Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Semantic labeling for improved vehicle detection in aerial imagery

: Sommer, L.; Nie, K.; Schumann, A.; Schuchert, Tobias; Beyerer, Jürgen

Volltext urn:nbn:de:0011-n-4819034 (605 KByte PDF)
MD5 Fingerprint: 30f6758fe4a3f23cd911b4e1c10e94b9
© IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE.
Erstellt am: 8.2.2018

Institute of Electrical and Electronics Engineers -IEEE-:
14th IEEE International Conference on Advanced Video and Signal Based Surveillance, AVSS 2017 : August 29, 2017-September 1, 2017, Lecce
Piscataway, NJ: IEEE, 2017
ISBN: 978-1-5386-2939-0
ISBN: 978-1-5386-2940-6 (Print)
International Conference on Advanced Video and Signal Based Surveillance (AVSS) <14, 2017, Lecce>
Konferenzbeitrag, Elektronische Publikation
Fraunhofer IOSB ()
contextual information; conventional detection; detection framework; detection performance; false positive detection; ground sampling distance; semantic labeling; traffic monitoring

Growing cities and increasing traffic densities result in an increased demand for applications such as traffic monitoring, traffic analysis, and support of rescue work. These applications share the need for accurate detection of relevant vehicles, e.g. in aerial imagery. Recently, the application of deep learning based detection frameworks like Faster R-CNN clearly outperformed conventional detection methods for vehicle detection in aerial images. In this paper, we propose a detection framework that fuses Faster R-CNN and semantic labeling to integrate contextual information. We achieve an improved detection performance by decreasing the number of false positive detections while the number of candidate regions to classify is reduced. To demonstrate the generalization of our approach, we evaluate our detection framework for various ground sampling distances on a publicly available dataset.