Traffic Light Detection at Night: Comparison of a Learning-based Detector and three Model-based Detectors

Publikation: Bidrag til bog/antologi/rapport/konference proceedingKonferenceartikel i proceedingForskningpeer review

5 Citationer (Scopus)
582 Downloads (Pure)

Resumé

Traffic light recognition (TLR) is an integral part of any in- telligent vehicle, it must function both at day and at night. However, the majority of TLR research is focused on day-time scenarios. In this paper we will focus on detection of traffic lights at night and evalu- ate the performance of three detectors based on heuristic models and one learning-based detector. Evaluation is done on night-time data from the public LISA Traffic Light Dataset. The learning-based detector out- performs the model-based detectors in both precision and recall. The learning-based detector achieves an average AUC of 51.4 % for the two night test sequences. The heuristic model-based detectors achieves AUCs ranging from 13.5 % to 15.0 %.
OriginalsprogEngelsk
TitelAdvances in Visual Computing : 11th International Symposium, ISVC 2015, Las Vegas, NV, USA, December 14-16, 2015, Proceedings, Part I
ForlagSpringer
Publikationsdatookt. 2015
Sider774-783
ISBN (Trykt)978-3-319-27856-8
ISBN (Elektronisk)978-3-319-27857-5
DOI
StatusUdgivet - okt. 2015
BegivenhedISVC 2015: 11th International Symposium on Visual Computing - Las Vegas, USA
Varighed: 14 dec. 201516 dec. 2015

Konference

KonferenceISVC 2015
LandUSA
ByLas Vegas
Periode14/12/201516/12/2015
NavnLecture Notes in Computer Science
Vol/bind9474
ISSN0302-9743

Fingerprint

Telecommunication traffic
Detectors
Intelligent vehicle highway systems

Citer dette

Jensen, M. B., Philipsen, M. P., Bahnsen, C., Møgelmose, A., Moeslund, T. B., & Trivedi, M. M. (2015). Traffic Light Detection at Night: Comparison of a Learning-based Detector and three Model-based Detectors. I Advances in Visual Computing: 11th International Symposium, ISVC 2015, Las Vegas, NV, USA, December 14-16, 2015, Proceedings, Part I (s. 774-783). Springer. Lecture Notes in Computer Science, Bind. 9474 https://doi.org/10.1007/978-3-319-27857-5_69
Jensen, Morten Bornø ; Philipsen, Mark Philip ; Bahnsen, Chris ; Møgelmose, Andreas ; Moeslund, Thomas B. ; Trivedi, Mohan M. / Traffic Light Detection at Night : Comparison of a Learning-based Detector and three Model-based Detectors. Advances in Visual Computing: 11th International Symposium, ISVC 2015, Las Vegas, NV, USA, December 14-16, 2015, Proceedings, Part I. Springer, 2015. s. 774-783 (Lecture Notes in Computer Science, Bind 9474).
@inproceedings{9db6a35f8efb4dc59fd953fdd25cdfa7,
title = "Traffic Light Detection at Night: Comparison of a Learning-based Detector and three Model-based Detectors",
abstract = "Traffic light recognition (TLR) is an integral part of any in- telligent vehicle, it must function both at day and at night. However, the majority of TLR research is focused on day-time scenarios. In this paper we will focus on detection of traffic lights at night and evalu- ate the performance of three detectors based on heuristic models and one learning-based detector. Evaluation is done on night-time data from the public LISA Traffic Light Dataset. The learning-based detector out- performs the model-based detectors in both precision and recall. The learning-based detector achieves an average AUC of 51.4 {\%} for the two night test sequences. The heuristic model-based detectors achieves AUCs ranging from 13.5 {\%} to 15.0 {\%}.",
author = "Jensen, {Morten Born{\o}} and Philipsen, {Mark Philip} and Chris Bahnsen and Andreas M{\o}gelmose and Moeslund, {Thomas B.} and Trivedi, {Mohan M.}",
year = "2015",
month = "10",
doi = "10.1007/978-3-319-27857-5_69",
language = "English",
isbn = "978-3-319-27856-8",
series = "Lecture Notes in Computer Science",
publisher = "Springer",
pages = "774--783",
booktitle = "Advances in Visual Computing",
address = "Germany",

}

Jensen, MB, Philipsen, MP, Bahnsen, C, Møgelmose, A, Moeslund, TB & Trivedi, MM 2015, Traffic Light Detection at Night: Comparison of a Learning-based Detector and three Model-based Detectors. i Advances in Visual Computing: 11th International Symposium, ISVC 2015, Las Vegas, NV, USA, December 14-16, 2015, Proceedings, Part I. Springer, Lecture Notes in Computer Science, bind 9474, s. 774-783, ISVC 2015, Las Vegas, USA, 14/12/2015. https://doi.org/10.1007/978-3-319-27857-5_69

Traffic Light Detection at Night : Comparison of a Learning-based Detector and three Model-based Detectors. / Jensen, Morten Bornø; Philipsen, Mark Philip; Bahnsen, Chris; Møgelmose, Andreas; Moeslund, Thomas B.; Trivedi, Mohan M.

Advances in Visual Computing: 11th International Symposium, ISVC 2015, Las Vegas, NV, USA, December 14-16, 2015, Proceedings, Part I. Springer, 2015. s. 774-783 (Lecture Notes in Computer Science, Bind 9474).

Publikation: Bidrag til bog/antologi/rapport/konference proceedingKonferenceartikel i proceedingForskningpeer review

TY - GEN

T1 - Traffic Light Detection at Night

T2 - Comparison of a Learning-based Detector and three Model-based Detectors

AU - Jensen, Morten Bornø

AU - Philipsen, Mark Philip

AU - Bahnsen, Chris

AU - Møgelmose, Andreas

AU - Moeslund, Thomas B.

AU - Trivedi, Mohan M.

PY - 2015/10

Y1 - 2015/10

N2 - Traffic light recognition (TLR) is an integral part of any in- telligent vehicle, it must function both at day and at night. However, the majority of TLR research is focused on day-time scenarios. In this paper we will focus on detection of traffic lights at night and evalu- ate the performance of three detectors based on heuristic models and one learning-based detector. Evaluation is done on night-time data from the public LISA Traffic Light Dataset. The learning-based detector out- performs the model-based detectors in both precision and recall. The learning-based detector achieves an average AUC of 51.4 % for the two night test sequences. The heuristic model-based detectors achieves AUCs ranging from 13.5 % to 15.0 %.

AB - Traffic light recognition (TLR) is an integral part of any in- telligent vehicle, it must function both at day and at night. However, the majority of TLR research is focused on day-time scenarios. In this paper we will focus on detection of traffic lights at night and evalu- ate the performance of three detectors based on heuristic models and one learning-based detector. Evaluation is done on night-time data from the public LISA Traffic Light Dataset. The learning-based detector out- performs the model-based detectors in both precision and recall. The learning-based detector achieves an average AUC of 51.4 % for the two night test sequences. The heuristic model-based detectors achieves AUCs ranging from 13.5 % to 15.0 %.

UR - http://isvc.net/Final_Program.pdf

U2 - 10.1007/978-3-319-27857-5_69

DO - 10.1007/978-3-319-27857-5_69

M3 - Article in proceeding

SN - 978-3-319-27856-8

T3 - Lecture Notes in Computer Science

SP - 774

EP - 783

BT - Advances in Visual Computing

PB - Springer

ER -

Jensen MB, Philipsen MP, Bahnsen C, Møgelmose A, Moeslund TB, Trivedi MM. Traffic Light Detection at Night: Comparison of a Learning-based Detector and three Model-based Detectors. I Advances in Visual Computing: 11th International Symposium, ISVC 2015, Las Vegas, NV, USA, December 14-16, 2015, Proceedings, Part I. Springer. 2015. s. 774-783. (Lecture Notes in Computer Science, Bind 9474). https://doi.org/10.1007/978-3-319-27857-5_69