Cargando…
Scene Recognition Using Deep Softpool Capsule Network Based on Residual Diverse Branch Block
With the improvement of the quality and resolution of remote sensing (RS) images, scene recognition tasks have played an important role in the RS community. However, due to the special bird’s eye view image acquisition mode of imaging sensors, it is still challenging to construct a discriminate repr...
Autores principales: | , , , |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
MDPI
2021
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8402264/ https://www.ncbi.nlm.nih.gov/pubmed/34451017 http://dx.doi.org/10.3390/s21165575 |
_version_ | 1783745748287881216 |
---|---|
author | Wang, Chunyuan Wu, Yang Wang, Yihan Chen, Yiping |
author_facet | Wang, Chunyuan Wu, Yang Wang, Yihan Chen, Yiping |
author_sort | Wang, Chunyuan |
collection | PubMed |
description | With the improvement of the quality and resolution of remote sensing (RS) images, scene recognition tasks have played an important role in the RS community. However, due to the special bird’s eye view image acquisition mode of imaging sensors, it is still challenging to construct a discriminate representation of diverse and complex scenes to improve RS image recognition performance. Capsule networks that can learn the spatial relationship between the features in an image has a good image classification performance. However, the original capsule network is not suitable for images with a complex background. To address the above issues, this paper proposes a novel end-to-end capsule network termed DS-CapsNet, in which a new multi-scale feature enhancement module and a new Caps-SoftPool method are advanced by aggregating the advantageous attributes of the residual convolution architecture, Diverse Branch Block (DBB), Squeeze and Excitation (SE) block, and the Caps-SoftPool method. By using the residual DBB, multiscale features can be extracted and fused to recover a semantic strong feature representation. By adopting SE, the informative features are emphasized, and the less salient features are weakened. The new Caps-SoftPool method can reduce the number of parameters that are needed in order to prevent an over-fitting problem. The novel DS-CapsNet achieves a competitive and promising performance for RS image recognition by using high-quality and robust capsule representation. The extensive experiments on two challenging datasets, AID and NWPU-RESISC45, demonstrate the robustness and superiority of the proposed DS-CapsNet in scene recognition tasks. |
format | Online Article Text |
id | pubmed-8402264 |
institution | National Center for Biotechnology Information |
language | English |
publishDate | 2021 |
publisher | MDPI |
record_format | MEDLINE/PubMed |
spelling | pubmed-84022642021-08-29 Scene Recognition Using Deep Softpool Capsule Network Based on Residual Diverse Branch Block Wang, Chunyuan Wu, Yang Wang, Yihan Chen, Yiping Sensors (Basel) Article With the improvement of the quality and resolution of remote sensing (RS) images, scene recognition tasks have played an important role in the RS community. However, due to the special bird’s eye view image acquisition mode of imaging sensors, it is still challenging to construct a discriminate representation of diverse and complex scenes to improve RS image recognition performance. Capsule networks that can learn the spatial relationship between the features in an image has a good image classification performance. However, the original capsule network is not suitable for images with a complex background. To address the above issues, this paper proposes a novel end-to-end capsule network termed DS-CapsNet, in which a new multi-scale feature enhancement module and a new Caps-SoftPool method are advanced by aggregating the advantageous attributes of the residual convolution architecture, Diverse Branch Block (DBB), Squeeze and Excitation (SE) block, and the Caps-SoftPool method. By using the residual DBB, multiscale features can be extracted and fused to recover a semantic strong feature representation. By adopting SE, the informative features are emphasized, and the less salient features are weakened. The new Caps-SoftPool method can reduce the number of parameters that are needed in order to prevent an over-fitting problem. The novel DS-CapsNet achieves a competitive and promising performance for RS image recognition by using high-quality and robust capsule representation. The extensive experiments on two challenging datasets, AID and NWPU-RESISC45, demonstrate the robustness and superiority of the proposed DS-CapsNet in scene recognition tasks. MDPI 2021-08-19 /pmc/articles/PMC8402264/ /pubmed/34451017 http://dx.doi.org/10.3390/s21165575 Text en © 2021 by the authors. https://creativecommons.org/licenses/by/4.0/Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/). |
spellingShingle | Article Wang, Chunyuan Wu, Yang Wang, Yihan Chen, Yiping Scene Recognition Using Deep Softpool Capsule Network Based on Residual Diverse Branch Block |
title | Scene Recognition Using Deep Softpool Capsule Network Based on Residual Diverse Branch Block |
title_full | Scene Recognition Using Deep Softpool Capsule Network Based on Residual Diverse Branch Block |
title_fullStr | Scene Recognition Using Deep Softpool Capsule Network Based on Residual Diverse Branch Block |
title_full_unstemmed | Scene Recognition Using Deep Softpool Capsule Network Based on Residual Diverse Branch Block |
title_short | Scene Recognition Using Deep Softpool Capsule Network Based on Residual Diverse Branch Block |
title_sort | scene recognition using deep softpool capsule network based on residual diverse branch block |
topic | Article |
url | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8402264/ https://www.ncbi.nlm.nih.gov/pubmed/34451017 http://dx.doi.org/10.3390/s21165575 |
work_keys_str_mv | AT wangchunyuan scenerecognitionusingdeepsoftpoolcapsulenetworkbasedonresidualdiversebranchblock AT wuyang scenerecognitionusingdeepsoftpoolcapsulenetworkbasedonresidualdiversebranchblock AT wangyihan scenerecognitionusingdeepsoftpoolcapsulenetworkbasedonresidualdiversebranchblock AT chenyiping scenerecognitionusingdeepsoftpoolcapsulenetworkbasedonresidualdiversebranchblock |