TU Darmstadt / ULB / TUbiblio

A Light and Faster Regional Convolutional Neural Network for Object Detection in Optical Remote Sensing Images

Ding, Peng ; Zhang, Ye ; Deng, Wei-Jian ; Jia, Ping ; Kuijper, Arjan (2018)
A Light and Faster Regional Convolutional Neural Network for Object Detection in Optical Remote Sensing Images.
In: ISPRS Journal of Photogrammetry and Remote Sensing, 141
doi: 10.1016/j.isprsjprs.2018.05.005
Artikel, Bibliographie

Kurzbeschreibung (Abstract)

Detection of objects from satellite optical remote sensing images is very important for many commercial and governmental applications. With the development of deep convolutional neural networks (deep CNNs), the field of object detection has seen tremendous advances. Currently, objects in satellite remote sensing images can be detected using deep CNNs. In general, optical remote sensing images contain many dense and small objects, and the use of the original Faster Regional CNN framework does not yield a suitably high precision. Therefore, after careful analysis we adopt dense convoluted networks, a multi-scale representation and various combinations of improvement schemes to enhance the structure of the base VGG16-Net for improving the precision. We propose an approach to reduce the test-time (detection time) and memory requirements. To validate the effectiveness of our approach, we perform experiments using satellite remote sensing image datasets of aircraft and automobiles. The results show that the improved network structure can detect objects in satellite optical remote sensing images more accurately and efficiently.

Typ des Eintrags: Artikel
Erschienen: 2018
Autor(en): Ding, Peng ; Zhang, Ye ; Deng, Wei-Jian ; Jia, Ping ; Kuijper, Arjan
Art des Eintrags: Bibliographie
Titel: A Light and Faster Regional Convolutional Neural Network for Object Detection in Optical Remote Sensing Images
Sprache: Englisch
Publikationsjahr: 2018
Titel der Zeitschrift, Zeitung oder Schriftenreihe: ISPRS Journal of Photogrammetry and Remote Sensing
Jahrgang/Volume einer Zeitschrift: 141
DOI: 10.1016/j.isprsjprs.2018.05.005
URL / URN: https://doi.org/10.1016/j.isprsjprs.2018.05.005
Kurzbeschreibung (Abstract):

Detection of objects from satellite optical remote sensing images is very important for many commercial and governmental applications. With the development of deep convolutional neural networks (deep CNNs), the field of object detection has seen tremendous advances. Currently, objects in satellite remote sensing images can be detected using deep CNNs. In general, optical remote sensing images contain many dense and small objects, and the use of the original Faster Regional CNN framework does not yield a suitably high precision. Therefore, after careful analysis we adopt dense convoluted networks, a multi-scale representation and various combinations of improvement schemes to enhance the structure of the base VGG16-Net for improving the precision. We propose an approach to reduce the test-time (detection time) and memory requirements. To validate the effectiveness of our approach, we perform experiments using satellite remote sensing image datasets of aircraft and automobiles. The results show that the improved network structure can detect objects in satellite optical remote sensing images more accurately and efficiently.

Freie Schlagworte: Convolutional Neural Networks (CNN), Deep learning, Object detection
Fachbereich(e)/-gebiet(e): 20 Fachbereich Informatik
20 Fachbereich Informatik > Mathematisches und angewandtes Visual Computing
Hinterlegungsdatum: 19 Jun 2019 11:18
Letzte Änderung: 19 Jun 2019 11:18
PPN:
Export:
Suche nach Titel in: TUfind oder in Google
Frage zum Eintrag Frage zum Eintrag

Optionen (nur für Redakteure)
Redaktionelle Details anzeigen Redaktionelle Details anzeigen