Vehicle Color Recognition via Representative Color Region Extraction and Convolutional Neural Network

  • Kwang Ju Kim
  • , Pyong Kun Kim
  • , Kil Taek Lim
  • , Yun Su Chung
  • , Yoon Jeong Song
  • , Soo In Lee
  • , Doo Hyun Choi

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

10 Scopus citations

Abstract

Vehicle color recognition is one of the important part in ITS (Intelligent Transportation System). This paper presents a new vehicle color classification technique for CCTV systems via representative color region extraction and Convolutional Neural Net (CNN). The Harris corner point detection method is used to generate a probability map of a representative color region. From the probability map, point are randomly selected to generate an input image for CNN. Finally, we trained CNN model with it. In order to evaluate the performance of the proposed method, we acquired a total of 5,941 images from camera on highway. We conducted 5-fold cross validation for performance evaluation. Our vehicle color recognition method performance of about 96.1 % was shown.

Original languageEnglish
Title of host publicationICUFN 2018 - 10th International Conference on Ubiquitous and Future Networks
PublisherIEEE Computer Society
Pages89-94
Number of pages6
ISBN (Print)9781538646465
DOIs
StatePublished - 14 Aug 2018
Event10th International Conference on Ubiquitous and Future Networks, ICUFN 2018 - Prague, Czech Republic
Duration: 3 Jul 20186 Jul 2018

Publication series

NameInternational Conference on Ubiquitous and Future Networks, ICUFN
Volume2018-July
ISSN (Print)2165-8528
ISSN (Electronic)2165-8536

Conference

Conference10th International Conference on Ubiquitous and Future Networks, ICUFN 2018
Country/TerritoryCzech Republic
CityPrague
Period3/07/186/07/18

Keywords

  • CNN
  • Color Recognition
  • Probability Map
  • Vehicle

Fingerprint

Dive into the research topics of 'Vehicle Color Recognition via Representative Color Region Extraction and Convolutional Neural Network'. Together they form a unique fingerprint.

Cite this