KIT | KIT-Bibliothek | Impressum | Datenschutz

On the classifier performance for simulation based debris detection in sar imagery

Kuny, S.; Hammer, H.; Schulz, K.


Urban areas struck by disasters such as earthquakes are in need of a fast damage detection assessment. A post-event SAR image often is the first available image, most likely with no matching pre-event image to perform change detection. In previous work we have introduced a debris detection algorithm for this scenario that is trained exclusively with synthetically generated training data. A classification step is employed to separate debris from similar textures such as vegetation. In order to verify the use of a random forest classifier for this context, we conduct a performance comparison with two alternative popular classifiers, a support vector machine and a convolutional neural network. With the direct comparison revealing the random forest classifier to be best suited, the effective performance on the prospect of debris detection is investigated for the post-earthquake Christchurch scene. Results show a good separation of debris from vegetation and gravel, thus reducing the false alarm rate in the damage detection operation considerably.

Verlagsausgabe §
DOI: 10.5445/IR/1000140392
Veröffentlicht am 26.11.2021
DOI: 10.5194/isprs-archives-XLIII-B1-2021-45-2021
Cover der Publikation
Zugehörige Institution(en) am KIT Institut für Photogrammetrie und Fernerkundung (IPF)
Publikationstyp Proceedingsbeitrag
Publikationsjahr 2021
Sprache Englisch
Identifikator ISSN: 1682-1750
KITopen-ID: 1000140392
Erschienen in Volume XLIII-B1-2021, XXIV ISPRS Congress Imaging today, foreseeing tomorrow, Commission I: 2021 edition, 5–9 July 2021. Ed.: N. Paparoditis
Veranstaltung 24th ISPRS Congress (2021), Online, 05.07.2021 – 09.07.2021
Verlag International Society for Photogrammetry and Remote Sensing (ISPRS)
Seiten 45-50
Serie The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences ; 43
Schlagwörter SAR simulation, debris, damage detection, texture features, classifier performance
Nachgewiesen in Dimensions
KIT – Die Forschungsuniversität in der Helmholtz-Gemeinschaft
KITopen Landing Page