A subclass supported convolutional neural network for object detection and localization in remote-sensing images

This article proposes a novel subclass-based classifier based on convolutional neural networks (CNNs) for detecting objects more accurately on remote-sensing images. The proposed classifier, called subclass supported CNN (SSCNN), is used to separate the representation of the objects into subclasses such as nearcentre, centre, and border depending on the distance of the object centre to obtain more effective feature extractor. A three-stage object recognition framework is used to evaluate the performance of the proposed classifier. In the first of these stages, the Selective Search algorithm generates object proposals from the image. Then, the proposed SSCNN classifies the proposals. Finally, subclass-based localization evaluation function has been proposed to calculate the localization of the object with classification results. Due to the limited number of satellite image samples, pretrained AlexNet is used by transfer learning approach to build effective feature extractor. The proposed method has been compared with region-based CNN (R-CNN) on a four-class remote-sensing test dataset consisting of 411 airplanes, 240 baseball diamonds, 468 storage tanks, and 83 ground track fields. In addition, Faster R-CNN has been trained with SSCNN features and the performances of the trained Faster R-CNNs are comparatively evaluated on 10-class remote-sensing image dataset. Experiment results have shown that the proposed framework can locate the objects precisely. © 2018, © 2018 Informa UK Limited, trading as Taylor & Francis Group.

Yazar Kilic E.
Ozturk S.
Yayın Türü Article
Tek Biçim Adres https://hdl.handle.net/20.500.12628/4080
Tek Biçim Adres 10.1080/01431161.2018.1562260
Koleksiyonlar Araştırma Çıktıları | WoS | Scopus | TR-Dizin | PubMed | SOBİAD
Scopus İndeksli Yayınlar Koleksiyonu
WoS İndeksli Yayınlar Koleksiyonu
Dergi Adı International Journal of Remote Sensing
Dergi Cilt Bilgisi 40
Dergi Sayısı 11
Sayfalar 4193 - 4212
Yayın Yılı 2019
Eser Adı
[dc.title]
A subclass supported convolutional neural network for object detection and localization in remote-sensing images
Yazar
[dc.contributor.author]
Kilic E.
Yazar
[dc.contributor.author]
Ozturk S.
Yayın Yılı
[dc.date.issued]
2019
Yayıncı
[dc.publisher]
Taylor and Francis Ltd.
Yayın Türü
[dc.type]
article
Özet
[dc.description.abstract]
This article proposes a novel subclass-based classifier based on convolutional neural networks (CNNs) for detecting objects more accurately on remote-sensing images. The proposed classifier, called subclass supported CNN (SSCNN), is used to separate the representation of the objects into subclasses such as nearcentre, centre, and border depending on the distance of the object centre to obtain more effective feature extractor. A three-stage object recognition framework is used to evaluate the performance of the proposed classifier. In the first of these stages, the Selective Search algorithm generates object proposals from the image. Then, the proposed SSCNN classifies the proposals. Finally, subclass-based localization evaluation function has been proposed to calculate the localization of the object with classification results. Due to the limited number of satellite image samples, pretrained AlexNet is used by transfer learning approach to build effective feature extractor. The proposed method has been compared with region-based CNN (R-CNN) on a four-class remote-sensing test dataset consisting of 411 airplanes, 240 baseball diamonds, 468 storage tanks, and 83 ground track fields. In addition, Faster R-CNN has been trained with SSCNN features and the performances of the trained Faster R-CNNs are comparatively evaluated on 10-class remote-sensing image dataset. Experiment results have shown that the proposed framework can locate the objects precisely. © 2018, © 2018 Informa UK Limited, trading as Taylor & Francis Group.
Kayıt Giriş Tarihi
[dc.date.accessioned]
2019-12-23
Açık Erişim Tarihi
[dc.date.available]
2019-12-23
Yayın Dili
[dc.language.iso]
eng
Haklar
[dc.rights]
info:eu-repo/semantics/closedAccess
ISSN
[dc.identifier.issn]
0143-1161
İlk Sayfa Sayısı
[dc.identifier.startpage]
4193
Son Sayfa Sayısı
[dc.identifier.endpage]
4212
Dergi Adı
[dc.relation.journal]
International Journal of Remote Sensing
Dergi Sayısı
[dc.identifier.issue]
11
Dergi Cilt Bilgisi
[dc.identifier.volume]
40
Tek Biçim Adres
[dc.identifier.uri]
https://dx.doi.org/10.1080/01431161.2018.1562260
Tek Biçim Adres
[dc.identifier.uri]
https://hdl.handle.net/20.500.12628/4080
Görüntülenme Sayısı ( Şehir )
Görüntülenme Sayısı ( Ülke )
Görüntülenme Sayısı ( Zaman Dağılımı )
Görüntülenme
74
09.12.2022 tarihinden bu yana
İndirme
1
09.12.2022 tarihinden bu yana
Son Erişim Tarihi
11 Haziran 2024 03:55
Google Kontrol
Tıklayınız
proposed object remote-sensing objects classifier proposals trained effective feature extractor Faster framework localization centre results dataset subclass-based method storage approach diamonds compared region-based baseball (R-CNN) airplanes four-class consisting performances ground Francis Taylor trading Limited Informa
6698 sayılı Kişisel Verilerin Korunması Kanunu kapsamında yükümlülüklerimiz ve çerez politikamız hakkında bilgi sahibi olmak için alttaki bağlantıyı kullanabilirsiniz.

creativecommons
Bu site altında yer alan tüm kaynaklar Creative Commons Alıntı-GayriTicari-Türetilemez 4.0 Uluslararası Lisansı ile lisanslanmıştır.
Platforms