Open AccessFast vanishing point detection method based on road border region estimation
Author(s) -
Wang Ende,
Sun Anan,
Li Yong,
Hou Xukui,
Zhu Yalong
Publication year - 2018
Publication title -
iet image processing
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.401
H-Index - 45
eISSN - 1751-9667
pISSN - 1751-9659
DOI - 10.1049/iet-ipr.2017.0030
Subject(s) - vanishing point , shadow (psychology) , point (geometry) , computer science , boundary (topology) , computer vision , line (geometry) , estimation , voting , artificial intelligence , road surface , line segment , image (mathematics) , algorithm , mathematics , geometry , engineering , mathematical analysis , civil engineering , systems engineering , politics , law , political science , psychotherapist , psychology
The vanishing point is the converging point at which the road boundaries extend to the distance. It can be used to estimate the road region in a complicated environment. As the vanishing point detection algorithms for the unstructured road often take long time and they could not adapt to the image with complex environment of the road, in this study, the authors propose a vanishing point estimation method based on road boundaries region estimation. This method uses the boundary direction of the voting points in the border region to estimate the vanishing point by the line‐soft‐voting based on maximum weight. This method has fast calculation speed, and it can overcome the influence of the shadow and the complex environment on the surface of the road in many cases.