We present a literature review to analyze the state of the art in the area of road detection based upon frontal images. For this purpose, a systematic literature review (SLR) was conducted that focuses on analyzing region-based works, since they can adapt to different surface types and do not depend on road geometry or lane markings. Through the comprehensive study of publications in a 11-year time frame, we analyze the methods that are being used, on which types of surface they are applied, whether they are adaptive in relation to surface changes, and whether they are able to distinguish possible faults or changes in the road, such as potholes, shadows, and puddles.

No abstract

Obstacle detection is a key issue in many current applications, especially in applications that have been increasingly highlighted such as: advanced driver assistance systems (ADAS), simultaneous localization and mapping (SLAM) and autonomous navigation system. This can be achieved by active and passive acquisition vision systems, for example: laser and cameras respectively. In this paper we present a comparison between low-cost active and passive devices, more specifically LIDAR and two cameras. To this comparison a disparity map is created by stereo correspondence through two images and a point cloud map created by LIDAR data values (distances measures). The obtained results shown that passive vision can be as good as or even better than active vision in low cost scenarios.

This paper describes a fast image segmentation approach designed for pavement detection in a moving camera. The method is based on a graph-oriented segmentation approach where gradient information is used temporally as a system of discontinuities to control merging between adjacent regions. The method presumes that the navigable path usually is located at specific positions on the scene, and a predefined set of seed points is used to locate the region of interest. The obtained results shown the proposed approach is able to accurately detect in an inexpensive computation manner the navigable path even in non-optimum scenarios such as miss-painted or unpaved dirt roads. Validation was conducted using a dataset with 701 samples of navigable paths, presenting a very high precision for real-time applications.