A new strategy of detecting traffic information based on traffic camera : modified inverse perspective mapping

Menéndez García, José Manuel and Yaghoobi Ershadi, Nastaran (2017). A new strategy of detecting traffic information based on traffic camera : modified inverse perspective mapping. "Journal of Electrical Engineering, Technology and Interface Utilities", v. 10 (n. 2); pp. 1101-1118. ISSN 2093-7423. https://doi.org/10.5271/JEETIU.2017.10.2.984.

Description

Title: A new strategy of detecting traffic information based on traffic camera : modified inverse perspective mapping
Author/s:
  • Menéndez García, José Manuel
  • Yaghoobi Ershadi, Nastaran
Item Type: Article
Título de Revista/Publicación: Journal of Electrical Engineering, Technology and Interface Utilities
Date: March 2017
Volume: 10
Subjects:
Freetext Keywords: Transportation Systems, Hough transform occlusions
Faculty: E.T.S.I. Telecomunicación (UPM)
Department: Señales, Sistemas y Radiocomunicaciones
Creative Commons Licenses: Recognition - No derivative works - Non commercial

Full text

[img]
Preview
PDF - Requires a PDF viewer, such as GSview, Xpdf or Adobe Acrobat Reader
Download (706kB) | Preview

Abstract

The development of Intelligent Transportation Systems (ITS) needs high quality traffic information such as intersections, but conventional image-based traffic detection methods have difficulties with perspective and background noise, shadows and lighting transitions. In this paper, we propose a new traffic information detection method based on Modified Inverse Perspective Mapping (MIPM) to perform under these challenging conditions. In our proposed method, first the perspective is removed from the images using the Modified Inverse Perspective Mapping (MIPM); afterward, Hough transform is applied to extract structural information like road lines and lanes; then, Gaussian Mixture Models are used to generate the binary image. Meanwhile, to tackle shadow effect in car areas, we have applied a chromacity-base strategy. To evaluate the performance of the proposed method, we used several video sequences as benchmarks. These videos are captured in normal weather from a high way, and contain different types of locations and occlusions between cars. Our simulation results indicate that the proposed algorithms and frameworks are effective, robust and more accurate compared to other frameworks, especially in facing different kinds of occlusions.

More information

Item ID: 49940
DC Identifier: http://oa.upm.es/49940/
OAI Identifier: oai:oa.upm.es:49940
DOI: 10.5271/JEETIU.2017.10.2.984
Official URL: http://jeet.us/firstonline.html
Deposited by: Memoria Investigacion
Deposited on: 14 Apr 2018 08:50
Last Modified: 25 Mar 2019 09:32
  • Logo InvestigaM (UPM)
  • Logo GEOUP4
  • Logo Open Access
  • Open Access
  • Logo Sherpa/Romeo
    Check whether the anglo-saxon journal in which you have published an article allows you to also publish it under open access.
  • Logo Dulcinea
    Check whether the spanish journal in which you have published an article allows you to also publish it under open access.
  • Logo de Recolecta
  • Logo del Observatorio I+D+i UPM
  • Logo de OpenCourseWare UPM