Parametric Model of the Perspective Projection of a Road with Applications to Lane Keeping and 3D Road Reconstruction

An Improved Linear-Parabolic Model for Lane Following and Curve Detection

“…Guiducci [7,8] worked on road models and automatic camera calibration. He assumed that for planar high speed roads (radius of curvature typically larger than 1000 m) and for vehicle heading directions forming a small angle with the road direction (typically smaller than 5…”

Camera Calibration for Road Applications

“…1, we envisage a CCD camera mounted on a vehicle running on a highway. As shown in [21], the lane borders can be modeled as hyperbolas on the image plane of the on-board camera with one asymptote almost coincident with the horizon line (see B/(Y − Y 0 ), is the same for both borders and represents the deviation from straightness due to the horizontal curvature of the road.…”

“…The algorithm depends on the results of a previous paper [21] and those results, together with the notations used throughout, are briefly described in Section 2. The calibration formulas are derived in Section 3, while the theoretical values of the algorithm's errors are given in Section 4 and compared with the experimental data in Section 5.…”

Automatic change detection of driving environments in a vision-based driver assistance system

“…Vision systems have been used in many applications for detecting, tracking, monitoring, inspecting, and recognizing objects. For the purpose of driver assistance, vision systems have been exploited to detect, track, and recognize objects such as roads [6], [7], [17], [20], [23], lane markings [4], [6], [13], [16], [22], [33], [38], traffic signs [14], [29], road conditions (e.g., dry, wet, fog, freezing, and snow) [1], [39], and obstacles (e.g., pedestrians, vehicles, motorcycles and other intruders) [4], [5], [13], [28]. In this paper, a vision system for detecting critical changes in driving environment is presented.…”