Traffic sign detection is an important task for transportation management. This task is usually done manually or semi-automatically using images captured by onboard cameras or LIDAR data obtained from airplanes, unmanned aerial vehicles, or vans. This is very time-consuming and expensive. This process includes the identification of objects that are on the road or its proximity. The previous works mainly focus on detecting and recognizing traffic signs based on images captured by onboard cameras. Visual features of traffic signs such as color, shape, and appearance, however, are often sensitive to illumination conditions, angle of view, etc. Except for the camera, LIDAR also provides important and alternative features of traffic signs. LIDAR is an active sensor that can capture a point cloud of XYZ points with intensity values. Intensity values correlate to the strength of the returned laser pulse, which depends on the reflectivity of the object and the wavelength used by the LIDAR. This characteristic can provide an important alternative approach for capturing traffic signs, since signs are required to have reflective material for nighttime driving, we can use this property to our advantage when discriminating point clouds for signs. In most previous works, different colors of traffic signs are individually handled in a specific color space, which generally results in the use of many thresholds or multiple classifiers. In this research, we will compare information that can be obtained from camera images and LiDAR measurements. This comparison will be presented for three example objects: traffic signs, road markings, and general pole-shaped objects (e.g. city lights or trees). Further, we will identify a process based on our algorithm that detects traffic signs in LiDAR measurement and transforms the results into a common format used in geographic information systems. This method will be tested on an approximately two-kilometer-long road in an urban area.
California State University, Fresno
This research will provide a good understanding of how Mobile LiDAR System in transportation is used and how these transportation features are stored and inventoried.