Real Time Stereo Image Registration for Planar Structure and 3D Sensor Pose Estimation

Abstract

In recent years, several techniques to on-board vision pose estimation have been proposed (Zhu et al., 1998; Labayrade & Aubert, 2003; Liu & Fujimura, 2004; Stein et al., 2000; Suzuki & Kanade, 1999). Vision system pose estimation is required for any advanced driver assistance application. The real-time estimation of on-board vision system pose-position and orientationis a challenging task since i) the sensor undergoes motions due to the vehicle dynamics and the road imperfections, and ii) the viewed scene is unknown and continuously changing. Of particular interest is the estimation of on-board camera's position and orientation related to the 3D road plane. Note that since the 3D plane parameters are expressed in the camera coordinate system, the camera's position and orientation are equivalent to the 3D plane parameters. Algorithms for fast road plane estimation are very useful for driver assistance applications as well as for augmented reality applications. The ability to use continuously updated plane parameters (camera pose) will considerably make the tasks of obstacle detection more efficient (Viola et al., 2005; Sun et al., 2006; Toulminet et al., 2006). However, dealing with an urban scenario is more diffcult than dealing with highways scenario since the prior knowledge as well as visual features are not always available in these scenes (Franke et al., 1999). In general, monocular vision systems avoid problems related to 3D Euclidean geometry by using the prior knowledge of the environment as an extra source of information. Although prior knowledge has been extensively used to tackle the driver assistance problem, it may lead to wrong results. Hence, considering a constant camera's position and orientation is not a valid assumption to be used in urban scenarios, since both of them are easily affected by road imperfections or artifacts (e.g., rough road, speed bumpers), car's accelerations, uphill/downhill driving, among others. The use of prior knowledge has also been considered by some stereo vision based techniques to simplify the problem and to speed up the whole processing by reducing the amount of information to be handled (Bertozzi & Broggi, 1998; Bertozzi et al. 2003; Nedevschi et al., 2006 ). In the literature, many application-oriented stereo systems have been proposed. For instance, the edge based v-disparity approach proposed in (Labayrade et al., 2002), for an automatic estimation of horizon lines and later on used for applications such as obstacle or pedestrian detection (e.g., (Bertozzi et al., 2005; Labayrade & Aubert, O pe n A cc es s D at ab as e w w w .ite ch on lin e. co m