Pattern recognition of soldier uniforms with dilated convolutions and a modified encoder-decoder neural network architecture

In this paper, we study a deep learning (DL)-based multimodal technology for military, surveillance, and defense applications based on a pixel-by-pixel classification of soldier's image dataset. We explore the acquisition of images from a remote tactical-robot to a ground station, where the det...

Full description

Saved in:
Bibliographic Details
Published in:Applied artificial intelligence Vol. 35; no. 6; pp. 476 - 487
Main Authors: Morocho-Cayamcela, Manuel Eugenio, Lim, Wansu
Format: Journal Article
Language:English
Published: Philadelphia Taylor & Francis 12.05.2021
Taylor & Francis Ltd
Taylor & Francis Group
Subjects:
ISSN:0883-9514, 1087-6545
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Abstract In this paper, we study a deep learning (DL)-based multimodal technology for military, surveillance, and defense applications based on a pixel-by-pixel classification of soldier's image dataset. We explore the acquisition of images from a remote tactical-robot to a ground station, where the detection and tracking of soldiers can help the operator to take actions or automate the tactical-robot in battlefield. The soldier detection is achieved by training a convolutional neural network to learn the patterns of the soldier's uniforms. Our CNN learns from the initial dataset and from the actions taken by the operator, as opposed to the old-fashioned and hard-coded image processing algorithms. Our system attains an accuracy of over 81% in distinguishing the specific soldier uniform and the background. These experimental results prove our hypothesis that dilated convolutions can increase the segmentation performance when compared with patch-based, and fully connected networks.
AbstractList In this paper, we study a deep learning (DL)-based multimodal technology for military, surveillance, and defense applications based on a pixel-by-pixel classification of soldier’s image dataset. We explore the acquisition of images from a remote tactical-robot to a ground station, where the detection and tracking of soldiers can help the operator to take actions or automate the tactical-robot in battlefield. The soldier detection is achieved by training a convolutional neural network to learn the patterns of the soldier’s uniforms. Our CNN learns from the initial dataset and from the actions taken by the operator, as opposed to the old-fashioned and hard-coded image processing algorithms. Our system attains an accuracy of over 81% in distinguishing the specific soldier uniform and the background. These experimental results prove our hypothesis that dilated convolutions can increase the segmentation performance when compared with patch-based, and fully connected networks.
Author Morocho-Cayamcela, Manuel Eugenio
Lim, Wansu
Author_xml – sequence: 1
  givenname: Manuel Eugenio
  orcidid: 0000-0002-4705-7923
  surname: Morocho-Cayamcela
  fullname: Morocho-Cayamcela, Manuel Eugenio
  organization: Smart Data Analysis Systems Group (SDAS) (www.sdas-group.com)
– sequence: 2
  givenname: Wansu
  orcidid: 0000-0003-2533-3496
  surname: Lim
  fullname: Lim, Wansu
  email: wansu.lim@kumoh.ac.kr
  organization: Mechanical and Electronic Convergence Engineering, Kumoh National Institute of Technology
BookMark eNqFUU1rFTEUDVLB1-pPEAKu55lkMpk3uFGK1UJBF7oON19tnvOSepOxFPzxZvqqCxe6yYHcc869nHNKTlJOnpCXnG0527HXbLfrp4HLrWCCb_nUXiGfkE0bjp0a5HBCNiunW0nPyGkpe8YYH0e-IT8_Q60eE0Vv83WKNeZEc6Alzy56pEuKIeOh0LtYb6iLM1TvqM3pR56XlVwoJEeBHrKLIbaZTzY7j53zD0iTXxDmBvUu4zcKaG9i9bYu6J-TpwHm4l884hn5evH-y_nH7urTh8vzd1edlQOvnQUVnBn7AL3pHThmFOemD6OUwSjDJz-CC5N0wdkggpsYVyCNcWK0AprqjFwefV2Gvb7FeAC81xmifvjIeK0Ba7Sz16AmI4VybFBKGi8mIYSygzFG-gl2rnm9OnrdYv6--FL1Pi-Y2vlatHinXdvOGms4sizmUtCHP1s502tp-ndpei1NP5bWdG_-0tlYYc25IsT5v-q3R3VMa2nQAp-drnA_ZwwIycai-39b_AL3NLY3
CitedBy_id crossref_primary_10_1080_09500340_2025_2484550
crossref_primary_10_1007_s00521_023_08419_x
crossref_primary_10_3390_app12146924
Cites_doi 10.1109/LCOMM.2020.2982887
10.1109/TWC.2020.2986202
10.1109/TGRS.2017.2740362
10.1109/TPAMI.2016.2644615
10.1109/ACCESS.2020.3023335
10.1109/IGARSS.2017.8127684
10.1049/el.2020.1977
10.1109/CVPR.2009.5206848
10.1109/CVPR.2011.5995316
10.1007/978-3-319-10602-1_48
10.1109/TPAMI.2016.2572683
10.1109/CVPR.2016.90
10.1109/PROC.1979.11328
10.1109/CVPR.2008.4587503
10.1007/s11042-017-4440-4
10.1109/TSMC.1973.4309314
10.1214/aoms/1177729586
10.1109/TKDE.2009.191
ContentType Journal Article
Copyright 2021 Taylor & Francis 2021
2021 Taylor & Francis
Copyright_xml – notice: 2021 Taylor & Francis 2021
– notice: 2021 Taylor & Francis
DBID AAYXX
CITATION
7SC
8FD
JQ2
L7M
L~C
L~D
DOA
DOI 10.1080/08839514.2021.1902124
DatabaseName 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
Directory of Open Access Journals
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: DOA
  name: DOAJ Directory of Open Access Journals
  url: https://www.doaj.org/
  sourceTypes: Open Website
DeliveryMethod fulltext_linktorsrc
Discipline Computer Science
EISSN 1087-6545
EndPage 487
ExternalDocumentID oai_doaj_org_article_a69b426d05664be292226c5bbb4e9a8d
10_1080_08839514_2021_1902124
1902124
Genre Research Article
GroupedDBID .4S
.7F
.DC
.QJ
0YH
23M
2DF
30N
4.4
5GY
5VS
8VB
AAENE
AAFWJ
AAJMT
ABCCY
ABDBF
ABFIM
ABHAV
ABIVO
ABPEM
ABTAI
ACGEJ
ACGFS
ACGOD
ACNCT
ACTIO
ACUHS
ADCVX
ADMLS
ADXPE
AEISY
AEMOZ
AENEX
AEOZL
AEPSL
AEYOC
AFKVX
AFPKN
AGMYJ
AHQJS
AIJEM
AIYEW
AJWEG
AKVCP
ALMA_UNASSIGNED_HOLDINGS
ALQZU
AQRUH
AQTUD
ARCSS
AVBZW
AWYRJ
BLEHA
CCCUG
CE4
CS3
DGEBU
DKSSO
EAP
EBR
EBS
EBU
ECS
EDO
EMK
EPL
EST
ESX
E~A
E~B
F5P
GTTXZ
H13
HF~
HZ~
H~9
H~P
I-F
IPNFZ
J.P
K1G
KYCEM
LJTGL
M4Z
MK~
NA5
NX~
O9-
P2P
PQQKQ
QWB
RIG
S-T
SNACF
TDBHL
TFL
TFW
TH9
TNC
TTHFI
TUS
TWF
UT5
UU3
ZL0
~S~
AAYXX
CITATION
7SC
8FD
JQ2
L7M
L~C
L~D
GROUPED_DOAJ
ID FETCH-LOGICAL-c451t-ca6fdb73fa3b3dad0b611b3f744fb6b19e7adf94dfdcf2fd9016a4bbd27c2afa3
IEDL.DBID DOA
ISICitedReferencesCount 4
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000635849700001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 0883-9514
IngestDate Fri Oct 03 12:53:28 EDT 2025
Sun Nov 09 06:26:06 EST 2025
Sat Nov 29 03:21:25 EST 2025
Tue Nov 18 22:24:47 EST 2025
Mon Oct 20 23:48:26 EDT 2025
IsDoiOpenAccess true
IsOpenAccess true
IsPeerReviewed true
IsScholarly true
Issue 6
Language English
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-c451t-ca6fdb73fa3b3dad0b611b3f744fb6b19e7adf94dfdcf2fd9016a4bbd27c2afa3
Notes ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ORCID 0000-0002-4705-7923
0000-0003-2533-3496
OpenAccessLink https://doaj.org/article/a69b426d05664be292226c5bbb4e9a8d
PQID 2514989010
PQPubID 53050
PageCount 12
ParticipantIDs doaj_primary_oai_doaj_org_article_a69b426d05664be292226c5bbb4e9a8d
crossref_primary_10_1080_08839514_2021_1902124
proquest_journals_2514989010
informaworld_taylorfrancis_310_1080_08839514_2021_1902124
crossref_citationtrail_10_1080_08839514_2021_1902124
PublicationCentury 2000
PublicationDate 2021-05-12
PublicationDateYYYYMMDD 2021-05-12
PublicationDate_xml – month: 05
  year: 2021
  text: 2021-05-12
  day: 12
PublicationDecade 2020
PublicationPlace Philadelphia
PublicationPlace_xml – name: Philadelphia
PublicationTitle Applied artificial intelligence
PublicationYear 2021
Publisher Taylor & Francis
Taylor & Francis Ltd
Taylor & Francis Group
Publisher_xml – name: Taylor & Francis
– name: Taylor & Francis Ltd
– name: Taylor & Francis Group
References cit0011
cit0012
cit0010
Geron A. (cit0006) 2017
Mitchell T. M. (cit0016) 1997
Morocho-Cayamcela M. (cit0018) 2017; 12
cit0019
cit0017
Goodfellow I. (cit0008) 2016
cit0015
cit0013
cit0014
cit0022
cit0001
cit0023
cit0020
cit0021
Cauchy M. A. (cit0002) 1976; 25
cit0009
cit0004
cit0026
Everingham M. (cit0005) 2012
Glorot X. (cit0007) 2010; 9
cit0024
cit0003
cit0025
References_xml – ident: cit0017
  doi: 10.1109/LCOMM.2020.2982887
– ident: cit0013
  doi: 10.1109/TWC.2020.2986202
– volume: 12
  start-page: 180
  volume-title: 2017 International Conference on Next Generation Computing (ICNGC)
  year: 2017
  ident: cit0018
– ident: cit0015
  doi: 10.1109/TGRS.2017.2740362
– ident: cit0001
  doi: 10.1109/TPAMI.2016.2644615
– ident: cit0023
  doi: 10.1109/ACCESS.2020.3023335
– volume: 9
  start-page: 249
  volume-title: Proceedings of the 13th International Conference on Artificial Intelligence and Statistics (AISTATS), Chia Laguna Resort, Sardinia, Italy
  year: 2010
  ident: cit0007
– volume-title: Machine Learning
  year: 1997
  ident: cit0016
– volume: 25
  start-page: 536
  issue: 10
  year: 1976
  ident: cit0002
  publication-title: Comptes Rendus Hedb. Séances Academic Science
– ident: cit0014
  doi: 10.1109/IGARSS.2017.8127684
– ident: cit0019
  doi: 10.1049/el.2020.1977
– ident: cit0004
  doi: 10.1109/CVPR.2009.5206848
– ident: cit0024
  doi: 10.1109/CVPR.2011.5995316
– ident: cit0012
  doi: 10.1007/978-3-319-10602-1_48
– volume-title: Hands-on machine learning with scikit-learn and tensorflow: concepts, tools, and techniques to build intelligent systems
  year: 2017
  ident: cit0006
– ident: cit0022
  doi: 10.1109/TPAMI.2016.2572683
– ident: cit0011
  doi: 10.1109/CVPR.2016.90
– ident: cit0003
– volume-title: Deep Learning
  year: 2016
  ident: cit0008
– year: 2012
  ident: cit0005
  publication-title: International Journal of Computer Vision
– ident: cit0009
  doi: 10.1109/PROC.1979.11328
– ident: cit0025
  doi: 10.1109/CVPR.2008.4587503
– ident: cit0026
  doi: 10.1007/s11042-017-4440-4
– ident: cit0010
  doi: 10.1109/TSMC.1973.4309314
– ident: cit0021
  doi: 10.1214/aoms/1177729586
– ident: cit0020
  doi: 10.1109/TKDE.2009.191
SSID ssj0001771
Score 2.2793407
Snippet In this paper, we study a deep learning (DL)-based multimodal technology for military, surveillance, and defense applications based on a pixel-by-pixel...
SourceID doaj
proquest
crossref
informaworld
SourceType Open Website
Aggregation Database
Enrichment Source
Index Database
Publisher
StartPage 476
SubjectTerms Algorithms
Artificial neural networks
Battlefields
Coders
Computer architecture
Datasets
Encoders-Decoders
Ground stations
Image acquisition
Image classification
Image processing
Image segmentation
Machine learning
Military applications
Military technology
Neural networks
Pattern recognition
Pixels
Robots
SummonAdditionalLinks – databaseName: Taylor & Francis Online Journals
  dbid: TFW
  link: http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwrV3BbtQwELVQxYFLC7QVW0rlA9eUdeJN4iMgVhxQ1UMpvVkej41WahO02XLi45lxnFUBVT1QKVKUKGM5zsz42X55FuItjRkiofgFa4bGQtfzqgBVhaKEVs2xRdCJm3P5pTk7a6-uzHlmEw6ZVslj6DgKRaRczcHtYJgYce8oMCoCBjwjUqpT6tEo_bIiKHX9HJoXy2_bXKyaNORii4JNpn947ivlj94pifj_JWH6T8pO_dBy7xHe4LnYzSBUvh-95oV4ErqXYm_a4EHmeN8Xv86T-mYntzSjvpN9lAMvWdGDtx3_13UzSJ7Nlbi6JuCKkonsk0NLqqJ08qbHVSSwK1k2E8O6wJDOkuU0qSbdSEaXd5c1DsTX5aeLj5-LvF1D4fVCbQrv6ojQVNFVUKHDOdRKQRUbrSPUoExoHEajMaKPZURCIrXTAFg2vnRkdSh2ur4Lr4QslTeejkhlaAPQOsJ9HqOOLkJoFjOhp89kfdYy5y01rq2aJE9zC1tuYZtbeCZOt2Y_RjGPhww-sA9sH2Yt7nSjX3-3ObStqw0QzkGCkrWGUBqCXLVfAIAOxrU4E-auB9lNmoqJ474ptnqgAseTu9mcXAZLkFSblnk1R_9R9GvxjC-ZCaHKY7GzWd-GN-Kp_7lZDeuTFEa_AUHpG5g
  priority: 102
  providerName: Taylor & Francis
Title Pattern recognition of soldier uniforms with dilated convolutions and a modified encoder-decoder neural network architecture
URI https://www.tandfonline.com/doi/abs/10.1080/08839514.2021.1902124
https://www.proquest.com/docview/2514989010
https://doaj.org/article/a69b426d05664be292226c5bbb4e9a8d
Volume 35
WOSCitedRecordID wos000635849700001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
journalDatabaseRights – providerCode: PRVAWR
  databaseName: Taylor & Francis Online Journals
  customDbUrl:
  eissn: 1087-6545
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0001771
  issn: 0883-9514
  databaseCode: TFW
  dateStart: 19870101
  isFulltext: true
  titleUrlDefault: https://www.tandfonline.com
  providerName: Taylor & Francis
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwrV1Lj9MwELbQigMX3ojCgnzgmqVO3CQ-AqLigFZ7WGBvlsfjkSrtpqjt7okfz4zjVAUOvSBFiWTFycgznoc9_kapdxwzEHvxC8EMpcq286YC06Sqht7MsUewOTfn-9fu_Ly_unIXB6W-JCdshAceB-59aB2wFUE21K2FVDs2aG1cAIBNLvQo2nfeuSmYKjrYdDnU4inUVOxD2OnsjqBqc5s0cWxYmzO2h6y87R9WKYP3_wVd-o-qzvZn-Vg9LI6j_jAS_ETdS8NT9WgqyqDLHH2mfl1kxMxB71OD1oNek97KNhO_eDvIWaybrZYVWI2ra3Y2UUvy-SSEOgyog75Z44rYQdUCdYlpU2HKTy0QmEzJMCaQ68OtiOfq2_Lz5acvVSmxUEW7MLsqhpYQuoZCAw0GnENrDDTUWUvQgnGpC0jOImGkmpC9hzZYAKy7WAfu9UKdDOshvVS6NtFFvoi_YR1AH9hXi0iWAkHqFjNlpyH2seCPSxmMa28mmNLCGS-c8YUzM3W27_ZzBOA41uGj8G__suBn5waWKl-kyh-Tqplyh9z3u7x8QmOtE98cIeB0EhVfFMLWsxtpXS-5MK_-B32v1QP5paQxmPpUnew2t-mNuh_vdqvt5m2eC3y_XP74DUfHDJk
linkProvider Directory of Open Access Journals
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwrV1Lj9MwELZgQYILy1MUFvCBa5Y6cR4-AqJaRKk4FNib5bE9q0q7CWq7nPjxzDhJVUBoDyBFipR4LMeZGX8ejz8L8ZLmDEgovmTOUMx0NS0yUEXMcmjUNDQBdMrN-TKvF4vm9NTs74XhtEqeQ2NPFJF8NRs3B6PHlLhXZBkFIQMOieTqmIY08r_6urhR0ljL_PnL2dedN1Z1mnSxSMYy4y6ev1Xzy_iUaPx_IzH9w2mnkWh2-D--4a64M-BQ-bpXnHviWmzvi8PxjAc5mPwD8eNTIuBs5S7TqGtlh3LDq1ZU8LLlrV0XG8kBXRlW54Rdg-Rc9lGnJbVROnnRhRUS3pXMnBniOgsx3SUzalJL2j4fXe6vbDwUn2fvlm9PsuHEhszrUm0z7yoMUBfoCiiCC1OolIICa60RKlAm1i6g0QGDxxwDgZHKaYCQ1z53JPVIHLRdGx8LmStvPF1IdWgD0DiCfj6gRocQ63Ii9PifrB_ozPlUjXOrRtbToYct97Adengijndi33o-j6sE3rAS7AozHXd60K3P7GDd1lUGCOoEQpOVhpgbQl2VLwFAR-OaMBFmX4XsNkVjsD86xRZXNOBo1Dc7-JeNJVSqTcOpNU_-oeoX4tbJ8uPczt8vPjwVt_kVJ0ao_EgcbNeX8Zm46b9vV5v182RTPwEelR_C
linkToPdf http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwrV1Nb9QwELWgIMSF8ikWCvjANWWdeJP4yNcKRLXaQ4HeLI_tQSu1SbXZcuLHM-N1VgWEegApUqQkYznOzPjZfnkW4iWNGZBQ_Iw1Q7HQ9bQqQFWxKKFV09AG0Imb8-WoWSzakxOzzGzCIdMqeQyNW6GIlKs5uM8Djoy4VxQYFQEDnhEp1SH1aJR-9XVxg6BzzU5-PP-6S8aqSWMuNinYZvyJ52_F_NI9JRX_3zRM_8jZqSOa7_-HV7gr7mQUKl9v3eaeuBa7-2J_3OFB5oB_IH4sk_xmJ3c8o76TPcqB16zowYuOf-w6GyRP58qwOiXkGiQz2UePllRF6eRZH1ZIaFeybmaI6yLEdJasp0k16bZsdHl5XeOh-Dx_f_z2Q5H3ayi8nqlN4V2NAZoKXQVVcGEKtVJQYaM1Qg3KxMYFNDpg8FhiIChSOw0QysaXjqweib2u7-JjIUvljacDqQxtAFpHwM8H1OgQYjObCD1-JuuzmDnvqXFq1ah5mlvYcgvb3MITcbgzO9-qeVxl8IZ9YPcwi3GnC_36m82xbV1tgIBOICxZa4ilIcxV-xkA6GhcGybCXPYgu0lzMbjdOMVWV1TgYHQ3m7PLYAmTatMysebJPxT9Qtxavpvbo4-LT0_Fbb7DrAhVHoi9zfoiPhM3_ffNalg_TxH1E6XyHnQ
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=Pattern+recognition+of+soldier+uniforms+with+dilated+convolutions+and+a+modified+encoder-decoder+neural+network+architecture&rft.jtitle=Applied+artificial+intelligence&rft.au=Morocho-Cayamcela%2C+Manuel+Eugenio&rft.au=Lim%2C+Wansu&rft.date=2021-05-12&rft.pub=Taylor+%26+Francis+Ltd&rft.issn=0883-9514&rft.eissn=1087-6545&rft.volume=35&rft.issue=6&rft.spage=476&rft.epage=487&rft_id=info:doi/10.1080%2F08839514.2021.1902124&rft.externalDBID=NO_FULL_TEXT
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0883-9514&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0883-9514&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0883-9514&client=summon