The search functionality is under construction.
The search functionality is under construction.

Vision Based Nighttime Vehicle Detection Using Adaptive Threshold and Multi-Class Classification

Yuta SAKAGAWA, Kosuke NAKAJIMA, Gosuke OHASHI

  • Full Text Views

    0

  • Cite this

Summary :

We propose a method that detects vehicles from in-vehicle monocular camera images captured during nighttime driving. Detecting vehicles from their shape is difficult at night; however, many vehicle detection methods focusing on light have been proposed. We detect bright spots by appropriate binarization based on the characteristics of vehicle lights such as brightness and color. Also, as the detected bright spots include lights other than vehicles, we need to distinguish the vehicle lights from other bright spots. Therefore, the bright spots were distinguished using Random Forest, a multiclass classification machine-learning algorithm. The features of bright spots not associated with vehicles were effectively utilized in the vehicle detection in our proposed method. More precisely vehicle detection is performed by giving weights to the results of the Random Forest based on the features of vehicle bright spots and the features of bright spots not related to the vehicle. Our proposed method was applied to nighttime images and confirmed effectiveness.

Publication
IEICE TRANSACTIONS on Fundamentals Vol.E102-A No.9 pp.1235-1245
Publication Date
2019/09/01
Publicized
Online ISSN
1745-1337
DOI
10.1587/transfun.E102.A.1235
Type of Manuscript
Special Section PAPER (Special Section on Image Media Quality)
Category

Authors

Yuta SAKAGAWA
  Shizuoka University
Kosuke NAKAJIMA
  Shizuoka University
Gosuke OHASHI
  Shizuoka University

Keyword