Abstract-Based on a road meteorology standard, we present a roadside camera-based system able to detect daytime fog and to estimate the visibility range. Two detection algorithms, both based on a daytime fog model, are presented along with a process to combine their outputs. Unlike previous methods, the system takes into account the 3-D scene structure and filters the moving objects from the region of interest through use of a background modelling approach and detects the cause of the visibility reduction. The study of the system accuracy with respect to the camera characteristics leads to a specification of the characteristics of the camera required for the system. Some results obtained using a reduced-scale prototyping of the system are presented. Finally, an outlook to future works is given.

No abstract

Abstract-Long-range detection of road surface in a sequence of images from a front camera aboard a vehicle is known as an unsolved problem. We propose an algorithm using a single camera and based on color segmentation which has interesting performance and which is stable along the sequence whatever its length. It is an off-line algorithm which makes good use of current and successive images to build reliable models of the color aspects of the road and of its environment at each vehicle position. To apply the proposed algorithm a radiometric calibration step is required to ensure uniform responses of the pixels over the image. The algorithm consists in three steps: image smoothing consistent with perspective effect on the road, building of the models of the road and non-road colors, and region growing of the road region. The relevance of the proposed algorithm is illustrated by an application to the roadway visibility estimation in stereovision and its performance are illustrated by experiments in difficult situations.

We present a new application of computer vision: continuous measurement of the geometrical visibility range on inter-urban roads, solely based on a monocular image acquisition system. To tackle this problem, we propose first a road segmentation scheme based on a Parzenwindowing of a color feature space with an original update that allows us to cope with heterogeneously paved-roads, shadows and reflections, observed under various and changing lighting conditions. Second, we address the under-constrained problem of retrieving the depth information along the road based on the flat word assumption. This is performed by a new region-fitting iterative least squares algorithm, derived from half-quadratic theory, able to cope with vanishing-point estimation, and allowing us to estimate the geometrical visibility range.⋆ Thanks to the French Department of Transportation for funding, within the SARI-PREDIT project (http://www.sari.prd.fr/HomeEN.html).