Classification of road conditions: From camera images and weather data

Author

Jonsson, Patrik

Author_Institution

Dept. of Inf. Technol. & Media, Mid Sweden Univ., Östersund, Sweden

fYear

2011

fDate

19-21 Sept. 2011

Firstpage

1

Lastpage

6

Abstract

It is important to correctly determine road condition as it contains essential information for improving traffic safety. Knowledge about the road condition is used by maintenance personnel as a trigger for snow removal and deicing tasks. The presence of severe road conditions is also communicated as warnings and speed reduction recommendations to road users. Previous research shows that road images and data from Road Weather information Systems (RWiS) give enough information to identify road conditions, such as dry, wet, snowy, icy and tracks. The hypothesis of the new model was that it should be possible to develop a model that could classify road conditions from existing RWiS road weather data and road images. This paper proposes a model that gives a correct classification of the road conditions dry, wet, snowy and icy at an accuracy rate of 91% to 100%.

Keywords

image classification; road safety; road traffic; traffic information systems; camera image; road condition classification; road weather information system; traffic safety; weather data; Atmospheric modeling; Data models; Feature extraction; Input variables; Roads; Snow; Road accidents; Traffic information systems; classification algorithms;

fLanguage

English

Publisher

ieee

Conference_Titel

Computational Intelligence for Measurement Systems and Applications (CIMSA), 2011 IEEE International Conference on

Conference_Location

Ottawa, ON, Canada

ISSN

2159-1547

Print_ISBN

978-1-61284-924-9

Type

conf

DOI

10.1109/CIMSA.2011.6059917

Filename

6059917