Fusion of Deep Learning and Compressed Domain Features for Content-Based Image Retrieval
This paper presents an effective image retrieval method by combining high-level features from convolutional neural network (CNN) model and low-level features from dot-diffused block truncation coding (DDBTC). The low-level features, e.g., texture and color, are constructed by vector quantization -in...
Saved in:
| Published in | IEEE transactions on image processing Vol. 26; no. 12; pp. 5706 - 5717 |
|---|---|
| Main Authors | , , , |
| Format | Journal Article |
| Language | English |
| Published |
United States
IEEE
01.12.2017
|
| Subjects | |
| Online Access | Get full text |
Cover
Loading…
| Abstract | This paper presents an effective image retrieval method by combining high-level features from convolutional neural network (CNN) model and low-level features from dot-diffused block truncation coding (DDBTC). The low-level features, e.g., texture and color, are constructed by vector quantization -indexed histogram from DDBTC bitmap, maximum, and minimum quantizers. Conversely, high-level features from CNN can effectively capture human perception. With the fusion of the DDBTC and CNN features, the extended deep learning two-layer codebook features is generated using the proposed two-layer codebook, dimension reduction, and similarity reweighting to improve the overall retrieval rate. Two metrics, average precision rate and average recall rate (ARR), are employed to examine various data sets. As documented in the experimental results, the proposed schemes can achieve superior performance compared with the state-of-the-art methods with either low-or high-level features in terms of the retrieval rate. Thus, it can be a strong candidate for various image retrieval related applications. |
|---|---|
| AbstractList | This paper presents an effective image retrieval method by combining high-level features from convolutional neural network (CNN) model and low-level features from dot-diffused block truncation coding (DDBTC). The low-level features, e.g., texture and color, are constructed by vector quantization -indexed histogram from DDBTC bitmap, maximum, and minimum quantizers. Conversely, high-level features from CNN can effectively capture human perception. With the fusion of the DDBTC and CNN features, the extended deep learning two-layer codebook features is generated using the proposed two-layer codebook, dimension reduction, and similarity reweighting to improve the overall retrieval rate. Two metrics, average precision rate and average recall rate (ARR), are employed to examine various data sets. As documented in the experimental results, the proposed schemes can achieve superior performance compared with the state-of-the-art methods with either low-or high-level features in terms of the retrieval rate. Thus, it can be a strong candidate for various image retrieval related applications.This paper presents an effective image retrieval method by combining high-level features from convolutional neural network (CNN) model and low-level features from dot-diffused block truncation coding (DDBTC). The low-level features, e.g., texture and color, are constructed by vector quantization -indexed histogram from DDBTC bitmap, maximum, and minimum quantizers. Conversely, high-level features from CNN can effectively capture human perception. With the fusion of the DDBTC and CNN features, the extended deep learning two-layer codebook features is generated using the proposed two-layer codebook, dimension reduction, and similarity reweighting to improve the overall retrieval rate. Two metrics, average precision rate and average recall rate (ARR), are employed to examine various data sets. As documented in the experimental results, the proposed schemes can achieve superior performance compared with the state-of-the-art methods with either low-or high-level features in terms of the retrieval rate. Thus, it can be a strong candidate for various image retrieval related applications. This paper presents an effective image retrieval method by combining high-level features from convolutional neural network (CNN) model and low-level features from dot-diffused block truncation coding (DDBTC). The low-level features, e.g., texture and color, are constructed by vector quantization -indexed histogram from DDBTC bitmap, maximum, and minimum quantizers. Conversely, high-level features from CNN can effectively capture human perception. With the fusion of the DDBTC and CNN features, the extended deep learning two-layer codebook features is generated using the proposed two-layer codebook, dimension reduction, and similarity reweighting to improve the overall retrieval rate. Two metrics, average precision rate and average recall rate (ARR), are employed to examine various data sets. As documented in the experimental results, the proposed schemes can achieve superior performance compared with the state-of-the-art methods with either low-or high-level features in terms of the retrieval rate. Thus, it can be a strong candidate for various image retrieval related applications. |
| Author | Jing-Ming Guo Peizhong Liu Chi-Yi Wu Danlin Cai |
| Author_xml | – sequence: 1 givenname: Peizhong surname: Liu fullname: Liu, Peizhong – sequence: 2 givenname: Jing-Ming surname: Guo fullname: Guo, Jing-Ming – sequence: 3 givenname: Chi-Yi surname: Wu fullname: Wu, Chi-Yi – sequence: 4 givenname: Danlin orcidid: 0000-0003-3083-5503 surname: Cai fullname: Cai, Danlin |
| BackLink | https://www.ncbi.nlm.nih.gov/pubmed/28866491$$D View this record in MEDLINE/PubMed |
| BookMark | eNp9kE1r3DAQhkVIyFd7LxSCjr14q09bOrabbLKwkBBS6M3I9iio2NJGkgP999Gy2xxy6GnE6Hk1o-cCHfvgAaEvlCwoJfr70_phwQhtFqzhNRf8CJ1TLWhFiGDH5UxkUzVU6DN0kdIfQqiQtD5FZ0ypuhaanqPfqzm54HGw-Bpgizdgonf-GRs_4GWYthFSggFfh8k4j1dg8lxa2IZYrn0Gn6ufZkesJ_MM-BFydPBqxk_oxJoxwedDvUS_VjdPy7tqc3-7Xv7YVD1XMlfSdtBZqQdlhVAg7ACSW83KpraThlndWG6bQYIgVDeC8b7ugGtlazXowfJL9G3_7jaGlxlSbieXehhH4yHMqaWaS0F4zVhBrw7o3E0wtNvoJhP_tv9sFIDsgT6GlCLYd4SSdie8LcLbnfD2ILxE6g-R3mWTi9IcjRv_F_y6DzoAeJ-jyicV4_wNeSGMjg |
| CODEN | IIPRE4 |
| CitedBy_id | crossref_primary_10_32604_csse_2022_021459 crossref_primary_10_1007_s11063_021_10537_3 crossref_primary_10_1109_ACCESS_2020_3010882 crossref_primary_10_1109_ACCESS_2021_3076074 crossref_primary_10_1109_TITS_2019_2909915 crossref_primary_10_1177_0020720920936834 crossref_primary_10_1109_ACCESS_2019_2950968 crossref_primary_10_1109_ACCESS_2019_2938000 crossref_primary_10_1109_JIOT_2021_3105647 crossref_primary_10_1016_j_ijleo_2021_167754 crossref_primary_10_1109_TIP_2019_2901407 crossref_primary_10_1109_TCSVT_2019_2959875 crossref_primary_10_3390_app10196829 crossref_primary_10_1016_j_dsp_2025_105138 crossref_primary_10_1049_iet_cvi_2018_5206 crossref_primary_10_1371_journal_pone_0232776 crossref_primary_10_1007_s11042_022_13120_7 crossref_primary_10_1080_17538947_2022_2036833 crossref_primary_10_1007_s12652_018_0963_4 crossref_primary_10_1016_j_eij_2024_100499 crossref_primary_10_1007_s11042_021_11045_1 crossref_primary_10_1007_s00371_021_02170_x crossref_primary_10_1016_j_patcog_2023_109748 crossref_primary_10_1007_s13369_019_03880_0 crossref_primary_10_1155_2019_4625371 crossref_primary_10_1007_s12065_020_00563_w crossref_primary_10_1109_ACCESS_2019_2947006 crossref_primary_10_1089_big_2021_0049 crossref_primary_10_3389_fninf_2021_805669 crossref_primary_10_3390_a12070135 crossref_primary_10_3390_info14030184 crossref_primary_10_1088_1742_6596_2833_1_012013 crossref_primary_10_1109_ACCESS_2020_2981720 crossref_primary_10_1016_j_eswa_2020_114545 crossref_primary_10_1016_j_sigpro_2019_107329 crossref_primary_10_1109_TCYB_2019_2894498 crossref_primary_10_1016_j_irbm_2020_06_007 crossref_primary_10_1109_ACCESS_2019_2911630 crossref_primary_10_1007_s11042_024_18488_2 crossref_primary_10_1007_s11063_022_11079_y crossref_primary_10_1007_s11760_023_02631_x crossref_primary_10_1049_iet_ipr_2017_0917 crossref_primary_10_1007_s00371_020_01839_z crossref_primary_10_1016_j_dibe_2022_100085 crossref_primary_10_1016_j_neucom_2020_04_125 crossref_primary_10_1007_s11042_019_08113_y crossref_primary_10_1177_0040517519829003 crossref_primary_10_1142_S0219467822500474 crossref_primary_10_1016_j_heliyon_2023_e22316 crossref_primary_10_1109_ACCESS_2019_2948388 crossref_primary_10_1016_j_compeleceng_2023_108647 crossref_primary_10_1007_s11042_020_09983_3 crossref_primary_10_1016_j_inffus_2019_07_009 crossref_primary_10_1016_j_dsp_2020_102765 crossref_primary_10_1007_s11042_022_12348_7 crossref_primary_10_1007_s11042_020_10269_x crossref_primary_10_1016_j_bspc_2022_103710 crossref_primary_10_1016_j_zemedi_2018_11_002 crossref_primary_10_1049_iet_ipr_2019_1023 crossref_primary_10_1007_s12652_020_02139_z crossref_primary_10_1049_iet_ipr_2018_6619 crossref_primary_10_1177_1473871619891062 crossref_primary_10_1016_j_ipm_2022_103119 crossref_primary_10_1142_S0219467821500467 crossref_primary_10_1541_ieejeiss_138_853 crossref_primary_10_1016_j_compag_2020_105714 crossref_primary_10_1016_j_engappai_2021_104256 crossref_primary_10_1016_j_patcog_2019_04_003 crossref_primary_10_1007_s11042_018_5943_3 crossref_primary_10_1117_1_JEI_32_2_023003 crossref_primary_10_1007_s00521_019_04279_6 crossref_primary_10_3390_electronics10131574 crossref_primary_10_1109_ACCESS_2020_2968982 crossref_primary_10_1007_s10044_020_00879_4 crossref_primary_10_1109_ACCESS_2020_2986157 crossref_primary_10_1109_JSEN_2021_3084618 crossref_primary_10_1007_s11042_022_13119_0 crossref_primary_10_1016_j_compeleceng_2019_02_006 crossref_primary_10_3390_math8061019 crossref_primary_10_1007_s10115_024_02152_0 crossref_primary_10_1016_j_neucom_2020_11_057 |
| Cites_doi | 10.1109/76.927424 10.1049/el:20052176 10.1109/TIP.2014.2329182 10.1109/SITIS.2009.43 10.1016/j.mcm.2011.11.064 10.1109/CVPR.2015.7298594 10.1109/TMM.2015.2449234 10.1109/TSMCB.2006.880137 10.1016/j.patcog.2010.02.012 10.1109/TIP.2014.2305072 10.1109/TIP.2013.2257812 10.1016/j.compeleceng.2012.11.023 10.1007/978-1-4471-2099-5_24 10.1016/j.imavis.2004.03.026 10.1016/j.imavis.2008.07.004 10.1016/j.jvcir.2012.10.003 10.1049/iet-ipr.2013.0375 10.1016/j.knosys.2010.06.001 10.1109/TIP.2014.2372619 10.1016/j.patcog.2011.11.009 10.1016/j.eswa.2011.03.014 10.1109/CVPR.2010.5539957 10.1016/S0031-3203(03)00010-4 10.1016/S0031-3203(02)00083-3 10.1109/34.531803 10.1016/j.patrec.2008.10.005 10.1016/j.patcog.2011.02.003 10.1016/j.ipm.2006.07.014 10.1109/TCSVT.2014.2358011 10.1016/0031-3203(95)00067-4 10.1109/TIP.2002.807356 10.1109/TIP.2010.2042645 10.1109/CVPR.2016.227 10.1109/TSMCB.2005.850176 10.1109/TIP.2012.2188809 10.1007/978-3-642-01307-2_90 10.1016/j.sigpro.2011.12.005 10.1016/j.patcog.2009.08.017 10.1109/CVPR.2006.264 10.1109/ICCV.2003.1238663 10.1016/j.patrec.2009.02.006 10.1049/el.2010.3232 10.1109/CBMI.2010.5529838 10.1145/2647868.2654948 10.1016/j.eswa.2011.11.029 |
| ContentType | Journal Article |
| DBID | 97E RIA RIE AAYXX CITATION NPM 7X8 |
| DOI | 10.1109/TIP.2017.2736343 |
| DatabaseName | IEEE All-Society Periodicals Package (ASPP) 2005–Present IEEE All-Society Periodicals Package (ASPP) 1998–Present IEEE Electronic Library (IEL) CrossRef PubMed MEDLINE - Academic |
| DatabaseTitle | CrossRef PubMed MEDLINE - Academic |
| DatabaseTitleList | MEDLINE - Academic PubMed |
| Database_xml | – sequence: 1 dbid: NPM name: PubMed url: https://proxy.k.utb.cz/login?url=http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed sourceTypes: Index Database – sequence: 2 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 | Applied Sciences Engineering |
| EISSN | 1941-0042 |
| EndPage | 5717 |
| ExternalDocumentID | 28866491 10_1109_TIP_2017_2736343 8019823 |
| Genre | orig-research Journal Article |
| GrantInformation_xml | – fundername: Natural Science Foundation of Fujian Province, China grantid: 2016J01300 funderid: 10.13039/501100003392 – fundername: National Science Council of Taiwan grantid: 105-2628-E-011 -002 -MY3 – fundername: Fundamental Research Funds for the Central Universities grantid: JB-ZR1202 – fundername: Fujian Provincial Key Laboratory of Data-intensive Computing and Fujian University Laboratory of Intelligent Computing and Information Processing – fundername: National Natural Science Foundation of China grantid: 61605048 funderid: 10.13039/501100001809 |
| GroupedDBID | --- -~X .DC 0R~ 29I 4.4 53G 5GY 5VS 6IK 97E AAJGR AARMG AASAJ AAWTH ABAZT ABFSI ABQJQ ABVLG ACGFO ACGFS ACIWK AENEX AETIX AGQYO AGSQL AHBIQ AI. AIBXA AKJIK AKQYR ALLEH ALMA_UNASSIGNED_HOLDINGS ASUFR ATWAV BEFXN BFFAM BGNUA BKEBE BPEOZ CS3 DU5 E.L EBS EJD F5P HZ~ H~9 ICLAB IFIPE IFJZH IPLJI JAVBF LAI M43 MS~ O9- OCL P2P RIA RIE RNS TAE TN5 VH1 AAYOK AAYXX CITATION RIG NPM 7X8 |
| ID | FETCH-LOGICAL-c385t-5fbebf59d8f448e4fde53f92145fb5a2f97f3f7d5e40197423c6be398f68d9df3 |
| IEDL.DBID | RIE |
| ISSN | 1057-7149 1941-0042 |
| IngestDate | Thu Jul 10 16:59:05 EDT 2025 Thu Apr 03 07:08:34 EDT 2025 Thu Apr 24 23:04:06 EDT 2025 Tue Jul 01 02:03:15 EDT 2025 Tue Aug 26 17:01:02 EDT 2025 |
| IsPeerReviewed | true |
| IsScholarly | true |
| Issue | 12 |
| Language | English |
| License | https://ieeexplore.ieee.org/Xplorehelp/downloads/license-information/IEEE.html |
| LinkModel | DirectLink |
| MergedId | FETCHMERGED-LOGICAL-c385t-5fbebf59d8f448e4fde53f92145fb5a2f97f3f7d5e40197423c6be398f68d9df3 |
| Notes | ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 23 |
| ORCID | 0000-0003-3083-5503 |
| PMID | 28866491 |
| PQID | 1935403622 |
| PQPubID | 23479 |
| PageCount | 12 |
| ParticipantIDs | ieee_primary_8019823 proquest_miscellaneous_1935403622 crossref_citationtrail_10_1109_TIP_2017_2736343 pubmed_primary_28866491 crossref_primary_10_1109_TIP_2017_2736343 |
| ProviderPackageCode | CITATION AAYXX |
| PublicationCentury | 2000 |
| PublicationDate | 2017-Dec. 2017-12-00 2017-Dec 20171201 |
| PublicationDateYYYYMMDD | 2017-12-01 |
| PublicationDate_xml | – month: 12 year: 2017 text: 2017-Dec. |
| PublicationDecade | 2010 |
| PublicationPlace | United States |
| PublicationPlace_xml | – name: United States |
| PublicationTitle | IEEE transactions on image processing |
| PublicationTitleAbbrev | TIP |
| PublicationTitleAlternate | IEEE Trans Image Process |
| PublicationYear | 2017 |
| Publisher | IEEE |
| Publisher_xml | – name: IEEE |
| References | ref57 ref59 ref15 ref58 ref53 saadatmand-tarzjan (ref34) 2006 ref52 ref55 ref54 (ref19) 2016 ref17 wang (ref51) 2011 silakari (ref24) 2009; 4 (ref13) 2016 ref50 ref46 ref45 ref48 simonyan (ref6) 2014 ref47 ref42 ref41 ref44 ref43 ref49 ref7 ref9 arandjelovic (ref11) 2013 ref5 ref40 gahroudi (ref22) 2007 ref35 guo (ref37) 2015; 25 krizhevsky (ref4) 2012 ref36 ref31 ref30 ref33 ref32 jia (ref8) 2013 ref2 ref1 ref39 ref38 (ref14) 2016 tan (ref56) 2010; 19 lin (ref10) 2013 (ref18) 2016 ref23 ref26 ref25 ref20 chiang (ref29) 2006; 1 ref21 ref28 ref27 zeiler (ref3) 2014 ref60 ref62 jegou (ref16) 2013 ref61 (ref12) 2016 |
| References_xml | – start-page: 1 year: 2007 ident: ref22 article-title: Image retrieval based on texture and color method in BTC-VQ compressed domain publication-title: Proc 9th Int Symp Signal Process Appl – ident: ref41 doi: 10.1109/76.927424 – ident: ref30 doi: 10.1049/el:20052176 – ident: ref50 doi: 10.1109/TIP.2014.2329182 – ident: ref47 doi: 10.1109/SITIS.2009.43 – ident: ref31 doi: 10.1016/j.mcm.2011.11.064 – ident: ref5 doi: 10.1109/CVPR.2015.7298594 – ident: ref1 doi: 10.1109/TMM.2015.2449234 – ident: ref33 doi: 10.1109/TSMCB.2006.880137 – year: 2016 ident: ref13 publication-title: SIPI-USC Brodatz Texture Image Database – ident: ref42 doi: 10.1016/j.patcog.2010.02.012 – ident: ref52 doi: 10.1109/TIP.2014.2305072 – year: 2016 ident: ref18 publication-title: Outex Texture Image Database – ident: ref2 doi: 10.1109/TIP.2013.2257812 – ident: ref60 doi: 10.1016/j.compeleceng.2012.11.023 – ident: ref49 doi: 10.1007/978-1-4471-2099-5_24 – ident: ref26 doi: 10.1016/j.imavis.2004.03.026 – ident: ref25 doi: 10.1016/j.imavis.2008.07.004 – volume: 4 start-page: 31 year: 2009 ident: ref24 article-title: Color image clustering using block truncation algorithm publication-title: Int J Comput Sci Issues – year: 2013 ident: ref10 publication-title: Network in Network – ident: ref43 doi: 10.1016/j.jvcir.2012.10.003 – volume: 1 start-page: 205 year: 2006 ident: ref29 article-title: Content-based image retrieval via the multiresolution wavelet features of interest publication-title: J Inf Technol Appl – ident: ref48 doi: 10.1049/iet-ipr.2013.0375 – ident: ref32 doi: 10.1016/j.knosys.2010.06.001 – ident: ref36 doi: 10.1109/TIP.2014.2372619 – start-page: 304 year: 2013 ident: ref16 article-title: Hamming embedding and weak geometric consistency for large scale image search publication-title: Proc Eur Conf Comput Vis (ECCV) – ident: ref17 doi: 10.1016/j.patcog.2011.11.009 – ident: ref39 doi: 10.1016/j.eswa.2011.03.014 – start-page: 209 year: 2011 ident: ref51 article-title: Contextual weighting for vocabulary tree based image retrieval publication-title: Proc IEEE Int Conf Comput Vis (ICCV) – ident: ref7 doi: 10.1109/CVPR.2010.5539957 – ident: ref38 doi: 10.1016/S0031-3203(03)00010-4 – ident: ref27 doi: 10.1016/S0031-3203(02)00083-3 – ident: ref53 doi: 10.1109/34.531803 – ident: ref15 doi: 10.1016/j.patrec.2008.10.005 – start-page: 925 year: 2006 ident: ref34 article-title: Gabor wavelet correlogram algorithm for image indexing and retrieval publication-title: Proc 18th Int Conf Pattern Recognit – ident: ref40 doi: 10.1016/j.patcog.2011.02.003 – ident: ref28 doi: 10.1016/j.ipm.2006.07.014 – volume: 25 start-page: 466 year: 2015 ident: ref37 article-title: Content-based image retrieval using error diffusion block truncation coding features publication-title: IEEE Trans Circuits Syst Video Technol doi: 10.1109/TCSVT.2014.2358011 – ident: ref55 doi: 10.1016/0031-3203(95)00067-4 – ident: ref21 doi: 10.1109/TIP.2002.807356 – volume: 19 start-page: 1635 year: 2010 ident: ref56 article-title: Enhanced local texture feature sets for face recognition under difficult lighting conditions publication-title: IEEE Trans Image Process doi: 10.1109/TIP.2010.2042645 – year: 2016 ident: ref12 publication-title: Corel Photo Collection Color Image Database – ident: ref62 doi: 10.1109/CVPR.2016.227 – ident: ref54 doi: 10.1109/TSMCB.2005.850176 – ident: ref59 doi: 10.1109/TIP.2012.2188809 – ident: ref46 doi: 10.1007/978-3-642-01307-2_90 – ident: ref58 doi: 10.1016/j.sigpro.2011.12.005 – start-page: 818 year: 2014 ident: ref3 article-title: Visualizing and understanding convolutional networks publication-title: Vision Computer – ident: ref57 doi: 10.1016/j.patcog.2009.08.017 – year: 2016 ident: ref19 publication-title: KTH-TIPS Texture Image Database – ident: ref20 doi: 10.1109/CVPR.2006.264 – year: 2014 ident: ref6 publication-title: Very Deep Convolutional Networks for Large-scale Image Recognition – year: 2013 ident: ref8 publication-title: Caffe An Open Source Convolutional Architecture for Fast Feature Embedding – ident: ref44 doi: 10.1109/ICCV.2003.1238663 – ident: ref61 doi: 10.1016/j.patrec.2009.02.006 – start-page: 1578 year: 2013 ident: ref11 article-title: All about VLAD publication-title: Proc IEEE Conf CVPR – ident: ref23 doi: 10.1049/el.2010.3232 – ident: ref45 doi: 10.1109/CBMI.2010.5529838 – start-page: 1097 year: 2012 ident: ref4 article-title: ImageNet classification with deep convolutional neural networks publication-title: Proc NIPS – ident: ref9 doi: 10.1145/2647868.2654948 – ident: ref35 doi: 10.1016/j.eswa.2011.11.029 – year: 2016 ident: ref14 publication-title: MIT-Vision Texture (VisTex) Image Database |
| SSID | ssj0014516 |
| Score | 2.54717 |
| Snippet | This paper presents an effective image retrieval method by combining high-level features from convolutional neural network (CNN) model and low-level features... |
| SourceID | proquest pubmed crossref ieee |
| SourceType | Aggregation Database Index Database Enrichment Source Publisher |
| StartPage | 5706 |
| SubjectTerms | block truncation coding Computational complexity Content-based image retrieval convolutional-neural network deep learning Feature extraction halftoning Histograms Image coding Image color analysis Image retrieval Machine learning |
| Title | Fusion of Deep Learning and Compressed Domain Features for Content-Based Image Retrieval |
| URI | https://ieeexplore.ieee.org/document/8019823 https://www.ncbi.nlm.nih.gov/pubmed/28866491 https://www.proquest.com/docview/1935403622 |
| Volume | 26 |
| hasFullText | 1 |
| inHoldings | 1 |
| isFullTextHit | |
| isPrint | |
| link | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwjV1Lb9QwEB61PcGBQstjaUFG4oKEd0Nsx_aRUlYtUhFCrbS3KI7HFYImFU0u_PqOnYcAAeIWRXYemvHMN57xNwAvLRl9bb3kAnXGpctrbq1HnhUhc0aqoNM57rOPxcmF_LBRmy14PZ-FQcRUfIbLeJly-b6t-7hVtiJrak0utmGbArfhrNacMYgNZ1NmU2muCfZPKcnMrs5PP8UaLr0kV10IGVvn5MYUhbRvfvFGqb3K35Fm8jjrXTibvnUoNPm67Du3rH_8RuP4vz9zH-6N0JO9HXTlAWxhswe7Iwxl4yK_2YO7P3EU7sNm3ccNNdYGdox4zUZC1ktWNZ5Fa5LYxz07bq-qLw2LkLKnW4zAMEvUV03Hj6o44vSKbBf7nFp4kX4_hIv1-_N3J3xsx8BrYVTHVXDogrLeBIrpUAaPSgQbqc6DU1UerA4iaK-QYjYbM8B14VBYEwrjrQ_iEew0bYNPgGkvTV7lucu8kEj-knBWZH8yBC5lLfUCVpNYynrkKo8tM76VKWbJbEkyLaNMy1GmC3g1z7geeDr-MXY_imMeN0piAS8myZe0xmLipGqw7W9KArkEbMnV5wt4PKjEPHnSpKd_fugB3ImvHgpgDmGn-97jM4IxnXue9PcWYPzp0A |
| linkProvider | IEEE |
| linkToHtml | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwjV1Lb9QwEB6VcoAeKLRQlqeRuCDh3TS2Y_sIlNUudCuEttLeoji2UVWaVDS59Nd37DwECBC3KLIjRzOe-eyZ-QbgtUajL7XllDmZUG7SkmptHU0ynxjFhZexjnt1ki1O-aeN2GzB27EWxjkXk8_cNDzGWL6tyzZclc3QmmqVsltwG_2-OOyqtcaYQWg5G2ObQlKJwH8ISiZ6tl5-CVlccorOOmM8NM9Jlcoyrg9_8UexwcrfsWb0OfNdWA2r7VJNzqdtY6bl9W9Ejv_7O_fhXg8-ybtOWx7Alqv2YLcHoqTf5ld7sPMTS-E-bOZtuFIjtSdHzl2SnpL1GykqS4I9ifzjlhzVF8VZRQKobPEVQThMIvlV1dD3RRixvEDrRb7GJl6o4Q_hdP5x_WFB-4YMtGRKNFR444wX2iqPpzrHvXWCeR3Izr0RReq19MxLKxye2nSIAZeZcUwrnymrrWePYLuqK_cYiLRcpUWamsQy7tBjItIK_E8K4SUvuZzAbBBLXvZs5aFpxvc8nloSnaNM8yDTvJfpBN6MMy47po5_jN0P4hjH9ZKYwKtB8jnushA6KSpXt1c5wlyEtujs0wkcdCoxTh406cmfP_oS7izWq-P8eHny-SncDcvo0mGewXbzo3XPEdQ05kXU5Ru9Hu0Z |
| 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=Fusion+of+Deep+Learning+and+Compressed+Domain+Features+for+Content-Based+Image+Retrieval&rft.jtitle=IEEE+transactions+on+image+processing&rft.au=Peizhong+Liu&rft.au=Jing-Ming+Guo&rft.au=Chi-Yi+Wu&rft.au=Danlin+Cai&rft.date=2017-12-01&rft.eissn=1941-0042&rft.volume=26&rft.issue=12&rft.spage=5706&rft_id=info:doi/10.1109%2FTIP.2017.2736343&rft_id=info%3Apmid%2F28866491&rft.externalDocID=28866491 |
| thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1057-7149&client=summon |
| thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1057-7149&client=summon |
| thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1057-7149&client=summon |