A vehicle detection using selective multi-stage features in convolutional neural networks

Won Jae Lee; Dong Sung Pae; Dong Won Kim; Myo Taeg Lim

doi:10.23919/ICCAS.2017.8204413

A vehicle detection using selective multi-stage features in convolutional neural networks

Won Jae Lee, Dong Sung Pae, Dong Won Kim, Myo Taeg Lim

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

5 Citations (Scopus)

Abstract

Vehicle detection is the most basic and important technology in advanced driver assistant system. Conventional methods do not reflect characteristic information of vehicle images, so they were vulnerable to noise. In order to improve the performance of vehicle detection, this paper proposes a vehicle detection framework using selective multi-stage features in convolutional neural networks. We design the convolutional neural network (CNN) model with 10 layers and use a visualization technique to selectively extract features from the activation feature map in CNN. Our proposed features have the characteristic information of vehicle images and are more robust to noise than traditional appearance based features. We train the Adaboost algorithm using these features to implement a vehicle detector. The result of the experiments proves that our proposed vehicle detection framework has a better performance than other frameworks.

Original language	English
Title of host publication	ICCAS 2017 - 2017 17th International Conference on Control, Automation and Systems - Proceedings
Publisher	IEEE Computer Society
Pages	1-3
Number of pages	3
ISBN (Electronic)	9788993215137
DOIs	https://doi.org/10.23919/ICCAS.2017.8204413
Publication status	Published - 2017 Dec 13
Event	17th International Conference on Control, Automation and Systems, ICCAS 2017 - Jeju, Korea, Republic of Duration: 2017 Oct 18 → 2017 Oct 21

Publication series

Name	International Conference on Control, Automation and Systems
Volume	2017-October
ISSN (Print)	1598-7833

Other

Other	17th International Conference on Control, Automation and Systems, ICCAS 2017
Country/Territory	Korea, Republic of
City	Jeju
Period	17/10/18 → 17/10/21

Bibliographical note

Funding Information:
This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (Grant No. NRF-2016R1D1A1B01016071)

Funding Information:
This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (Grant No. NRF-2016R1D1A1B01016071).

Publisher Copyright:
© 2017 Institute of Control, Robotics and Systems - ICROS.

Keywords

Adaboost
Advanced driver assistant system
CNN
Vehicle detection

ASJC Scopus subject areas

Artificial Intelligence
Computer Science Applications
Control and Systems Engineering
Electrical and Electronic Engineering

Access to Document

10.23919/ICCAS.2017.8204413

Cite this

Lee, W. J., Pae, D. S., Kim, D. W., & Lim, M. T. (2017). A vehicle detection using selective multi-stage features in convolutional neural networks. In ICCAS 2017 - 2017 17th International Conference on Control, Automation and Systems - Proceedings (pp. 1-3). (International Conference on Control, Automation and Systems; Vol. 2017-October). IEEE Computer Society. https://doi.org/10.23919/ICCAS.2017.8204413

A vehicle detection using selective multi-stage features in convolutional neural networks. / Lee, Won Jae; Pae, Dong Sung; Kim, Dong Won et al.
ICCAS 2017 - 2017 17th International Conference on Control, Automation and Systems - Proceedings. IEEE Computer Society, 2017. p. 1-3 (International Conference on Control, Automation and Systems; Vol. 2017-October).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Lee, WJ, Pae, DS, Kim, DW & Lim, MT 2017, A vehicle detection using selective multi-stage features in convolutional neural networks. in ICCAS 2017 - 2017 17th International Conference on Control, Automation and Systems - Proceedings. International Conference on Control, Automation and Systems, vol. 2017-October, IEEE Computer Society, pp. 1-3, 17th International Conference on Control, Automation and Systems, ICCAS 2017, Jeju, Korea, Republic of, 17/10/18. https://doi.org/10.23919/ICCAS.2017.8204413

@inproceedings{ea4e0c52482143d88793a88f85dc5d0f,

title = "A vehicle detection using selective multi-stage features in convolutional neural networks",

abstract = "Vehicle detection is the most basic and important technology in advanced driver assistant system. Conventional methods do not reflect characteristic information of vehicle images, so they were vulnerable to noise. In order to improve the performance of vehicle detection, this paper proposes a vehicle detection framework using selective multi-stage features in convolutional neural networks. We design the convolutional neural network (CNN) model with 10 layers and use a visualization technique to selectively extract features from the activation feature map in CNN. Our proposed features have the characteristic information of vehicle images and are more robust to noise than traditional appearance based features. We train the Adaboost algorithm using these features to implement a vehicle detector. The result of the experiments proves that our proposed vehicle detection framework has a better performance than other frameworks.",

keywords = "Adaboost, Advanced driver assistant system, CNN, Vehicle detection",

author = "Lee, {Won Jae} and Pae, {Dong Sung} and Kim, {Dong Won} and Lim, {Myo Taeg}",

note = "Funding Information: This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (Grant No. NRF-2016R1D1A1B01016071) Funding Information: This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (Grant No. NRF-2016R1D1A1B01016071). Publisher Copyright: {\textcopyright} 2017 Institute of Control, Robotics and Systems - ICROS.; 17th International Conference on Control, Automation and Systems, ICCAS 2017 ; Conference date: 18-10-2017 Through 21-10-2017",

year = "2017",

month = dec,

day = "13",

doi = "10.23919/ICCAS.2017.8204413",

language = "English",

series = "International Conference on Control, Automation and Systems",

publisher = "IEEE Computer Society",

pages = "1--3",

booktitle = "ICCAS 2017 - 2017 17th International Conference on Control, Automation and Systems - Proceedings",

}

TY - GEN

T1 - A vehicle detection using selective multi-stage features in convolutional neural networks

AU - Lee, Won Jae

AU - Pae, Dong Sung

AU - Kim, Dong Won

AU - Lim, Myo Taeg

N1 - Funding Information: This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (Grant No. NRF-2016R1D1A1B01016071) Funding Information: This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (Grant No. NRF-2016R1D1A1B01016071). Publisher Copyright: © 2017 Institute of Control, Robotics and Systems - ICROS.

PY - 2017/12/13

Y1 - 2017/12/13

N2 - Vehicle detection is the most basic and important technology in advanced driver assistant system. Conventional methods do not reflect characteristic information of vehicle images, so they were vulnerable to noise. In order to improve the performance of vehicle detection, this paper proposes a vehicle detection framework using selective multi-stage features in convolutional neural networks. We design the convolutional neural network (CNN) model with 10 layers and use a visualization technique to selectively extract features from the activation feature map in CNN. Our proposed features have the characteristic information of vehicle images and are more robust to noise than traditional appearance based features. We train the Adaboost algorithm using these features to implement a vehicle detector. The result of the experiments proves that our proposed vehicle detection framework has a better performance than other frameworks.

AB - Vehicle detection is the most basic and important technology in advanced driver assistant system. Conventional methods do not reflect characteristic information of vehicle images, so they were vulnerable to noise. In order to improve the performance of vehicle detection, this paper proposes a vehicle detection framework using selective multi-stage features in convolutional neural networks. We design the convolutional neural network (CNN) model with 10 layers and use a visualization technique to selectively extract features from the activation feature map in CNN. Our proposed features have the characteristic information of vehicle images and are more robust to noise than traditional appearance based features. We train the Adaboost algorithm using these features to implement a vehicle detector. The result of the experiments proves that our proposed vehicle detection framework has a better performance than other frameworks.

KW - Adaboost

KW - Advanced driver assistant system

KW - CNN

KW - Vehicle detection

UR - http://www.scopus.com/inward/record.url?scp=85044447229&partnerID=8YFLogxK

U2 - 10.23919/ICCAS.2017.8204413

DO - 10.23919/ICCAS.2017.8204413

M3 - Conference contribution

AN - SCOPUS:85044447229

T3 - International Conference on Control, Automation and Systems

SP - 1

EP - 3

BT - ICCAS 2017 - 2017 17th International Conference on Control, Automation and Systems - Proceedings

PB - IEEE Computer Society

T2 - 17th International Conference on Control, Automation and Systems, ICCAS 2017

Y2 - 18 October 2017 through 21 October 2017

ER -

A vehicle detection using selective multi-stage features in convolutional neural networks

Abstract

Publication series

Other

Bibliographical note

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this