Transfer Linear Subspace Learning for Cross-Corpus Speech Emotion Recognition
Speech emotion recognition has received an increasing interest in recent years, which is often conducted on the assumption that speech utterances in training and testing datasets are obtained under the same conditions. However, in reality, this assumption does not hold as the speech data are often c...
Saved in:
| Published in: | IEEE transactions on affective computing Vol. 10; no. 2; pp. 265 - 275 |
|---|---|
| Main Author: | |
| Format: | Journal Article |
| Language: | English |
| Published: |
Piscataway
IEEE
01-04-2019
The Institute of Electrical and Electronics Engineers, Inc. (IEEE) |
| Subjects: | |
| Online Access: | Get full text |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Abstract | Speech emotion recognition has received an increasing interest in recent years, which is often conducted on the assumption that speech utterances in training and testing datasets are obtained under the same conditions. However, in reality, this assumption does not hold as the speech data are often collected from different devices or environments. Hence, there exists discrepancy between the training and testing data, which will have an adverse effect on recognition performance. In this paper, we examine the problem of cross-corpus speech emotion recognition. To address it, we present a novel transfer linear subspace learning (TLSL) framework to learn a common feature subspace for source and target datasets. In TLSL, a nearest neighbor graph algorithm is used to measure the similarity between different corpora, and a feature grouping strategy is introduced to divide the emotional features into two categories, i.e., high transferable part (HTP) versus low transferable part (LTP). To explore the proposed TLSL with different scenarios, we propose two kinds of TLSL approaches, called transfer unsupervised linear subspace learning (TULSL) and transfer supervised linear subspace learning (TSLSL), and provide the corresponding solutions for the optimization problems. Extensive experiments on several benchmark datasets validate the effectiveness of TLSL for cross-corpus speech emotion recognition. |
|---|---|
| AbstractList | Speech emotion recognition has received an increasing interest in recent years, which is often conducted on the assumption that speech utterances in training and testing datasets are obtained under the same conditions. However, in reality, this assumption does not hold as the speech data are often collected from different devices or environments. Hence, there exists discrepancy between the training and testing data, which will have an adverse effect on recognition performance. In this paper, we examine the problem of cross-corpus speech emotion recognition. To address it, we present a novel transfer linear subspace learning (TLSL) framework to learn a common feature subspace for source and target datasets. In TLSL, a nearest neighbor graph algorithm is used to measure the similarity between different corpora, and a feature grouping strategy is introduced to divide the emotional features into two categories, i.e., high transferable part (HTP) versus low transferable part (LTP). To explore the proposed TLSL with different scenarios, we propose two kinds of TLSL approaches, called transfer unsupervised linear subspace learning (TULSL) and transfer supervised linear subspace learning (TSLSL), and provide the corresponding solutions for the optimization problems. Extensive experiments on several benchmark datasets validate the effectiveness of TLSL for cross-corpus speech emotion recognition. |
| Author | Song, Peng |
| Author_xml | – sequence: 1 givenname: Peng orcidid: 0000-0002-6567-663X surname: Song fullname: Song, Peng email: pengsongseu@gmail.com organization: Yantai University, Yantai Shi, Shandong Sheng, China |
| BookMark | eNpNUMtOwzAQtFCRKKU_ABdLnFP8iO3mWEUtIAUh0XK2nGRTUlE72MmBv8ehFWIvs7ua2R3NNZpYZwGhW0oWlJLsYbfabPIFI1QtmCJCZvICTWmWZgknqZj866_QPIQDicU5l0xN0cvOGxsa8LhoLRiPt0MZOlMBLuJkW7vHjfM49y6EJHe-GwLedgDVB14fXd86i9-gcnvbjv0NumzMZ4D5GWfofbPe5U9J8fr4nK-KpOKS9tEKoyVUhC1TY1ImaxVBNFLJuAUFdSVKIpiogRMQKWQNN8qQlC3L2qSU8Bm6P93tvPsaIPT64AZv40vNGGdKMiZYZLETqxrde2h059uj8d-aEj0mp3-T02Ny-pxcFN2dRC0A_AlUxqnIOP8B_-1rew |
| CODEN | ITACBQ |
| CitedBy_id | crossref_primary_10_1016_j_knosys_2022_108659 crossref_primary_10_1109_ACCESS_2021_3069818 crossref_primary_10_1109_TAFFC_2022_3146325 crossref_primary_10_1016_j_apacoust_2020_107519 crossref_primary_10_3390_electronics11172745 crossref_primary_10_1109_TII_2020_3023677 crossref_primary_10_3233_JIFS_231263 crossref_primary_10_1109_TASLP_2020_3006331 crossref_primary_10_3390_e25010124 crossref_primary_10_1587_transinf_2020EDL8074 crossref_primary_10_1016_j_dsp_2020_102906 crossref_primary_10_1109_TAFFC_2022_3192899 crossref_primary_10_1109_TAFFC_2021_3077489 crossref_primary_10_1016_j_dsp_2021_103121 crossref_primary_10_3233_JIFS_230766 crossref_primary_10_1016_j_neucom_2023_126866 crossref_primary_10_1109_ACCESS_2022_3163712 crossref_primary_10_1109_TAFFC_2018_2800046 crossref_primary_10_1109_TAFFC_2022_3168834 crossref_primary_10_1109_TAFFC_2019_2916092 crossref_primary_10_1007_s13369_023_07920_8 crossref_primary_10_1109_JPROC_2023_3309299 crossref_primary_10_1016_j_specom_2020_12_009 crossref_primary_10_1109_ACCESS_2020_3000066 crossref_primary_10_1109_ACCESS_2020_3047395 crossref_primary_10_3390_e24081046 crossref_primary_10_1016_j_knosys_2023_110851 crossref_primary_10_32604_csse_2022_021635 crossref_primary_10_3390_e24091250 crossref_primary_10_1109_TCDS_2021_3055524 crossref_primary_10_1109_ACCESS_2020_2969032 crossref_primary_10_1109_ACCESS_2020_2984508 crossref_primary_10_1186_s13636_022_00264_5 crossref_primary_10_3390_s21041249 crossref_primary_10_1109_TMM_2021_3087098 crossref_primary_10_1016_j_mlwa_2021_100132 crossref_primary_10_1109_TASLP_2019_2955252 crossref_primary_10_1016_j_apacoust_2024_110118 crossref_primary_10_1109_TAFFC_2021_3109485 crossref_primary_10_1007_s00138_018_0960_9 crossref_primary_10_1109_TAFFC_2023_3290795 crossref_primary_10_1109_TCSS_2021_3130401 crossref_primary_10_1016_j_neucom_2020_10_021 crossref_primary_10_1109_TAFFC_2019_2947464 crossref_primary_10_1016_j_cogsys_2020_10_001 crossref_primary_10_1016_j_apacoust_2022_108919 crossref_primary_10_1109_TAFFC_2022_3167013 crossref_primary_10_1109_TAFFC_2022_3183166 crossref_primary_10_1002_cpe_7319 crossref_primary_10_1016_j_advengsoft_2023_103412 |
| Cites_doi | 10.1109/TASL.2013.2255278 10.1109/ICASSP.1996.541103 10.1109/TPAMI.2007.250598 10.3115/1610075.1610094 10.1109/TKDE.2009.191 10.1007/s11263-014-0696-6 10.21437/Interspeech.2010-739 10.1109/TIP.2010.2090535 10.1109/ICCV.2013.398 10.1016/S0167-6393(03)00099-2 10.1109/TASLP.2014.2319157 10.1109/89.279278 10.1587/transinf.E96.D.2286 10.1109/TASL.2007.902859 10.1109/TKDE.2014.2373376 10.1109/T-AFFC.2011.31 10.1109/CVPR.2014.183 10.1109/ICASSP.2016.7472665 10.1109/ICASSP.2007.366937 10.1049/el.2014.3339 10.1126/science.290.5500.2319 10.1109/ICASSP.2016.7472149 10.1145/1273496.1273521 10.1109/TNN.2010.2091281 10.1109/TNN.2005.860852 10.1145/1273496.1273592 10.1613/jair.1872 10.1109/TKDE.2013.111 10.1109/TNN.2009.2015760 10.1109/LSP.2014.2308954 10.1109/TKDE.2009.126 10.1109/T-AFFC.2010.8 10.1109/ICASSP.2006.1659966 10.1006/csla.1998.0043 10.1126/science.290.5500.2323 10.1109/89.876308 10.1007/978-3-540-45167-9_41 10.1109/LSP.2014.2324759 10.1109/TPAMI.2010.231 10.3115/1075671.1075688 10.1109/LSP.2016.2537926 10.1016/j.dsp.2012.05.007 10.1109/MLSP.2016.7738905 10.1109/APSIPA.2015.7415532 10.1109/ICASSP.2011.5947651 10.1145/1390156.1390182 10.1109/TNNLS.2014.2330900 10.1109/TKDE.2014.2382599 10.2197/ipsjjip.17.138 10.1109/ICASSP.2015.7178934 |
| ContentType | Journal Article |
| Copyright | Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2019 |
| Copyright_xml | – notice: Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2019 |
| DBID | 97E RIA RIE AAYXX CITATION 7SC 8FD JQ2 L7M L~C L~D |
| DOI | 10.1109/TAFFC.2017.2705696 |
| DatabaseName | IEEE All-Society Periodicals Package (ASPP) 2005-present IEEE All-Society Periodicals Package (ASPP) Online IEEE Electronic Library Online CrossRef Computer and Information Systems Abstracts Technology Research Database ProQuest Computer Science Collection Advanced Technologies Database with Aerospace Computer and Information Systems Abstracts Academic Computer and Information Systems Abstracts Professional |
| DatabaseTitle | CrossRef Computer and Information Systems Abstracts Technology Research Database Computer and Information Systems Abstracts – Academic Advanced Technologies Database with Aerospace ProQuest Computer Science Collection Computer and Information Systems Abstracts Professional |
| DatabaseTitleList | Computer and Information Systems Abstracts |
| Database_xml | – sequence: 1 dbid: RIE name: IEEE Electronic Library Online url: http://ieeexplore.ieee.org/Xplore/DynWel.jsp sourceTypes: Publisher |
| DeliveryMethod | fulltext_linktorsrc |
| Discipline | Computer Science |
| EISSN | 1949-3045 |
| EndPage | 275 |
| ExternalDocumentID | 10_1109_TAFFC_2017_2705696 7931593 |
| Genre | orig-research |
| GrantInformation_xml | – fundername: National Natural Science Foundation of China grantid: 61572419; 61602399 funderid: 10.13039/501100001809 – fundername: Fundamental Research Funds for the Southeast University grantid: CDLS-2017-02 – fundername: Natural Science Foundation of Shandong Province grantid: ZR2014FQ016 funderid: 10.13039/501100007129 |
| GroupedDBID | 0R~ 4.4 5VS 6IK 97E AAJGR AASAJ ABQJQ AENEX AKJIK ALMA_UNASSIGNED_HOLDINGS ATWAV BEFXN BFFAM BGNUA BKEBE BPEOZ EBS EJD HZ~ IEDLZ IFIPE IPLJI JAVBF M43 O9- OCL PQQKQ RIA RIC RIE RIG RNI RZB AAYXX CITATION 7SC 8FD JQ2 L7M L~C L~D |
| ID | FETCH-LOGICAL-c361t-3021bec0284aa426d7aa45f676bece7edc5b0525de30e54e9f3a7a0428bda4103 |
| IEDL.DBID | RIE |
| ISSN | 1949-3045 |
| IngestDate | Thu Oct 10 17:45:40 EDT 2024 Fri Aug 23 01:40:18 EDT 2024 Mon Nov 04 12:07:00 EST 2024 |
| IsPeerReviewed | true |
| IsScholarly | true |
| Issue | 2 |
| Language | English |
| LinkModel | DirectLink |
| MergedId | FETCHMERGED-LOGICAL-c361t-3021bec0284aa426d7aa45f676bece7edc5b0525de30e54e9f3a7a0428bda4103 |
| ORCID | 0000-0002-6567-663X |
| PQID | 2232762252 |
| PQPubID | 2040414 |
| PageCount | 11 |
| ParticipantIDs | proquest_journals_2232762252 ieee_primary_7931593 crossref_primary_10_1109_TAFFC_2017_2705696 |
| PublicationCentury | 2000 |
| PublicationDate | 2019-04-01 |
| PublicationDateYYYYMMDD | 2019-04-01 |
| PublicationDate_xml | – month: 04 year: 2019 text: 2019-04-01 day: 01 |
| PublicationDecade | 2010 |
| PublicationPlace | Piscataway |
| PublicationPlace_xml | – name: Piscataway |
| PublicationTitle | IEEE transactions on affective computing |
| PublicationTitleAbbrev | T-AFFC |
| PublicationYear | 2019 |
| 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 | ref57 ref13 belkin (ref53) 2001 ref12 fukunaga (ref28) 2013 ref59 ref15 niyogi (ref29) 2004 ref58 ref52 ref55 ref11 ref54 ref10 bishop (ref27) 2006 cai (ref56) 2011; 33 ref17 ref19 ref18 ben-david and (ref46) 2003 ref50 rosenstein (ref45) 2005 bakker (ref47) 2003; 4 ref48 kanamori (ref21) 2009; 10 ref44 kenny (ref14) 2005 ref43 jia (ref60) 2009; 20 sanchez (ref16) 2010 ref8 ref7 ref9 ref4 ref3 ref6 ref40 huang (ref42) 2006 ref35 ref34 ref37 ref36 ref31 ref30 ref33 han (ref5) 2014 ref32 ref2 ref1 wang (ref39) 2008 ref38 gretton (ref20) 2009; 3 ref24 ref23 ref26 jiang (ref41) 2007; 7 ref25 ref64 ref63 ref22 ref65 roweis (ref51) 2000; 290 ref62 pan (ref49) 2008 ref61 |
| References_xml | – year: 2013 ident: ref28 publication-title: Introduction to statistical pattern recognition contributor: fullname: fukunaga – ident: ref17 doi: 10.1109/TASL.2013.2255278 – year: 2006 ident: ref27 publication-title: Pattern Recognition and Machine Learning contributor: fullname: bishop – ident: ref8 doi: 10.1109/ICASSP.1996.541103 – ident: ref31 doi: 10.1109/TPAMI.2007.250598 – ident: ref43 doi: 10.3115/1610075.1610094 – ident: ref32 doi: 10.1109/TKDE.2009.191 – ident: ref54 doi: 10.1007/s11263-014-0696-6 – ident: ref62 doi: 10.21437/Interspeech.2010-739 – ident: ref55 doi: 10.1109/TIP.2010.2090535 – ident: ref48 doi: 10.1109/ICCV.2013.398 – ident: ref1 doi: 10.1016/S0167-6393(03)00099-2 – ident: ref63 doi: 10.1109/TASLP.2014.2319157 – ident: ref10 doi: 10.1109/89.279278 – ident: ref65 doi: 10.1587/transinf.E96.D.2286 – year: 2004 ident: ref29 article-title: Locality preserving projections publication-title: Proc Neural Inf Process Syst contributor: fullname: niyogi – ident: ref15 doi: 10.1109/TASL.2007.902859 – ident: ref50 doi: 10.1109/TKDE.2014.2373376 – start-page: 601 year: 2006 ident: ref42 article-title: Correcting sample selection bias by unlabeled data publication-title: Proc Adv Neural Inf Process Syst contributor: fullname: huang – ident: ref64 doi: 10.1109/T-AFFC.2011.31 – ident: ref58 doi: 10.1109/CVPR.2014.183 – start-page: 585 year: 2001 ident: ref53 article-title: Laplacian eigenmaps and spectral techniques for embedding and clustering publication-title: Proc 14th Int Conf Neural Inf Process Syst Natural Synthetic contributor: fullname: belkin – ident: ref61 doi: 10.1109/ICASSP.2016.7472665 – start-page: 550 year: 2008 ident: ref39 article-title: Transferred dimensionality reduction publication-title: Proc Knowl Discovery Data Mining contributor: fullname: wang – ident: ref2 doi: 10.1109/ICASSP.2007.366937 – ident: ref18 doi: 10.1049/el.2014.3339 – ident: ref52 doi: 10.1126/science.290.5500.2319 – ident: ref25 doi: 10.1109/ICASSP.2016.7472149 – volume: 10 start-page: 1391 year: 2009 ident: ref21 article-title: A least-squares approach to direct importance estimation publication-title: J Mach Learn Res contributor: fullname: kanamori – ident: ref35 doi: 10.1145/1273496.1273521 – ident: ref44 doi: 10.1109/TNN.2010.2091281 – ident: ref59 doi: 10.1109/TNN.2005.860852 – start-page: 223 year: 2014 ident: ref5 article-title: Speech emotion recognition using deep neural network and extreme learning machine publication-title: Proc INTERSPEECH contributor: fullname: han – ident: ref36 doi: 10.1145/1273496.1273592 – ident: ref37 doi: 10.1613/jair.1872 – ident: ref38 doi: 10.1109/TKDE.2013.111 – volume: 20 start-page: 729 year: 2009 ident: ref60 article-title: Trace ratio problem revisited publication-title: IEEE Trans Neural Netw doi: 10.1109/TNN.2009.2015760 contributor: fullname: jia – ident: ref6 doi: 10.1109/LSP.2014.2308954 – year: 2005 ident: ref45 article-title: To transfer or not to transfer publication-title: Proc NIPS Workshop Transfer Learning contributor: fullname: rosenstein – ident: ref30 doi: 10.1109/TKDE.2009.126 – ident: ref19 doi: 10.1109/T-AFFC.2010.8 – ident: ref13 doi: 10.1109/ICASSP.2006.1659966 – start-page: 677 year: 2008 ident: ref49 article-title: Transfer learning via dimensionality reduction publication-title: Proc 23rd Nat Conf Artif Intell contributor: fullname: pan – ident: ref9 doi: 10.1006/csla.1998.0043 – volume: 290 start-page: 2323 year: 2000 ident: ref51 article-title: Nonlinear dimensionality reduction by locally linear embedding publication-title: Science doi: 10.1126/science.290.5500.2323 contributor: fullname: roweis – ident: ref11 doi: 10.1109/89.876308 – start-page: 567 year: 2003 ident: ref46 article-title: Exploiting task relatedness for multiple task learning publication-title: Proc Conf Learning Theory and Kernel Machines doi: 10.1007/978-3-540-45167-9_41 contributor: fullname: ben-david and – volume: 7 start-page: 264 year: 2007 ident: ref41 article-title: Instance weighting for domain adaptation in NLP publication-title: ACL contributor: fullname: jiang – year: 2005 ident: ref14 article-title: Joint factor analysis of speaker and session variability: Theory and algorithms contributor: fullname: kenny – start-page: 2874 year: 2010 ident: ref16 article-title: Domain adaptation and compensation for emotion detection publication-title: Proc Interspeech Int Conf Spoken Language Process contributor: fullname: sanchez – ident: ref23 doi: 10.1109/LSP.2014.2324759 – volume: 33 start-page: 1548 year: 2011 ident: ref56 article-title: Graph regularized nonnegative matrix factorization for data representation publication-title: IEEE Trans Pattern Anal Mach Intell doi: 10.1109/TPAMI.2010.231 contributor: fullname: cai – ident: ref7 doi: 10.3115/1075671.1075688 – ident: ref26 doi: 10.1109/LSP.2016.2537926 – ident: ref4 doi: 10.1016/j.dsp.2012.05.007 – ident: ref12 doi: 10.1109/MLSP.2016.7738905 – ident: ref33 doi: 10.1109/APSIPA.2015.7415532 – ident: ref3 doi: 10.1109/ICASSP.2011.5947651 – ident: ref40 doi: 10.1145/1390156.1390182 – ident: ref34 doi: 10.1109/TNNLS.2014.2330900 – ident: ref57 doi: 10.1109/TKDE.2014.2382599 – ident: ref22 doi: 10.2197/ipsjjip.17.138 – volume: 3 year: 2009 ident: ref20 article-title: Covariate shift by kernel mean matching publication-title: Dataset Shift Mach Learn contributor: fullname: gretton – volume: 4 start-page: 83 year: 2003 ident: ref47 article-title: Task clustering and gating for Bayesian multitask learning publication-title: J Mach Learn Res contributor: fullname: bakker – ident: ref24 doi: 10.1109/ICASSP.2015.7178934 |
| SSID | ssj0000333627 |
| Score | 2.4763532 |
| Snippet | Speech emotion recognition has received an increasing interest in recent years, which is often conducted on the assumption that speech utterances in training... |
| SourceID | proquest crossref ieee |
| SourceType | Aggregation Database Publisher |
| StartPage | 265 |
| SubjectTerms | Algorithm design and analysis Algorithms Datasets dimensionality reduction Emotion recognition Learning Optimization Principal component analysis Speech Speech recognition subspace learning Subspaces Testing Training transfer learning |
| Title | Transfer Linear Subspace Learning for Cross-Corpus Speech Emotion Recognition |
| URI | https://ieeexplore.ieee.org/document/7931593 https://www.proquest.com/docview/2232762252 |
| Volume | 10 |
| hasFullText | 1 |
| inHoldings | 1 |
| isFullTextHit | |
| isPrint | |
| link | http://sdu.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV2xTsMwED3RTiwUKIhCQR7YwG0Sx3YyVqURCwy0SGxREl9gSiva_H_PTlIhwcKUyIoj686-d2ef7wHcF7JQnhHIY5krTh5xxvMyjHge-1Ekc2H8zJHYLvXrR_S0sGVyHg93YRDRJZ_hxL66s3yzLmq7VTaluUToK3rQ03HU3NU67Kd4QpAt1t29GC-ermZJMrfJW3oSaMJ5W5f_B_Y4MpVfFtjBSjL434BO4aR1H9ms0fcZHGF1DoOOmoG1K3UILw6DSmqiYJMmM7MGgsJjZG1B1U9G3iqb23FyW8243rLlBrH4YouG2Ie9dalF6-oC3pPFav7MW-YEXgjl77gg5CblkO8QZhlhsNH0kKXSilpRoylkbgnsDAoPZYhxKTKd2fApN1noe-IS-tW6witgCqkvueGWyDMUJYUj0lb583Xk5UKFagQPnUzTTVMgI3WBhRenTgOp1UDaamAEQyvFw5etAEcw7tSQtmtom5LjEpCpDmRw_XevGzimf8dNHs0Y-rvvGm-htzX1nZsbe1rlt3k |
| link.rule.ids | 315,782,786,798,27935,27936,54770 |
| linkProvider | IEEE |
| linkToHtml | http://sdu.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV09T8MwED1BGWChQEEUCnhgg7ROHMfJWJVGRbQdaJHYonxcYEor2vx_zk5SIcHClMiKFevOvndnn-8B3Kcy9Xgm0Apk4lnkEcdWkru-lQS278tEZHZsSGwXav7uP411mZzH3V0YRDTJZ9jXr-YsP1ulpd4qG9BcIvQV-3AgXaV4dVtrt6PChSBrrJqbMTwYLIdhONLpW6rvKEJ6XZn_B_oYOpVfNtgAS9j-35BO4Lh2INmw0vgp7GFxBu2GnIHVa7UDM4NCOTVRuEnTmWkTQQEysrqk6gcjf5WN9DgtXc-43LDFGjH9ZOOK2oe9NslFq-Ic3sLxcjSxau4EKxWevbUEYTeph7wHN44JhTNFD5l7yqNWVJilMtEUdhkKjtLFIBexinUAlWSxa3NxAa1iVeAlMA-pLznimsrTFTkFJFLX-bOVzxPhuV4XHhqZRuuqREZkQgseREYDkdZAVGugCx0txd2XtQC70GvUENWraBOR6-KQsXakc_V3rzs4nCxn02j6PH-5hiP6T1Bl1fSgtf0q8Qb2N1l5a-bJNyNcusQ |
| 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=Transfer+Linear+Subspace+Learning+for+Cross-Corpus+Speech+Emotion+Recognition&rft.jtitle=IEEE+transactions+on+affective+computing&rft.au=Song%2C+Peng&rft.date=2019-04-01&rft.pub=IEEE&rft.eissn=1949-3045&rft.volume=10&rft.issue=2&rft.spage=265&rft.epage=275&rft_id=info:doi/10.1109%2FTAFFC.2017.2705696&rft.externalDocID=7931593 |
| thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1949-3045&client=summon |
| thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1949-3045&client=summon |
| thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1949-3045&client=summon |