A Grad-CAM and capsule network hybrid method for remote sensing image scene classification

IF 1.8 4区 地球科学 Q3 GEOSCIENCES, MULTIDISCIPLINARY Frontiers of Earth Science Pub Date : 2024-07-04 DOI:10.1007/s11707-022-1079-x
Zhan He, Chunju Zhang, Shu Wang, Jianwei Huang, Xiaoyun Zheng, Weijie Jiang, Jiachen Bo, Yucheng Yang
{"title":"A Grad-CAM and capsule network hybrid method for remote sensing image scene classification","authors":"Zhan He, Chunju Zhang, Shu Wang, Jianwei Huang, Xiaoyun Zheng, Weijie Jiang, Jiachen Bo, Yucheng Yang","doi":"10.1007/s11707-022-1079-x","DOIUrl":null,"url":null,"abstract":"<p>Remote sensing image scene classification and remote sensing technology applications are hot research topics. Although CNN-based models have reached high average accuracy, some classes are still misclassified, such as “freeway,” “spare residential,” and “commercial_area.” These classes contain typical decisive features, spatial-relation features, and mixed decisive and spatial-relation features, which limit high-quality image scene classification. To address this issue, this paper proposes a Grad-CAM and capsule network hybrid method for image scene classification. The Grad-CAM and capsule network structures have the potential to recognize decisive features and spatial-relation features, respectively. By using a pre-trained model, hybrid structure, and structure adjustment, the proposed model can recognize both decisive and spatial-relation features. A group of experiments is designed on three popular data sets with increasing classification difficulties. In the most advanced experiment, 92.67% average accuracy is achieved. Specifically, 83%, 75%, and 86% accuracies are obtained in the classes of “church,” “palace,” and “commercial_area,” respectively. This research demonstrates that the hybrid structure can effectively improve performance by considering both decisive and spatial-relation features. Therefore, Grad-CAM-CapsNet is a promising and powerful structure for image scene classification.</p>","PeriodicalId":48927,"journal":{"name":"Frontiers of Earth Science","volume":null,"pages":null},"PeriodicalIF":1.8000,"publicationDate":"2024-07-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Frontiers of Earth Science","FirstCategoryId":"89","ListUrlMain":"https://doi.org/10.1007/s11707-022-1079-x","RegionNum":4,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"GEOSCIENCES, MULTIDISCIPLINARY","Score":null,"Total":0}
引用次数: 0

Abstract

Remote sensing image scene classification and remote sensing technology applications are hot research topics. Although CNN-based models have reached high average accuracy, some classes are still misclassified, such as “freeway,” “spare residential,” and “commercial_area.” These classes contain typical decisive features, spatial-relation features, and mixed decisive and spatial-relation features, which limit high-quality image scene classification. To address this issue, this paper proposes a Grad-CAM and capsule network hybrid method for image scene classification. The Grad-CAM and capsule network structures have the potential to recognize decisive features and spatial-relation features, respectively. By using a pre-trained model, hybrid structure, and structure adjustment, the proposed model can recognize both decisive and spatial-relation features. A group of experiments is designed on three popular data sets with increasing classification difficulties. In the most advanced experiment, 92.67% average accuracy is achieved. Specifically, 83%, 75%, and 86% accuracies are obtained in the classes of “church,” “palace,” and “commercial_area,” respectively. This research demonstrates that the hybrid structure can effectively improve performance by considering both decisive and spatial-relation features. Therefore, Grad-CAM-CapsNet is a promising and powerful structure for image scene classification.

查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
用于遥感图像场景分类的 Grad-CAM 和胶囊网络混合方法
遥感图像场景分类和遥感技术应用是研究热点。虽然基于 CNN 的模型已经达到了较高的平均准确率,但仍有一些类被误分类,如 "高速公路"、"闲置住宅 "和 "商业区"。这些类别包含典型的决定性特征、空间相关特征以及混合决定性特征和空间相关特征,从而限制了高质量的图像场景分类。针对这一问题,本文提出了一种用于图像场景分类的 Grad-CAM 和胶囊网络混合方法。Grad-CAM 和胶囊网络结构分别具有识别决定性特征和空间相关特征的潜力。通过使用预训练模型、混合结构和结构调整,所提出的模型可以识别决定性特征和空间相关特征。我们在三个分类难度不断增加的流行数据集上设计了一组实验。在最先进的实验中,平均准确率达到了 92.67%。具体来说,"教堂"、"宫殿 "和 "商业区 "类别的准确率分别为 83%、75% 和 86%。这项研究表明,混合结构可以通过同时考虑决定性特征和空间相关特征来有效提高性能。因此,Grad-CAM-CapsNet 是一种用于图像场景分类的前景广阔且功能强大的结构。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 去求助
来源期刊
Frontiers of Earth Science
Frontiers of Earth Science GEOSCIENCES, MULTIDISCIPLINARY-
CiteScore
3.50
自引率
5.00%
发文量
627
期刊介绍: Frontiers of Earth Science publishes original, peer-reviewed, theoretical and experimental frontier research papers as well as significant review articles of more general interest to earth scientists. The journal features articles dealing with observations, patterns, processes, and modeling of both innerspheres (including deep crust, mantle, and core) and outerspheres (including atmosphere, hydrosphere, and biosphere) of the earth. Its aim is to promote communication and share knowledge among the international earth science communities
期刊最新文献
Case studies of hailstorms in Shandong Province using hail size discrimination algorithm based on dual Polarimetric parameters Experimental analysis of dust deposition and physical weathering intensity in the Gobi Desert Sedimentary architecture of a sandy braided river with seasonal hydrodynamic variations: insights from the Permian Lower Shihezi Formation, Ordos Basin, China Projected changes of runoff in the Upper Yellow River Basin under shared socioeconomic pathways Applying 3D geological modeling to predict favorable areas for coalbed methane accumulation: a case study in the Qinshui Basin
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1