Accuracy Evaluation and Sensitivity Analysis of Estimating 3D Road Centerline Length using Lidar and NED

Highway networks are represented by linear spatial objects (road segments). Having accurate length information of road centerlines is critical in transportation. This paper presents a geographic information system (GIS)-based approach that overlays planimetric road centerlines and elevation data to model road centerlines in a 3D space and estimate their lengths. Elevation sources included light detection and ranging (lidar) and the National Elevation Dataset (NED). The estimated distances were compared to distance measurement instrument (DMI)-measured distances to evaluate the accuracy. The effects of elevation datasets with varying vertical accuracies were assessed. The relationship between road geometric properties and the accuracy of distance estimates was examined. We found that (a) the proposed 3D approach is efficient in estimating 3D road centerline distances, (b) using lidar point data improves the accuracy by 28 percent over the use of NED, and (c) certain road geometric properties have direct relationship with the accuracy of distance estimates. Introduction Highway networks are represented by a rich set of linear spatial objects (road segments). These geo-spatial transportation segments are connected to form topological networks that provide the basis for location referencing systems to locate features such as bridges, signs, pavement conditions, and traffic accidents, and can be used to more accurately measure over-the-road travel distances between geographic locations (FGDC, 1994). Furthermore, these topological networks can be used to find shortest paths, to determine efficient routes, and to estimate traffic volumes when combined with the variety of network analysis tools. Transportation data is usually referenced to road networks by using a one-dimensional (1D) linear referencing model (Quiroga, 1999). With this model, objects along a network are located using a set of known points on the network and distances and directions from the known points to the objects along the linear objects. All linear PHOTOGRAMMETRIC ENGINEER ING & REMOTE SENS ING J u n e 2009 657 Hubo Cai is with the Department of Civil and Construction Engineering, Western Michigan University, 1903 West Michigan Avenue, Kalamazoo, MI 49009 ([email protected]). William Rasdorf is with the Department of Civil, Construction, and Environmental Engineering, North Carolina State University, Raleigh, NC 27695 ([email protected]). Photogrammetric Engineering & Remote Sensing Vol. 75, No. 6, June 2009, pp. 657–665. 0099-1112/09/7506–0657/$3.00/0 © 2009 American Society for Photogrammetry and Remote Sensing Accuracy Evaluation and Sensitivity Analysis of Estimating 3D Road Centerline Length using Lidar and NED Hubo Cai and William Rasdorf referencing methods are based on this concept (Baker and Blessing, 1994) which makes distance critical in transportation spatial databases. Examining various existing transportation spatial databases reveals that there is no correct and consistent distance information (Rasdorf et al., 2003). The integration of linear referencing data with different linear referencing methods and the integration of linear referencing data with other datasets (global positioning system (GPS) data, for example) reveal that errors vary widely across databases. There is a need to obtain accurate and consistent distance information and to use this information to update existing transportation spatial databases. This is not a trivial task taking into consideration the number of roads and the mileages of these roads in the United States (Rasdorf et al., 2004). A number of techniques are available to obtain distances along linear objects. Distance can be extracted from construction and design drawings. This approach suffers from the problems of currency and completeness. A distance measurement instrument (DMI) and an inertial navigation system (INS) might be used to measure the distance directly along lines. The DMI itself is an inexpensive technology consisting of a mechanical device attached to one or more of the wheels of a van and connected to an in-vehicle recorder (Karimi et al., 2000). However, a DMI requires frequent calibration to minimize errors that accumulate as the result of travel. An INS uses accelerometers and gyroscopes to provide pitch, roll, and heading information to derive a relative geo-referencing for a point. An INS has high accuracy over small distances and is usually more accurate than GPS (depending on types of GPS) but much more costly (Karimi et al., 2000). Ground surveying is another technology that can be used to obtain distances along linear objects with a high accuracy when working with straight lines only. GPS is a very promising technique for measuring linear objects; however, the most notable difficulties in using this technology stem from signal blockage in certain areas caused by thick tree canopies, bridges, or multi-path problems caused by signal deflection by high-rise buildings. In addition to the difficulties and limitations mentioned above, all these approaches suffer from a common disadvantage: they are very time-consuming to use to obtain distance information for a large number of roads. 657–665_07-104.qxd 5/16/09 4:48 PM Page 657