Connexion
S'inscrire
Réinitialiser le mot de passe
Publier & Distribuer
Solutions d'édition
Solutions de distribution
Thèmes
Architecture et design
Arts
Business et économie
Chimie
Chimie industrielle
Droit
Géosciences
Histoire
Informatique
Ingénierie
Intérêt général
Linguistique et sémiotique
Littérature
Mathématiques
Musique
Médecine
Pharmacie
Philosophie
Physique
Sciences bibliothécaires et de l'information, études du livre
Sciences des matériaux
Sciences du vivant
Sciences sociales
Sport et loisirs
Théologie et religion
Études classiques et du Proche-Orient ancient
Études culturelles
Études juives
Publications
Journaux
Livres
Comptes-rendus
Éditeurs
Blog
Contact
Chercher
EUR
USD
GBP
Français
English
Deutsch
Polski
Español
Français
Italiano
Panier
Home
Journaux
International Journal of Advanced Network, Monitoring and Controls
Édition 7 (2022): Edition 4 (January 2022)
Accès libre
Super-resolution Reconstruction Based on Capsule Generative Adversarial Network
Ziyi Wu
Ziyi Wu
,
Hongge Yao
Hongge Yao
,
Hualong Yang
Hualong Yang
,
Hong Jiang
Hong Jiang
,
Wei Zhang
Wei Zhang
et
Jun Yu
Jun Yu
| 26 mai 2023
International Journal of Advanced Network, Monitoring and Controls
Édition 7 (2022): Edition 4 (January 2022)
À propos de cet article
Article précédent
Article suivant
Résumé
Article
Figures et tableaux
Références
Auteurs
Articles dans cette édition
Aperçu
PDF
Citez
Partagez
Publié en ligne:
26 mai 2023
Pages:
69 - 81
DOI:
https://doi.org/10.2478/ijanmc-2022-0038
Mots clés
Generative Adversarial Network
,
Capsule Network
,
Capsule Generative Adversarial Network
,
Capsule Discriminator
,
Super-resolution Reconstruction
© 2022 Ziyi Wu et al., published by Sciendo
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
Figure 1.
Structure diagram of CSRGAN.
Figure 2.
Structure diagram of RDB.
Figure 3.
u is n 8-dimensional vectors obtained by encapsulating and flattening the data through the PrimaryCaps layer, v1 is a 16-dimensional digital capsule vector after a routing process and v2 is the 32-dimensional vector output at the end of the second routing process. wij is the fully connected weight matrix of the connected vector. The vectors u' and v' are the prediction vectors of u and v1 obtained through a linear hierarchical relationship, the subscripts i, j correspond to the number of vector u and the next layer vector v1, respectively. The parameter bij is the log prior probability from the lower capsule to the upper capsule. The parameter bij is the logarithmic prior probability from the low-level capsule to the high-level capsule and is used to update each cij correspondingly. cij is the coupling coefficient connecting the two layers' vectors before and after, representing the degree of correlation between the i-th vector of the previous layer and the j-th vector of the next layer. Squash is a linear rectification function, which normalizes the results obtained.
Figure 4.
Panda.
Figure 5.
Rome.
Figure 6.
Nut.
Figure 7.
Train PSNR is evaluated on Set5.
Figure 8.
The Roman Colosseum(a).
Figure 9.
The Roman Colosseum(b).
Figure 10.
castle(a).
Figure 11.
castle(b).
PSNR and SSIM
Method
Bicubic
SRCNN
SRGAN
RDN
CSRGAN
PSNR
28.39
30.45
29.43
32.44
32.58
SSIM
0.8102
0. 8616
0.8477
0.8988
0.9003