Estimating ambient visibility in the presence of fog: a deep convolutional neural network approach

ORCID Identifiers

0000-0002-3740-1452

Document Type

Article

Source of Publication

Personal and Ubiquitous Computing

Publication Date

1-1-2019

Abstract

© 2019, Springer-Verlag London Ltd., part of Springer Nature. Next-generation intelligent transportation systems are based on the acquisition of ambient data that influence traffic flow and safety. Among these, is the ambient visibility range whose estimation, in the presence of fog, is extremely useful for next-generation intelligent transportation systems. However, existing camera-based approaches are based on “engineered features” extraction methods that use computer algorithms and procedures from the image processing field. In this contribution, a novel approach to estimate visibility range under foggy weather conditions is proposed which is based on “learned features” instead. More precisely, we use AlexNet deep convolutional neural network (DCNN), trained with raw image data, for feature extraction and a support vector machine (SVM) for visibility range estimation. Our quantitative analysis showed that the proposed approach is very promising in estimating the visibility range with very good accuracy. The proposed solution can pave the way towards intelligent driveway assistance systems to enhance awareness of driving weather conditions and hence mitigate the safety risks emanating from fog-induced low visibility conditions.

ISSN

1617-4909

Publisher

Springer London

Last Page

12

Disciplines

Computer Sciences

Keywords

Ambient intelligence, Atmospheric visibility, Deep convolutional neural networks, Intelligent transportation systems, Road safety, Ubiquitous technologies

Scopus ID

85074830838

Indexed in Scopus

yes

Open Access

no

Share

COinS