Search

High-resolution digital elevation models (HR-DEMs) originating from airborne laser scanning (ALS) point clouds must be transformed into Culvert-modified DEMs for hydrological and geomorphological analysis. To produce a culvert-modified DEM, information on the locations of drainage structures (DSs) (e.g., bridges and culverts) is essential. Nevertheless, DS mapping techniques, whether in connection with the development of new methods or an application setting of existing methods, have always been complicated. Consequently, wide area DS data are rare, making it challenging to produce a culvert-modified DEM in a wide area capacity. Alternatively, the breach algorithm (BA) method is a standard procedure to obtain culvert-modified DEMs in the absence of DS data, solving the problem to some extent. This paper addresses this shortcoming using a newly developed drainage structure mapping algorithm (DSMA) for obtaining a culvert-modified DEM for an area of 36 km2 in Vermont, USA. Benchmark DS data are used as a standard reference to assess the performance of the DSMA method compared to the BA method. A consistent methodological framework is formulated to obtain a culvert-modified DEM using DS data, mapped using the DSMA and resultant culvert-modified DEM is then compared with BA method respectively. The DSs found from the culvert-modified DEMs were reported as true positive (TP), false positive (FP), and false negative (FN). Based on TP, FP, and FN originating from the culvert-modified DEMs of both methods, the evaluation metrics of the false positive rate (FPR) (i.e., the commission error) and false negative rate (FNR) (i.e., the omission error) were computed. Our evaluation showed that the newly developed DSMA-based DS data resulted in an FPR of 0.05 with federal highway authorities (FHWA) roads and 0.12 with non-FHWA roads. The FNR with FHWA roads was 0.07, and with non-FHWA roads, it was 0.38. The BA method showed an FPR of 0.28 with FHWA roads and 0.62 with non-FHWA roads. Similarly, the FNR for the BA method was 0.32 with FHWA roads and 0.61 with non-FHWA roads. The statistics based on the FPR and FNR showed that the DSMA-based culvert-modified DEM was more accurate compared with the BA method, and the formulated framework for producing culvert-modified DEMs using DSMA-based DS data was robust. [ABSTRACT FROM AUTHOR]

Wide-area drainage structure (DS) mapping is of great concern, as many DSs are reaching the end of their design life and information on their location is usually absent. Recently, airborne laser scanning (ALS) has been proven useful for DS mapping through manual methods using ALS-derived digital elevation models (DEMs) and hillshade images. However, manual methods are slow and labor-intensive. To overcome these limitations, this paper proposes an automated DS mapping algorithm (DSMA) using classified ALS point clouds and road centerline information. The DSMA begins with removing ALS ground points within the buffer of the road centerlines; the size of the buffer varies according to different road classes. An ALS-modified DEM (ALS-mDEM) is then generated from the remaining ground points. A drainage network (DN) is derived from the ALS-mDEM. Candidate DSs are then obtained by intersecting the DN with the road centerlines. Finally, a refinement buffer of 15 m is placed around each candidate DS to prevent duplicate DS from being generated in close proximity. A total area of 50 km2, including an urban site and a rural site, in Vermont, USA, was used to assess the DSMA. Based on the road functional classification scheme of the Federal Highway Administration (FHWA), the centerline information regarding FHWA roads was obtained from a public data portal. The centerline information on non-FHWA roads, i.e., private roads and streets, was derived from the impervious surface data of a land cover dataset. A benchmark DS dataset was gathered from the transport agency of Vermont and was further augmented using Google Earth Street View images by the authors. The one-to-one correspondence between the benchmark DS and mapped DS for these two sites was then established. The positional accuracy was assessed by computing the Euclidian distance between the benchmark DS and mapped DS. The mean positional accuracy for the urban site and rural site were 13.5 m and 15.8 m, respectively. F1-scores were calculated to assess the prediction accuracy. For FHWA roads, the F1-scores were 0.87 and 0.94 for the urban site and rural site, respectively. For non-FHWA roads, the F1-scores were 0.72 and 0.74 for the urban site and rural site, respectively. [ABSTRACT FROM AUTHOR]