Compared with daytime, a larger proportion of road accidents happens during nighttime. The altered visibility for drivers partially explains this situation. It becomes worse when dense fog is present. In this paper, we first define a standard night visibility index, which allows specifying the type of fog that an advanced driver assistance system should recognize. A methodology to detect the presence of night fog and characterize its density in images grabbed by an in-vehicle camera is then proposed. The detection method relies on the visual effects of night fog. A first approach evaluates the presence of fog around a vehicle due to the detection of the backscattered veil created by the headlamps. In this aim, a correlation index is computed between the current image and a reference image where the fog density is known. It works when the vehicle is alone on a highway without external light sources. A second approach evaluates the presence of fog due to the detection of halos around light sources ahead of the vehicle. It works with oncoming traffic and public lighting. Both approaches are illustrated with actual images of fog. Their complementarity makes it possible to envision a complete night-fog detection system. If fog is detected, its characterization is achieved by fitting the different correlation indexes with an empirical model. Experimental results show the efficiency of the proposed method. The main applications for such a system are, for instance, automation or adaptation of vehicle lights, contextual speed computation, and reliability improvement for camera-based systems.

The employment of embedded passive sensors in order to perceive environment for reducing the accident risk level is a tendency of intelligent vehicles research. From such sensors, one can extract useful informations which can assist the driver to identify hazardous situations. While safety improvement is a substantial requirement for driving assistance, localizing and tracking obstacles in complex road environment became an important task. One promising approach is to use the V-disparity based on the stereovision technique. It is a cumulative space estimated from the disparity image. We propose a sound framework and a complete system based on a real-time stereovision for detection, 3D localization and tracking of dynamic obstacles in highway environment. The main contribution we propose is the improvement of the V-disparity approach by extending the basic approach by merging it with a confidence term. This consists on weighting each pixel in the V-disparity space according to a confidence value which measures the probability of associating a pair of pixels. Furthermore, we propose a tracking system which is based on the belief theory. The tracking task is done on the image space which takes into account uncertainties, handles conflicts, and automatically dealt with targets appearance and disappearce as well as their spatial and temporal propogation. Extensive experiments on simulated and real dataset demonstrate the effectiveness and the robustness of the weighted V-disparity approach.

Advanced Driver Assistance Systems (ADAS) based on video camera tends to be generalized in today's automotive. However, if most of these systems perform nicely in good weather conditions, they perform very poorly under adverse weather particularly under rain. We present a novel approach that aims at detecting raindrops on a car windshield using only images from an in-vehicle camera. Based on the photometric properties of raindrops, the algorithm relies on image processing technics to highlight raindrops. Its results can be further used for image restoration and vision enhancement and hence it is a valuable tool for ADAS.

Compared to daytime, a larger proportion of accident happens during nighttime. The altered visibility of the road scene for the drivers may partially explain this situation. The latter becomes worse in fog presence. In this paper, two camera-based methods are proposed to detect the presence of night fog in images grabbed by in-vehicle multipurpose cameras. They rely on the visual effects of night fog. A first approach can assess the presence of fog around the vehicle thanks to the detection of the backscattered veil induced by the vehicle ego lights. It aims at detecting fog when the vehicle is alone in absence of exterior public lighting. A second approach can assess the presence of fog thanks to the detection of halos around light sources in the vehicle environment. It aims at detecting fog in presence of road traffic or public lighting. Both methods are presented and illustrated with actual images of fog. Their complementarity makes it possible to envision a complete night fog detection system. There are numerous applications for such a system: automation or adaptation of vehicle lights, contextual speed computation and reliability improvement for camera-based systems.

Vision-based autonomous vehicles must face numerous challenges in order to be effective in practical areas. Among these lies the detection and localization of independent-moving objects, so as to track or avoid them. In this paper a method that address this particular issue is presented. Information from stereo and motion is used to extract the ego-motion of the vehicle. Known defects of this estimation are exploited to detect independent-moving obstacles. This method allows an early and reliable detection, even for objects partially occluded. Besides, it highlights the errors in the disparity map, which can be used, in future works, to correct depth-estimation, through motion-estimation.

Stereo-vision is extensively used for intelligent vehicles, mainly for obstacle detection, as it provides a large amount of data. Many authors use it as a classical 3D sensor which provides a large tri-dimensional cloud of metric measurements, and apply methods usually designed for other sensors, such as clustering based on a distance. For stereo-vision, the measurement uncertainty is related to the range. For medium to long range, often necessary in the field of intelligent vehicles, this uncertainty has a significant impact, limiting the use of this kind of approaches. On the other hand, some authors consider stereo-vision more like a vision sensor and choose to directly work in the disparity space. This provides the ability to exploit the connectivity of the measurements, but roughly takes into consideration the actual size of the objects. In this paper, we propose a probabilistic representation of the specific uncertainty for stereo-vision, which takes advantage of both aspects - distance and disparity. The model is presented and then applied to obstacle detection, using the occupancy grid framework. For this purpose, a computationally-efficient implementation based on the u-disparity approach is given.

In adverse weather conditions, in particular, in daylight fog, the contrast of images grabbed by in-vehicle cameras in the visible light range is drastically degraded, which makes current driver assistance that relies on cameras very sensitive to weather conditions. An onboard vision system should take weather effects into account. The effects of daylight fog vary across the scene and are exponential with respect to the depth of scene points. Because it is not possible in this context to compute the road scene structure beforehand, contrary to fixed camera surveillance, a new scheme is proposed. Fog density is first estimated and then used to restore the contrast using a flat-world assumption on the segmented free space in front of a moving vehicle. A scene structure is estimated and used to refine the restoration process. Results are presented using sample road scenes under foggy weather and assessed by computing the visibility level enhancement that is gained by the method. Finally, we show applications to the enhancement in daylight fog of low-level algorithms that are used in advanced camera-based driver assistance.

This paper presents a new method for the detection and 3D reconstruction of the road lane using onboard stereovision. The proposed algorithm makes it possible to overcome the assumptions commonly used in most of the detection systems using monocular vision such as: flat road, constant pitch angle or absence of roll angle. The proposed method of detection and 3D reconstruction is based on two modules. The first one is aimed at detecting the road markings of the lane in each image of the stereoscopic pair. It is based on a recognition algorithm driven by a statistical model of the road lane: the initial state of the model is obtained after a learning stage; it is updated using the results of a feature extraction stage. Our model takes into account the intrinsic links between the projections of the road in the two images of the stereoscopic pair, so that its update from a feature extracted in one of two images drives the detection of the features in the other image. No disparity map is required since the matching of the road features is directly obtained as the result of the model update. The second module is aimed at 3D reconstruction and relative localization of the vehicle with respect to its lane. The parameters of a 3D surface including the horizontal and vertical profiles of the road lane are estimated from the road borders previously detected. The robustness of the algorithm is evaluated from synthetic and real images.

Driving safety enhancement could be achieved by better understanding of risk situations from the knowledge of vehicle dynamic states as well as road geometry. Among the parameters of the road that have an impact on vehicle dynamics, one can find the bank and the slope angles, which can not however be measured by mean of low cost onboard sensors. This work of this paper is aimed to improve the controllability of the vehicle and the development of a low cost mapping system of the roadway. Two observers are developed firstly to estimate these two variables and secondly to localize the vehicle. The road attributes are estimated using an extended Kalman filter (EKF) and a Proportional Integral (PI) observer with unknown inputs based vehicle models and the measurements obtained from inertial (INS) and ABS sensors. The vehicle localization is performed at each time sample using an algorithm based on the IMM technique (Interacting Multiple Models), it allows to reconstruct the road shape in 3 dimensions. Testing on measurements obtained with a prototype vehicle show the good behavior of the proposed estimation scheme.

In foggy weather, the contrast of images grabbed by in-vehicle cameras in the visible light range is drastically degraded, which makes the current applications very sensitive to weather conditions. An onboard vision system should take fog effects into account. The effects of fog varies across the scene and are exponential with respect to the depth of scene points. Because it is not possible in this context to compute the road scene structure beforehand contrary to fixed camera surveillance, a new scheme is proposed. Weather conditions are first estimated and then used to restore the contrast according to a scene structure which is inferred a priori and refined during the restoration process. Based on the aimed application, different algorithms with increasing complexities are proposed. Results are presented using sample road scenes under foggy weather and assessed by computing the contrast before and after restoration.

A stereovision method is presented in this paper, to compute reliable and quasi-dense disparity maps of road scenes using in-vehicle cameras. It combines the advantages of the "v-disparity" approach and a quasi-dense matching algorithm. In this aim, road surface and vertical planes of the scene are first extracted using the sparse "v-disparity" approach. The knowledge of these global surfaces of the scene is then used to guide a quasi-dense matching algorithm and to propagate disparity information on horizontal edges. Both algorithms are presented and compared. Then, our approach is presented and examples of quasi-dense disparity maps are given. Finally, the efficiency of the method is illustrated by the accurate positioning of a bounding box around a vehicle in a bad contrasted video sequence

An atmospheric visibility measurement system capable of quantifying the most common operating range of onboard exteroceptive sensors is a key parameter in the creation of driving assistance systems. This information is then utilized to adapt sensor operations and processing or to alert the driver that his onboard assistance system is momentarily inoperative. Moreover, a system capable of either detecting the presence of fog or estimating visibility distances constitutes in itself a driving assistance. In this paper, the authors present a technique to estimate the mobilized visibility distance through a use of onboard charge-coupled device cameras. The latter represents the distance to the most distant object on the road surface having a contrast above 5%. This definition is very close to the definition of the meteorological visibility distance proposed by the International Commission on Illumination. The method combines the computations of local contrasts above 5% and of a depth map of the vehicle environment using stereovision within 60 ms on a current-day computer. In this paper, both methods are described separately. Then, their combination is detailed. The method is operative night and day in every kind of meteorological condition and is evaluated; thanks to video sequences under sunny weather and foggy weather.