Stereo vision-based road estimation assisted by efficient planar patch calculation

Abstract

A robust and accurate road model estimation algorithm can greatly improve the performance of many Advanced Driver Assistance Systems (ADAS) applications such as lane detection, obstacle detection and road marking recognition. To estimate the road model, the proposed algorithm employs a stereo vision camera system. In this paper, local planar patches are efficiently estimated in the disparity domain rather than conventionally in the Euclidean domain. Then, the estimated planar patch orientations are integrated to the fitting stage and orientation differences are minimized along with height differences. Moreover, patch orientation differences are exploited for weighting data points. Thus, outliers become insignificant in the fitting stage and the road model is estimated robustly and accurately without any prior knowledge of any extrinsic camera parameters. Experiments have been carried out for a free space calculation application and the road is segmented with a True Positive Rate (T P R) of 88%.

Self-Supervised Drivable Area and Road Anomaly Segmentation Using RGB-D Data For Robotic Wheelchairs

“…In recent years, some researchers have tried to solve this segmentation problem from other perspectives. For example, Ozgunalp et al [12] proposed to use the estimated planar patches and patch orientations to reduce the impact of outliers. Liu et al [13] proposed an approach to fuse the information of light detection and ranging (LIDAR) and vision data, respectively.…”

Robust lane‐detection algorithm based on improved symmetrical local threshold for feature extraction and inverse perspective mapping

“…While noise sources (cars, trees, pedestrians, traffic signs, and etc.) can be eliminated with a system using a 3D input such as the one we described in [8, 27], directional arrows appear on the same plane as the lane markings (on the road). Thus, this can be an issue even with a more complex hardware input.…”

Iterative roll angle estimation from dense disparity map

“…Imaging sensors such as digital cameras are usually used to obtain 3D images that can be used to monitor surrounding information. While some road fitting methods assume that the camera parameters are known [1] or the cameras are installed in such a way that their effects are negligible [2] [3], inaccurate estimation of camera parameters can compromise a system's reliability. Accurate estimation of roll angle is particularly important in systems that use the v-disparity map method [4].…”