Deep learning radar object detection and classification for urban automotive scenarios

Rodrigo Pérez, Falk Schubert, Ralph Rasshofer, Erwin Biebl

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

27 Scopus citations

Abstract

This paper presents a single shot detection and classification system in urban automotive scenarios with a 77 GHz frequency modulated continuous wave radar sensor. The detection and classification of road users is based on the real-time object detection system YOLO (You Only Look Once) applied to the pre-processed radar range-Doppler-angle power spectrum. To train and test the proposed system, data is gathered with a test vehicle parked on urban roads. A mean average precision of 70.64% is achieved on a separate test data set.

Original languageEnglish
Title of host publication2019 Kleinheubach Conference, KHB 2019
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9783948571009
StatePublished - Sep 2019
Event2019 Kleinheubach Conference, KHB 2019 - Miltenberg, Germany
Duration: 23 Sep 201925 Sep 2019

Publication series

Name2019 Kleinheubach Conference, KHB 2019

Conference

Conference2019 Kleinheubach Conference, KHB 2019
Country/TerritoryGermany
CityMiltenberg
Period23/09/1925/09/19

Fingerprint

Dive into the research topics of 'Deep learning radar object detection and classification for urban automotive scenarios'. Together they form a unique fingerprint.

Cite this