Detection of Road Images Containing a Counterlight Using Multilevel Analysis

In this paper, a method for detecting real-time images that include counterlight produced by the sun, is proposed. It involves applying a multistep analysis of the size, location, and distribution of bright areas in the image. In general, images containing counterlight have a symmetrically high brig...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autor principal: JongBae Kim
Formato: article
Lenguaje:EN
Publicado: MDPI AG 2021
Materias:
ITS
Acceso en línea:https://doaj.org/article/f4d47b98e95a417d91aee7c86fc5ac51
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
id oai:doaj.org-article:f4d47b98e95a417d91aee7c86fc5ac51
record_format dspace
spelling oai:doaj.org-article:f4d47b98e95a417d91aee7c86fc5ac512021-11-25T19:07:37ZDetection of Road Images Containing a Counterlight Using Multilevel Analysis10.3390/sym131122102073-8994https://doaj.org/article/f4d47b98e95a417d91aee7c86fc5ac512021-11-01T00:00:00Zhttps://www.mdpi.com/2073-8994/13/11/2210https://doaj.org/toc/2073-8994In this paper, a method for detecting real-time images that include counterlight produced by the sun, is proposed. It involves applying a multistep analysis of the size, location, and distribution of bright areas in the image. In general, images containing counterlight have a symmetrically high brightness value at a specific location spread over an extremely large region. In addition, the distribution and change in brightness in that specific region have a symmetrically large difference compared with other regions. Through a multistep analysis of these symmetrical features, it is determined whether counterlight is included in the image. The proposed method presents a processing time of approximately 0.7 s and a detection accuracy of 88%, suggesting that the approach can be applied to a safe driving support system for autonomous vehicles.JongBae KimMDPI AGarticlecounterlight detectionmultilevel analysisITSsafe driving support systemsMathematicsQA1-939ENSymmetry, Vol 13, Iss 2210, p 2210 (2021)
institution DOAJ
collection DOAJ
language EN
topic counterlight detection
multilevel analysis
ITS
safe driving support systems
Mathematics
QA1-939
spellingShingle counterlight detection
multilevel analysis
ITS
safe driving support systems
Mathematics
QA1-939
JongBae Kim
Detection of Road Images Containing a Counterlight Using Multilevel Analysis
description In this paper, a method for detecting real-time images that include counterlight produced by the sun, is proposed. It involves applying a multistep analysis of the size, location, and distribution of bright areas in the image. In general, images containing counterlight have a symmetrically high brightness value at a specific location spread over an extremely large region. In addition, the distribution and change in brightness in that specific region have a symmetrically large difference compared with other regions. Through a multistep analysis of these symmetrical features, it is determined whether counterlight is included in the image. The proposed method presents a processing time of approximately 0.7 s and a detection accuracy of 88%, suggesting that the approach can be applied to a safe driving support system for autonomous vehicles.
format article
author JongBae Kim
author_facet JongBae Kim
author_sort JongBae Kim
title Detection of Road Images Containing a Counterlight Using Multilevel Analysis
title_short Detection of Road Images Containing a Counterlight Using Multilevel Analysis
title_full Detection of Road Images Containing a Counterlight Using Multilevel Analysis
title_fullStr Detection of Road Images Containing a Counterlight Using Multilevel Analysis
title_full_unstemmed Detection of Road Images Containing a Counterlight Using Multilevel Analysis
title_sort detection of road images containing a counterlight using multilevel analysis
publisher MDPI AG
publishDate 2021
url https://doaj.org/article/f4d47b98e95a417d91aee7c86fc5ac51
work_keys_str_mv AT jongbaekim detectionofroadimagescontainingacounterlightusingmultilevelanalysis
_version_ 1718410316417597440