Hybrid deep learning model for image de-noising and de-mosaicking with adaptive Gannet optimization algorithm

Image reconstruction is a critical step in various applications, such as art restoration, medical image processing, and agriculture, but it faces challenges due to noise and mosaic artefacts. In this research, a novel approach is introduced for de-noising and de-mosaicking images to enhance image re...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Network (Bristol) s. 1 - 27
Hlavní autoři: K, John Peter, S.R, SylajaVallee Narayan, N, Muthuvairavan Pillai, S.P, Predeep Kumar
Médium: Journal Article
Jazyk:angličtina
Vydáno: England 23.07.2025
Témata:
ISSN:0954-898X, 1361-6536, 1361-6536
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Abstract Image reconstruction is a critical step in various applications, such as art restoration, medical image processing, and agriculture, but it faces challenges due to noise and mosaic artefacts. In this research, a novel approach is introduced for de-noising and de-mosaicking images to enhance image reconstruction quality. The proposed model consists of three main steps: detail layer extraction, image de-noising using an Efficient Generative Adversarial Network (E-GAN), and de-mosaicking using an Adaptive Gannet-based Residual DenseNet (AG_DenseResNet). The publicly available Kodak dataset is utilized for the evaluation of the proposed model. The results show that the proposed outperforms conventional methods in terms of Peak Signal-to-Noise Ratio (PSNR), Structural Similarity Index (SSIM), Mean Squared Error (MSE), and Learned Perceptual Image Patch Similarity (LPIPS) and acquired the values of 53.93, 0.98, 2.76, and 0.23, respectively.
AbstractList Image reconstruction is a critical step in various applications, such as art restoration, medical image processing, and agriculture, but it faces challenges due to noise and mosaic artefacts. In this research, a novel approach is introduced for de-noising and de-mosaicking images to enhance image reconstruction quality. The proposed model consists of three main steps: detail layer extraction, image de-noising using an Efficient Generative Adversarial Network (E-GAN), and de-mosaicking using an Adaptive Gannet-based Residual DenseNet (AG_DenseResNet). The publicly available Kodak dataset is utilized for the evaluation of the proposed model. The results show that the proposed outperforms conventional methods in terms of Peak Signal-to-Noise Ratio (PSNR), Structural Similarity Index (SSIM), Mean Squared Error (MSE), and Learned Perceptual Image Patch Similarity (LPIPS) and acquired the values of 53.93, 0.98, 2.76, and 0.23, respectively.Image reconstruction is a critical step in various applications, such as art restoration, medical image processing, and agriculture, but it faces challenges due to noise and mosaic artefacts. In this research, a novel approach is introduced for de-noising and de-mosaicking images to enhance image reconstruction quality. The proposed model consists of three main steps: detail layer extraction, image de-noising using an Efficient Generative Adversarial Network (E-GAN), and de-mosaicking using an Adaptive Gannet-based Residual DenseNet (AG_DenseResNet). The publicly available Kodak dataset is utilized for the evaluation of the proposed model. The results show that the proposed outperforms conventional methods in terms of Peak Signal-to-Noise Ratio (PSNR), Structural Similarity Index (SSIM), Mean Squared Error (MSE), and Learned Perceptual Image Patch Similarity (LPIPS) and acquired the values of 53.93, 0.98, 2.76, and 0.23, respectively.
Image reconstruction is a critical step in various applications, such as art restoration, medical image processing, and agriculture, but it faces challenges due to noise and mosaic artefacts. In this research, a novel approach is introduced for de-noising and de-mosaicking images to enhance image reconstruction quality. The proposed model consists of three main steps: detail layer extraction, image de-noising using an Efficient Generative Adversarial Network (E-GAN), and de-mosaicking using an Adaptive Gannet-based Residual DenseNet (AG_DenseResNet). The publicly available Kodak dataset is utilized for the evaluation of the proposed model. The results show that the proposed outperforms conventional methods in terms of Peak Signal-to-Noise Ratio (PSNR), Structural Similarity Index (SSIM), Mean Squared Error (MSE), and Learned Perceptual Image Patch Similarity (LPIPS) and acquired the values of 53.93, 0.98, 2.76, and 0.23, respectively.
Author S.R, SylajaVallee Narayan
S.P, Predeep Kumar
N, Muthuvairavan Pillai
K, John Peter
Author_xml – sequence: 1
  givenname: John Peter
  surname: K
  fullname: K, John Peter
– sequence: 2
  givenname: SylajaVallee Narayan
  surname: S.R
  fullname: S.R, SylajaVallee Narayan
– sequence: 3
  givenname: Muthuvairavan Pillai
  surname: N
  fullname: N, Muthuvairavan Pillai
– sequence: 4
  givenname: Predeep Kumar
  surname: S.P
  fullname: S.P, Predeep Kumar
BackLink https://www.ncbi.nlm.nih.gov/pubmed/40697071$$D View this record in MEDLINE/PubMed
BookMark eNo90U1PwzAMBuAIDcEY_ARQjlw6UqdfOaIJGNIkLiBxq9zUHYE2GUkHGr-eVmycIsePLNnvGZtYZ4mxy1jMY1GIG6HSpFDF6xwEpHNIQYFSR2wayyyOslRmEzYdTTSiU3YWwrsQIodcnrDTRGQqF3k8Zd1yV3lT85pow1tCb41d887V1PLGeW46XNPQjawzYWyhHXHUuYBGf4w_36Z_41jjpjdfxB_QWuq5G6rO_GBvnOXYrp0fVHfOjhtsA13s3xl7ub97Xiyj1dPD4-J2FWlIVR9VCaGUCRLEStVY6AwaUqrIE0GV0FLISgPqTCEkuaa8gCaVUAFopXKFWs7Y9d_cjXefWwp92ZmgqW3RktuGUoKE4XoxFAO92tNt1VFdbvywst-VhxMNIP0D2rsQPDX_JBblGEV5iKIcoyj3UchfX_983w
Cites_doi 10.1109/ACCESS.2022.3192451
10.1016/j.engappai.2022.105373
10.1364/OE.483268
10.3390/jimaging8110295
10.1109/TIP.2022.3211471
10.1109/ACCESS.2021.3086984
10.1155/2022/6200931
10.3390/technologies11040111
10.1364/OE.460495
10.1109/TCYB.2021.3124231
10.1109/ICIP49359.2023.10221948
10.1109/TII.2023.3316180
10.3390/bioengineering11040399
10.3390/s21093265
10.1109/JSTSP.2022.3172865
10.3390/electronics12143038
10.3390/s22051767
10.3934/mbe.2023329
10.1137/23M1545859
10.1109/OJSP.2022.3172842
10.23919/cje.2022.00.414
10.1016/j.matcom.2022.06.007
10.1007/s11633-023-1466-0
10.1016/j.cam.2023.115330
ContentType Journal Article
DBID AAYXX
CITATION
NPM
7X8
DOI 10.1080/0954898X.2025.2529299
DatabaseName CrossRef
PubMed
MEDLINE - Academic
DatabaseTitle CrossRef
PubMed
MEDLINE - Academic
DatabaseTitleList MEDLINE - Academic
PubMed
Database_xml – sequence: 1
  dbid: NPM
  name: PubMed
  url: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed
  sourceTypes: Index Database
– sequence: 2
  dbid: 7X8
  name: MEDLINE - Academic
  url: https://search.proquest.com/medline
  sourceTypes: Aggregation Database
DeliveryMethod fulltext_linktorsrc
Discipline Engineering
Mathematics
Computer Science
EISSN 1361-6536
EndPage 27
ExternalDocumentID 40697071
10_1080_0954898X_2025_2529299
Genre Journal Article
GroupedDBID ---
-~X
.4S
.DC
00X
03L
0R~
123
29N
36B
4.4
AAGDL
AALUX
AAMIU
AAPUL
AAQRR
AAYXX
ABBKH
ABEIZ
ABIVO
ABJNI
ABLIJ
ABLKL
ABUPF
ABWVI
ABXYU
ACENM
ACGEJ
ACGFS
ACIEZ
ADCVX
ADRBQ
ADXPE
AECIN
AEOZL
AFKVX
AFRVT
AGDLA
AGFJD
AGRBW
AGYJP
AIJEM
AIRBT
AJWEG
AKBVH
ALMA_UNASSIGNED_HOLDINGS
ALQZU
ALYBC
AMDAE
BABNJ
BLEHA
BOHLJ
CCCUG
CITATION
CS3
DKSSO
EBS
EMB
F5P
H13
HZ~
KRBQP
KWAYT
KYCEM
M4Z
O9-
P2P
RNANH
RO9
RVRKI
TASJS
TBQAZ
TDBHL
TERGH
TFDNU
TFL
TFW
TUROJ
TUS
UEQFS
V1S
~1N
ADYSH
NPM
7X8
ID FETCH-LOGICAL-c259t-b4ea334ae2199da8c62fe998740eb0c303bc2ac69a247ce782f532b22c9979ac3
ISICitedReferencesCount 0
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001533729500001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 0954-898X
1361-6536
IngestDate Fri Sep 05 15:42:15 EDT 2025
Thu Jul 24 02:15:12 EDT 2025
Sat Nov 29 07:42:09 EST 2025
IsPeerReviewed true
IsScholarly true
Keywords De-noising
deep learning
de-mosaicking
image reconstruction
generative adversarial learning
Language English
LinkModel OpenURL
MergedId FETCHMERGED-LOGICAL-c259t-b4ea334ae2199da8c62fe998740eb0c303bc2ac69a247ce782f532b22c9979ac3
Notes ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
PMID 40697071
PQID 3232489128
PQPubID 23479
PageCount 27
ParticipantIDs proquest_miscellaneous_3232489128
pubmed_primary_40697071
crossref_primary_10_1080_0954898X_2025_2529299
PublicationCentury 2000
PublicationDate 2025-Jul-23
PublicationDateYYYYMMDD 2025-07-23
PublicationDate_xml – month: 07
  year: 2025
  text: 2025-Jul-23
  day: 23
PublicationDecade 2020
PublicationPlace England
PublicationPlace_xml – name: England
PublicationTitle Network (Bristol)
PublicationTitleAlternate Network
PublicationYear 2025
References Huang Y (e_1_3_2_8_1) 2023
e_1_3_2_27_1
e_1_3_2_28_1
e_1_3_2_29_1
Chen S (e_1_3_2_3_1) 2023
e_1_3_2_20_1
e_1_3_2_21_1
e_1_3_2_22_1
Qian G (e_1_3_2_23_1) 2022
Rao CR (e_1_3_2_24_1) 2022; 13
e_1_3_2_26_1
Guo S (e_1_3_2_5_1) 2022
e_1_3_2_16_1
e_1_3_2_9_1
e_1_3_2_18_1
e_1_3_2_7_1
Koshelev I (e_1_3_2_14_1) 2023
e_1_3_2_19_1
Chen H (e_1_3_2_2_1) 2023
e_1_3_2_31_1
e_1_3_2_30_1
e_1_3_2_10_1
e_1_3_2_33_1
e_1_3_2_32_1
e_1_3_2_6_1
e_1_3_2_12_1
Samuel S (e_1_3_2_25_1) 2022
e_1_3_2_35_1
e_1_3_2_13_1
e_1_3_2_34_1
e_1_3_2_4_1
Kerepecký T (e_1_3_2_11_1) 2023
e_1_3_2_15_1
Li Y (e_1_3_2_17_1) 2023
References_xml – ident: e_1_3_2_12_1
  doi: 10.1109/ACCESS.2022.3192451
– ident: e_1_3_2_21_1
  doi: 10.1016/j.engappai.2022.105373
– ident: e_1_3_2_28_1
  doi: 10.1364/OE.483268
– ident: e_1_3_2_7_1
  doi: 10.3390/jimaging8110295
– ident: e_1_3_2_32_1
  doi: 10.1109/TIP.2022.3211471
– start-page: 1692
  volume-title: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition
  year: 2023
  ident: e_1_3_2_2_1
– ident: e_1_3_2_16_1
  doi: 10.1109/ACCESS.2021.3086984
– ident: e_1_3_2_13_1
– ident: e_1_3_2_26_1
  doi: 10.1155/2022/6200931
– ident: e_1_3_2_10_1
  doi: 10.3390/technologies11040111
– ident: e_1_3_2_20_1
  doi: 10.1364/OE.460495
– ident: e_1_3_2_33_1
  doi: 10.1109/TCYB.2021.3124231
– start-page: 379
  volume-title: Microelectronics, Communication Systems, Machine Learning and Internet of Things: Select Proceedings of MCMI 2020
  year: 2022
  ident: e_1_3_2_25_1
– start-page: 1735
  volume-title: 2023 IEEE International Conference on Image Processing (ICIP)
  year: 2023
  ident: e_1_3_2_11_1
  doi: 10.1109/ICIP49359.2023.10221948
– ident: e_1_3_2_34_1
  doi: 10.1109/TII.2023.3316180
– start-page: 18278
  volume-title: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition
  year: 2023
  ident: e_1_3_2_17_1
– ident: e_1_3_2_18_1
  doi: 10.3390/bioengineering11040399
– year: 2023
  ident: e_1_3_2_3_1
  article-title: Exploring asymmetric tunable blind-spots for self-supervised denoising in real-world scenarios
  publication-title: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR); June 16–22, 2024; Seattle, Washington, USA
– ident: e_1_3_2_27_1
  doi: 10.3390/s21093265
– ident: e_1_3_2_19_1
  doi: 10.1109/JSTSP.2022.3172865
– volume: 13
  start-page: 665
  issue: 1
  year: 2022
  ident: e_1_3_2_24_1
  article-title: Hybrid color image demosaicking using densely connected residual sub-pixel CNN with iterative ring resonator-based Gaussian filter
  publication-title: J Algebr Stat
– ident: e_1_3_2_30_1
  doi: 10.3390/electronics12143038
– ident: e_1_3_2_15_1
  doi: 10.3390/s22051767
– ident: e_1_3_2_29_1
  doi: 10.3934/mbe.2023329
– year: 2023
  ident: e_1_3_2_14_1
  article-title: Iterative reweighted least squares networks with convergence guarantees for solving inverse imaging problems
  publication-title: In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition
– start-page: 188
  volume-title: In Proceedings of the IEEE International Conference on Multimedia and Expo Workshops (ICMEW Workshops)
  year: 2023
  ident: e_1_3_2_8_1
– ident: e_1_3_2_4_1
  doi: 10.1137/23M1545859
– start-page: 17472
  volume-title: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition
  year: 2022
  ident: e_1_3_2_5_1
– ident: e_1_3_2_9_1
  doi: 10.1109/OJSP.2022.3172842
– ident: e_1_3_2_35_1
  doi: 10.23919/cje.2022.00.414
– ident: e_1_3_2_22_1
  doi: 10.1016/j.matcom.2022.06.007
– ident: e_1_3_2_31_1
  doi: 10.1007/s11633-023-1466-0
– ident: e_1_3_2_6_1
  doi: 10.1016/j.cam.2023.115330
– start-page: 1
  volume-title: 2022 IEEE International Conference on Computational Photography (ICCP)
  year: 2022
  ident: e_1_3_2_23_1
SSID ssj0007273
Score 2.4011495
SecondaryResourceType online_first
Snippet Image reconstruction is a critical step in various applications, such as art restoration, medical image processing, and agriculture, but it faces challenges...
SourceID proquest
pubmed
crossref
SourceType Aggregation Database
Index Database
StartPage 1
Title Hybrid deep learning model for image de-noising and de-mosaicking with adaptive Gannet optimization algorithm
URI https://www.ncbi.nlm.nih.gov/pubmed/40697071
https://www.proquest.com/docview/3232489128
WOSCitedRecordID wos001533729500001&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 Journals Complete
  customDbUrl:
  eissn: 1361-6536
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0007273
  issn: 0954-898X
  databaseCode: TFW
  dateStart: 19900101
  isFulltext: true
  titleUrlDefault: https://www.tandfonline.com
  providerName: Taylor & Francis
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV1bb9MwFLa6gdB44DJu5TIZibcqVepcHD8CYkxCqyqtQN8ix3G2QHNR0lbrH-B3c-zEaYo2CR54iWonclV_n06-454LQu_8mLrSo8SSPmOWK0DDRTBnJcxO_CCRThIlutkEnU6DxYLNBoNfJhdms6R5Hlxfs_K_Qg1zALZKnf0HuLtFYQI-A-hwBdjh-lfAn21VEtYolrI0PSEum4Y3OqQwzVSUTiytvEhrk6EIw6yoeSp-dmezPOalDiv6zFXmzqiAUdYmbY748rKo4Kmsr22nTUS50qwfdMWCZe-Y4YuJ-92PCL4Y6UCMi-2S_-DfVF8XCQa_4tsdaXUcwvl6dbXe8LTiKt9qplolpbs1ZloLV1L_Zh003j_MIJ46JW3yjceyMcCOP7F8z-kb1cmNpr6NjVQF61iwGKvVxsQjIPdY_3kAp8w01CrHl9pNw5c_amybWwfoDqEeU6GB89Pv3RtdaTyT_aXqst_0nUfonlllX-Lc4rdo_TJ_hB60jgd-3xDmMRrI_Bg9NE09cGvjj9H9XoVKGJ13ZX3rJyhruIXVPmPDLay5hYFbWHML77iFgVt4j1tYcQsbbuGGW7jPLdxx6yn6evpp_vHMavt1WAKc6JUVuZI7jsslvAVZzAPhk0SCO09dW0a2ALEUCcKFzzhxqZCgTRPPIREhgjHKuHCeocO8yOULhH3HpeBpB7Dz0g1in0e254nYp8ILGJiWIRqbLQ7LpixLODHVblt4QgVP2MIzRG8NECEYUPWvGM9lsa5DR_kUAQOdNkTPG4S6JQ2iL2-98wod7Xj8Gh2uqrV8g-6KzSqtqxN0QBfBiSbTb3wmlII
linkProvider Taylor & Francis
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=Hybrid+deep+learning+model+for+image+de-noising+and+de-mosaicking+with+adaptive+Gannet+optimization+algorithm&rft.jtitle=Network+%28Bristol%29&rft.au=K%2C+John+Peter&rft.au=S+R%2C+SylajaVallee+Narayan&rft.au=N%2C+Muthuvairavan+Pillai&rft.au=S+P%2C+Predeep+Kumar&rft.date=2025-07-23&rft.eissn=1361-6536&rft.spage=1&rft_id=info:doi/10.1080%2F0954898X.2025.2529299&rft_id=info%3Apmid%2F40697071&rft.externalDocID=40697071
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0954-898X&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0954-898X&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0954-898X&client=summon