1109_TITS_2019_2934346 crossref_primary_10_1002_adts_202200027 crossref_primary_10_1016_j_autcon_2022_104425 crossref_primary_10_1109_ACCESS_2020_3000068 crossref_primary_10_1007_s11042_023_15753_8 crossref_primary_10_1109_JSEN_2021_3132460 crossref_primary_10_1016_j_hspr_2022_11_001 crossref_primary_10_1088_1748_9326_ab66cb crossref_primary_10_1080_23248378_2020_1825128 crossref_primary_10_1109_ACCESS_2020_3017125 crossref_primary_10_1109_ACCESS_2022_3183102 crossref_primary_10_1080_15732479_2021_1994617 crossref_primary_10_1038_s41578_020_00255_y crossref_primary_10_3390_app12136296 crossref_primary_10_1080_00423114_2022_2103436 crossref_primary_10_1016_j_engappai_2023_106825 crossref_primary_10_1109_ACCESS_2021_3130261 crossref_primary_10_1016_j_eswa_2017_05_039 crossref_primary_10_3390_s20061562 crossref_primary_10_3390_s22051983 crossref_primary_10_1016_j_scs_2021_102898 crossref_primary_10_1155_2021_2565500 crossref_primary_10_1109_ACCESS_2020_3019318 crossref_primary_10_3390_app13116681 crossref_primary_10_1061__ASCE_IS_1943_555X_0000477 crossref_primary_10_1109_TITS_2021_3102613 crossref_primary_10_3390_app122010572 crossref_primary_10_3390_electronics13091781 crossref_primary_10_1109_ACCESS_2023_3273485 crossref_primary_10_3390_s23010544 crossref_primary_10_1093_iti_liad016 crossref_primary_10_1109_TIM_2018_2816800 crossref_primary_10_1088_1361_6501_abfdde crossref_primary_10_1109_ACCESS_2019_2957427 crossref_primary_10_3390_app13127330 crossref_primary_10_1177_09544097211030373 crossref_primary_10_3390_app112110373 crossref_primary_10_1016_j_ijmecsci_2023_108244 |
| Cites_doi | 10.3390/s110807364 10.1109/TITS.2014.2311123 10.1109/ICOIP.2010.303 10.1109/WACV.2012.6163021 10.1109/TSMCC.2007.893278 10.1109/CVPR.2014.81 10.1016/j.cemconres.2004.01.027 10.1109/34.531803 10.1109/WACV.2015.98 10.1109/ICIP.2009.5414438 10.1109/TPAMI.2002.1017623 10.1023/A:1007379606734 10.1109/ITSC.2014.6957919 10.1007/BF00344251 10.1016/S0008-8846(00)00283-0 10.1109/CVPR.2005.177 10.1109/TPAMI.2006.79 10.1145/2647868.2654889 10.1115/JRC/ICE2007-40080 10.1109/CVPR.2001.990517 10.1109/ITSC.2002.1041195 10.1145/3065386 10.1007/978-3-642-02230-2_7 10.1162/neco.1989.1.4.541 10.1109/TITS.2013.2287155 10.1364/AO.33.003751 10.1007/978-3-642-40246-3_55 10.1109/TIM.2013.2283741 10.1109/5.726791 10.1109/TITS.2012.2236555 10.1109/CVPR.2015.7298965 10.1109/CIHSPS.2006.313313 10.1186/1687-6180-2014-64 10.1109/CVPR.2015.7298594 |
| ContentType | Journal Article |
| Copyright | Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2017 |
| Copyright_xml | – notice: Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2017 |
| DBID | 97E RIA RIE AAYXX CITATION 7SC 7SP 8FD FR3 JQ2 KR7 L7M L~C L~D |
| DOI | 10.1109/TITS.2016.2568758 |
| DatabaseName | IEEE All-Society Periodicals Package (ASPP) 2005-present IEEE All-Society Periodicals Package (ASPP) 1998-Present IEEE Electronic Library (IEL) CrossRef Computer and Information Systems Abstracts Electronics & Communications Abstracts Technology Research Database Engineering Research Database ProQuest Computer Science Collection Civil Engineering Abstracts Advanced Technologies Database with Aerospace Computer and Information Systems Abstracts Academic Computer and Information Systems Abstracts Professional |
| DatabaseTitle | CrossRef Civil Engineering Abstracts Technology Research Database Computer and Information Systems Abstracts – Academic Electronics & Communications Abstracts ProQuest Computer Science Collection Computer and Information Systems Abstracts Engineering Research Database Advanced Technologies Database with Aerospace Computer and Information Systems Abstracts Professional |
| DatabaseTitleList | Civil Engineering Abstracts |
| Database_xml | – sequence: 1 dbid: RIE name: IEEE Electronic Library (IEL) url: https://proxy.k.utb.cz/login?url=https://ieeexplore.ieee.org/ sourceTypes: Publisher |
| DeliveryMethod | fulltext_linktorsrc |
| Discipline | Engineering |
| EISSN | 1558-0016 |
| EndPage | 164 |
| ExternalDocumentID | 10_1109_TITS_2016_2568758 7506117 |
| Genre | orig-research |
| GrantInformation_xml | – fundername: Federal Railroad Administration funderid: 10.13039/100006283 |
| GroupedDBID | -~X 0R~ 29I 4.4 5GY 5VS 6IK 97E AAJGR AASAJ ABQJQ ABTAH ABVLG ACGFO ACGFS ACIWK ACNCT AENEX AETIX AIBXA AKJIK ALMA_UNASSIGNED_HOLDINGS ATWAV BEFXN BFFAM BGNUA BKEBE BPEOZ CS3 DU5 EBS EJD HZ~ H~9 IFIPE IPLJI JAVBF LAI M43 O9- OCL P2P PQQKQ RIA RIE RIG RNS XFK ZY4 AAYXX CITATION 7SC 7SP 8FD FR3 JQ2 KR7 L7M L~C L~D |
| ID | FETCH-LOGICAL-c341t-c908220ffda8032931e7fd7dc63d9c2de50e233f67603b3ed0782f7392bca5973 |
| IEDL.DBID | RIE |
| ISSN | 1524-9050 |
| IngestDate | Thu Oct 10 19:23:15 EDT 2024 Fri Aug 23 03:47:01 EDT 2024 Wed Jun 26 19:18:18 EDT 2024 |
| IsPeerReviewed | true |
| IsScholarly | true |
| Issue | 1 |
| Language | English |
| LinkModel | DirectLink |
| MergedId | FETCHMERGED-LOGICAL-c341t-c908220ffda8032931e7fd7dc63d9c2de50e233f67603b3ed0782f7392bca5973 |
| PQID | 1855620753 |
| PQPubID | 75735 |
| PageCount | 12 |
| ParticipantIDs | crossref_primary_10_1109_TITS_2016_2568758 proquest_journals_1855620753 ieee_primary_7506117 |
| PublicationCentury | 2000 |
| PublicationDate | 2017-Jan. 2017-1-00 20170101 |
| PublicationDateYYYYMMDD | 2017-01-01 |
| PublicationDate_xml | – month: 01 year: 2017 text: 2017-Jan. |
| PublicationDecade | 2010 |
| PublicationPlace | New York |
| PublicationPlace_xml | – name: New York |
| PublicationTitle | IEEE transactions on intelligent transportation systems |
| PublicationTitleAbbrev | TITS |
| PublicationYear | 2017 |
| 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 ref34 muja (ref44) 2009 ref37 ref15 ref36 ref14 ref31 ref30 smak (ref1) 0 haghighat (ref46) 2013 ref32 gibert (ref5) 0 ref2 ref17 ref16 ref19 ref18 platt (ref48) 2000 pratt (ref38) 0 cunningham (ref6) 2000 berry (ref11) 0 gibert (ref10) 0 ref24 lv (ref35) 2015; 16 ref45 ref23 ref26 ref47 hinton (ref41) 0 ref25 hinton (ref39) 0 ref20 ref42 ref22 ref21 ref28 ref27 ref29 ref8 ahonen (ref43) 2009 trosino (ref7) 2002 ref9 ref4 ref3 collobert (ref33) 0 ref40 babenko (ref12) 2009 |
| References_xml | – ident: ref22 doi: 10.3390/s110807364 – ident: ref36 doi: 10.1109/TITS.2014.2311123 – ident: ref21 doi: 10.1109/ICOIP.2010.303 – ident: ref16 doi: 10.1109/WACV.2012.6163021 – ident: ref8 doi: 10.1109/TSMCC.2007.893278 – ident: ref31 doi: 10.1109/CVPR.2014.81 – ident: ref3 doi: 10.1016/j.cemconres.2004.01.027 – ident: ref47 doi: 10.1109/34.531803 – year: 2000 ident: ref6 publication-title: Automated Track Inspection Vehicle and Method contributor: fullname: cunningham – ident: ref4 doi: 10.1109/WACV.2015.98 – year: 2002 ident: ref7 publication-title: Automated Track Inspection Vehicle and Method contributor: fullname: trosino – ident: ref9 doi: 10.1109/ICIP.2009.5414438 – year: 2009 ident: ref12 publication-title: Visual Inspection of Railroad Tracks contributor: fullname: babenko – ident: ref45 doi: 10.1109/TPAMI.2002.1017623 – ident: ref37 doi: 10.1023/A:1007379606734 – start-page: 547 year: 2000 ident: ref48 article-title: Large margin DAGs for multiclass classification publication-title: Advances in Neural Information Systems (NIPS) contributor: fullname: platt – ident: ref25 doi: 10.1109/ITSC.2014.6957919 – start-page: 1 year: 0 ident: ref39 article-title: Learning distributed representation of concepts publication-title: Proc 8th Conf Cognitive Sci Soc contributor: fullname: hinton – ident: ref26 doi: 10.1007/BF00344251 – start-page: 289 year: 0 ident: ref10 article-title: A machine vision system for automated joint bar inspection from a moving rail vehicle publication-title: Proc ASME/IEEE Joint Rail Conf Internal Combustion Engine Spring Tech Conf contributor: fullname: gibert – ident: ref2 doi: 10.1016/S0008-8846(00)00283-0 – ident: ref42 doi: 10.1109/CVPR.2005.177 – volume: 16 start-page: 865 year: 2015 ident: ref35 article-title: Traffic flow prediction with big data: A deep learning approach publication-title: IEEE Trans Intell Transp Syst contributor: fullname: lv – start-page: 331 year: 2009 ident: ref44 article-title: Fast approximate nearest neighbors with automatic algorithm configuration publication-title: International Conference on Computer Vision Theory and Application VISSAPP'09) contributor: fullname: muja – ident: ref40 doi: 10.1109/TPAMI.2006.79 – ident: ref32 doi: 10.1145/2647868.2654889 – ident: ref20 doi: 10.1115/JRC/ICE2007-40080 – ident: ref17 doi: 10.1109/CVPR.2001.990517 – start-page: 1 year: 0 ident: ref1 article-title: Evolution of Amtrak's concrete crosstie and fastening system program publication-title: Proc Int Concrete Crosstie Fastening Syst Symp contributor: fullname: smak – ident: ref18 doi: 10.1109/ITSC.2002.1041195 – ident: ref29 doi: 10.1145/3065386 – start-page: 61 year: 2009 ident: ref43 article-title: Rotation invariant image description with local binary pattern histogram Fourier features publication-title: Image Analysis doi: 10.1007/978-3-642-02230-2_7 contributor: fullname: ahonen – ident: ref27 doi: 10.1162/neco.1989.1.4.541 – start-page: 1 year: 0 ident: ref33 article-title: Torch7: A Matlab-like environment for machine learning publication-title: Proc Advances NIPS contributor: fullname: collobert – ident: ref15 doi: 10.1109/TITS.2013.2287155 – start-page: 1 year: 0 ident: ref11 article-title: High speed video inspection of joint bars using advanced image collection and processing techniques publication-title: Proc World Congr Railway Res contributor: fullname: berry – ident: ref13 doi: 10.1364/AO.33.003751 – start-page: 440 year: 2013 ident: ref46 article-title: Identification using encrypted biometrics publication-title: Computer Analysis of Images and Patterns doi: 10.1007/978-3-642-40246-3_55 contributor: fullname: haghighat – ident: ref23 doi: 10.1109/TIM.2013.2283741 – ident: ref28 doi: 10.1109/5.726791 – start-page: 584 year: 0 ident: ref38 article-title: Direct transfer of learned information among neural networks publication-title: Proc AAAI contributor: fullname: pratt – ident: ref14 doi: 10.1109/TITS.2012.2236555 – ident: ref34 doi: 10.1109/CVPR.2015.7298965 – year: 0 ident: ref41 publication-title: Improving Neural Networks by Preventing Co-adaptation of Feature Detectors contributor: fullname: hinton – ident: ref19 doi: 10.1109/CIHSPS.2006.313313 – start-page: 621 year: 0 ident: ref5 article-title: Material classification and semantic segmentation of railway track images with deep convolutional neural networks publication-title: Proc IEEE ICIP contributor: fullname: gibert – ident: ref24 doi: 10.1186/1687-6180-2014-64 – ident: ref30 doi: 10.1109/CVPR.2015.7298594 |
| SSID | ssj0014511 |
| Score | 2.6573381 |
| Snippet | Railroad tracks need to be periodically inspected and monitored to ensure safe transportation. Automated track inspection using computer vision and pattern... |
| SourceID | proquest crossref ieee |
| SourceType | Aggregation Database Publisher |
| StartPage | 153 |
| SubjectTerms | Computer vision Deep convolutional neural networks Detectors Fasteners Inspection Machine learning material identification multitask learning Multitasking Neural networks Rail transportation Rails railway track inspection Railway tracks Training |
| Title | Deep Multitask Learning for Railway Track Inspection |
| URI | https://ieeexplore.ieee.org/document/7506117 https://www.proquest.com/docview/1855620753 |
| Volume | 18 |
| hasFullText | 1 |
| inHoldings | 1 |
| isFullTextHit | |
| isPrint | |
| link | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwjV07T8MwED61nWDgVRCFgjIwIZI6sePEI-KhFqkM0ErdIr-CUFFa0VQIfj22k1a8BjYPsWTd2fk---6-AziTuTS4SbhvuHXqE6yELyTVfhJzlCqqQ-kKhYf3tD8md5N40oCLdS2M1toln-nADl0sX83k0j6V9Qy60TBMmtBMUVTVaq0jBlZny2mjRsRnKF5FMEPEeqPB6NEmcdHA4Lvh5-k3DHJNVX79iR283G7DcLWwKqtkGixLEciPH5qN_135DmzVPNO7rDbGLjR0sQebX9QH20CutZ57VQUuX0y9Wmr1yTM81nvgzy9v_N0zWCan3qCoKjJnxT6Mb29GV32_bqLgSwNQpS9tT_MI5bniKcIG3EOd5CpRkmLFZKR0jHSEcU4TirDAWlnOkCeGNgnJzW0DH0CrmBX6EDykbBfwNGEiJYQRxJlAeUKVNBwnpEx34Hxl1mxeaWVk7o6BWGZ9kFkfZLUPOtC2Zlp_WFuoA92VI7L6NC0ywykMTTPkBh_9PesYNiILt-5ppAut8nWpTwxZKMWp2yWfWd65tA |
| link.rule.ids | 315,786,790,802,27955,27956,55107 |
| linkProvider | IEEE |
| linkToHtml | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwjV09T8MwED2VMgADXwVRKJCBCZHixKmdjAioWmg7QCp1ixLbQagorWgrBL-es5NUfA1sHmLJurPznn137wDORCoQN73YRm7t2x6ViZ0IpmzeiokvmXKEKRTuD1hn6N2NWqMKXCxrYZRSJvlMNfXQxPLlRCz0U9klohtzHL4Cq4jzhOfVWsuYgVbaMuqormcHpFXGMB0SXIbd8FGncbEmIjwydP8bCpm2Kr_-xQZg2lvQL5eW55WMm4t50hQfP1Qb_7v2bdgsmKZ1lW-NHaiobBc2vugP1sC7UWpq5TW48WxsFWKrTxYyWeshfn55i98tRDMxtrpZXpM5yfZg2L4Nrzt20UbBFghRc1voruYuSVMZ-4QivDuKp5JLwagMhCtViyiX0pRxRmhCldSsIeVInBIR432D7kM1m2TqACwidR9wnweJ73mBR-IgISlnUiDLcVig6nBemjWa5moZkbllkCDSPoi0D6LCB3WoaTMtPywsVIdG6YioOE-zCFkFEjWkN_Tw71mnsNYJ-72o1x3cH8G6q8HXPJQ0oDp_XahjpA7z5MTsmE9Vib0I |
| 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=Deep+Multitask+Learning+for+Railway+Track+Inspection&rft.jtitle=IEEE+transactions+on+intelligent+transportation+systems&rft.au=Gibert%2C+Xavier&rft.au=Patel%2C+Vishal+M.&rft.au=Chellappa%2C+Rama&rft.date=2017-01-01&rft.issn=1524-9050&rft.eissn=1558-0016&rft.volume=18&rft.issue=1&rft.spage=153&rft.epage=164&rft_id=info:doi/10.1109%2FTITS.2016.2568758&rft.externalDBID=n%2Fa&rft.externalDocID=10_1109_TITS_2016_2568758 |
| thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1524-9050&client=summon |
| thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1524-9050&client=summon |
| thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1524-9050&client=summon |