Traffic visibility is an essential reference for safe driving. Nighttime conditions add to the difficulty of estimating traffic visibility. To estimate the visibility in nighttime traffic images, we propose a Traffic Sensibility Visibility Estimation (TSVE) algorithm that combines laser transmission and image processing and needs no reference to the corresponding fog-free images and camera calibration. The information required is first obtained via the roadside equipment which collects environmental data and captures road images and then analyzed locally or remotely. The proposed analysis includes calculating the current atmospheric transmissivity with the laser atmospheric transmission theory and acquiring image features by using the cameras and the adjustable brightness target. Image analysis is performed using two image processing algorithms, namely, dark channel prior (DCP) and image brightness contrast. Finally, to improve the accuracy of visibility estimation, multiple nonlinear regression (MNLR) is performed on the various visibility indicators obtained by the two methods. Extensive on-site measurements analysis confirms the advantages of TSVE. Compared with other visibility estimation methods, such as the laser atmospheric transmission theory and image analysis method, TSVE significantly decreases the estimation errors.
Traffic visibility detection plays a vital role in intelligent transportation, autonomous driving, safe driving, etc. Convolutional neural networks (CNNs) based regression and classification algorithms have been shown competitive performance in many applications, but little attention has been paid to traffic visibility identification. In this paper, we propose a trainable end-to-end system called traffic visibility regression network (TVRNet). TVRNet takes a road image as input and outputs its visibility value. TVRNet adopts CNNs based deep architecture, uses appropriate filters to extract fog density-related features, and exploits the parallel convolution for multi-scale mapping. Later, a new type of non-linear activation function called Modified_sigmoid function is used. We synthesize labeled visibility datasets comprised of multi-scene and single-scene based on the actual road sense to train the visibility regression network. Extensive experiments and comparisons with other popular algorithms are performed to verify our method in road visibility estimation. INDEX TERMSTraffic visibility; convolutional neural networks; deep learning.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.