Rapid automatic vehicle manufacturer recognition using Random
forest

D 2017

Rapid automatic vehicle manufacturer recognition using Random forest

SEDLÁK, Jan a Lubomír POPELÍNSKÝ

Základní údaje

Originální název

Rapid automatic vehicle manufacturer recognition using Random forest

Autoři

SEDLÁK, Jan (203 Česká republika, garant, domácí) a Lubomír POPELÍNSKÝ (203 Česká republika, domácí)

Vydání

Bristol, Proceedings of the 21st International Database Engineering Applications Symposium, IDEAS, od s. 161-168, 8 s. 2017

Nakladatel

ACM

Další údaje

Jazyk

angličtina

Typ výsledku

Stať ve sborníku

Obor

10201 Computer sciences, information science, bioinformatics

Stát vydavatele

Spojené státy

Utajení

není předmětem státního či obchodního tajemství

Forma vydání

elektronická verze "online"

Kód RIV

RIV/00216224:14330/17:00099501

Organizační jednotka

Fakulta informatiky

ISBN

978-1-4503-5220-8

DOI

http://dx.doi.org/10.1145/3105831.3105869

Klíčová slova anglicky

machine learning; vehicle manufacturer classification; SVM; Random forest

Štítky

core_B, firank_B

Příznaky

Mezinárodní význam, Recenzováno

Změněno: 30. 9. 2018 21:21, doc. RNDr. Lubomír Popelínský, Ph.D.

Anotace

V originále

This paper studies the applicability of machine learning methods in identifying the individual vehicle ttributes based on camera images from the real environment. We focus on a vehicle manufacturer recognition. Classfication based on the front vehicle mask makes possible to identify also vehicles without manufacturer’s logo. THe algorithm has been evaluated on 2988 samples collected directly from cameras in real environment. Random forest algorithm has achieved the best results in classiffication. Accuracy for classifying the most frequent two manufacturers, ˇSkoda and Volkswagen has been 97.21% and 98.10% respectively. It is also fast enough to use it in real-time, even on low-cost devices like mobile phones or single-board computers like Raspberry Pi. Functional implementation of this method has been successfully deployed in a real-world environment.

Návaznosti

MUNI/A/0897/2016, interní kód MU

Název: Rozsáhlé výpočetní systémy: modely, aplikace a verifikace VI.

Investor: Masarykova univerzita, Rozsáhlé výpočetní systémy: modely, aplikace a verifikace VI., DO R. 2020_Kategorie A - Specifický výzkum - Studentské výzkumné projekty

Citovat

SEDLÁK, Jan a Lubomír POPELÍNSKÝ. Rapid automatic vehicle manufacturer recognition using Random forest. Online. In Bipin C. Desai and Jun Hong and Richard McClatchey. Proceedings of the 21st International Database Engineering Applications Symposium, IDEAS. Bristol: ACM, 2017, s. 161-168. ISBN 978-1-4503-5220-8. Dostupné z: https://dx.doi.org/10.1145/3105831.3105869.

@inproceedings{1404191,
   author = {Sedlák, Jan and Popelínský, Lubomír},
   address = {Bristol},
   booktitle = {Proceedings of the 21st International Database Engineering Applications Symposium, IDEAS},
   doi = {http://dx.doi.org/10.1145/3105831.3105869},
   editor = {Bipin C. Desai and Jun Hong and Richard McClatchey},
   keywords = {machine learning; vehicle manufacturer classification; SVM; Random forest},
   howpublished = {elektronická verze "online"},
   language = {eng},
   location = {Bristol},
   isbn = {978-1-4503-5220-8},
   pages = {161-168},
   publisher = {ACM},
   title = {Rapid automatic vehicle manufacturer recognition using Random forest},
   year = {2017}
}

TY  - JOUR
ID  - 1404191
AU  - Sedlák, Jan - Popelínský, Lubomír
PY  - 2017
TI  - Rapid automatic vehicle manufacturer recognition using Random forest
PB  - ACM
CY  - Bristol
SN  - 9781450352208
KW  - machine learning
KW  - vehicle manufacturer classification
KW  - SVM
KW  - Random forest
N2  - This paper studies the applicability of machine learning methods in identifying the individual vehicle ttributes based on camera images from the real environment. We focus on a vehicle manufacturer recognition. Classfication based on the front vehicle mask makes possible to identify also vehicles without manufacturer’s logo. THe algorithm has been evaluated on 2988 samples collected directly from cameras in real environment. Random forest algorithm has achieved the best results in classiffication. Accuracy for classifying the most frequent two manufacturers, ˇSkoda and Volkswagen has been 97.21% and 98.10% respectively. It is also fast enough to use it in real-time, even on low-cost devices like mobile phones or single-board computers like Raspberry Pi. Functional implementation of this method has been successfully deployed in a real-world environment.
ER  -

SEDLÁK, Jan a Lubomír POPELÍNSKÝ. Rapid automatic vehicle manufacturer recognition using Random forest. Online. In Bipin C. Desai and Jun Hong and Richard McClatchey. \textit{Proceedings of the 21st International Database Engineering Applications Symposium, IDEAS}. Bristol: ACM, 2017, s.~161-168. ISBN~978-1-4503-5220-8. Dostupné z: https://dx.doi.org/10.1145/3105831.3105869.

Podrobný výpis o publikaci