Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Utilizing temporal information in UAV surveillance videos for distant moving object recognition

: Teutsch, Michael

Volltext urn:nbn:de:0011-n-2384780 (236 KByte PDF)
MD5 Fingerprint: 85c9523ad1eceb6881593d39f26403b8
Erstellt am: 25.4.2013

Beyerer, Jürgen (Ed.); Pak, Alexey (Ed.) ; Fraunhofer-Institut für Optronik, Systemtechnik und Bildauswertung -IOSB-, Karlsruhe; Karlsruhe Institute of Technology -KIT-, Lehrstuhl für Interaktive Echtzeitsysteme -IES-:
Joint Workshop of Fraunhofer IOSB and Institute for Anthropomatics, Vision and Fusion Laboratory 2012. Proceedings : Triberg-Nussbach, Germany, July, 22 to 28, 2012
Karlsruhe: KIT Scientific Publishing, 2013 (Karlsruher Schriften zur Anthropomatik 13)
ISBN: 978-3-86644-988-6
DOI: 10.5445/KSP/1000032956
Fraunhofer Institute of Optronics, System Technologies and Image Exploitation and Institute for Anthropomatics, Vision and Fusion Laboratory (Joint Workshop) <2012, Triberg-Nussbach>
Konferenzbeitrag, Elektronische Publikation
Fraunhofer IOSB ()

Image sequences contain more information than single images due to the temporal context. There are many potential benefits for the automatic analysis of especially distant moving objects in surveillance videos such as temporal noise reduction, track-before-detect, estimating motion information of the camera itself and objects in the scene, or acquiring different appearances of an object for classification. In this report, example approaches are presented for utilizing the temporal information to make the detection, segmentation, and classification of such objects more robust. Using real surveillance datasets, various algorithms for independent motion detection and moving object segmentation are presented and evaluated. Some ideas for considering temporal information for object classification are discussed in a conceptual manner.