Remote Sensing (Jul 2022)
Similarity Analysis between Contour Lines by Remotely Piloted Aircraft and Topography Using Hausdorff Distance: Application on Contour Planting
Abstract
Contour planting minimizes soil degradation, making agricultural production more sustainable. Currently, geotechnologies can provide more precise and fast data from relief than rudimentary data acquisition for agricultural management. Thus, the objective of this work was to analyze the similarities between contour lines from topography and Remotely Piloted Aircraft, using the Hausdorff distance algorithm. This study was carried out in the period between January 2020 and November 2021 in four localities in the State of Rio de Janeiro, Brazil: two areas located in the municipality of Bom Jardim and two areas in the municipality of Seropédica. Data were acquired through a conventional topographic survey and an aerial photogrammetric survey by Remotely Piloted Aircraft. From the acquired field data for the studied areas, the Digital Elevation Models were generated with a spatial resolution of 0.20 m and the contour lines with an equidistance of one meter. The contour lines obtained by both techniques were superimposed and their similarity was verified using the Hausdorff distance. The results show that there was a better similarity among the contour lines in areas with a very rugged relief than in a smooth relief. Also, the lowest altimetric differences observed in the Digital Elevation Models were associated with the smallest Hausdorff distance. These adjustments correspond, respectively, to the segments between the contour lines with the best and the worst individual similarity for each area. We observed that the similarity between the contour lines from topography and RPA yielded slope differences lower than 6.1% for at least 95% of all studied areas. The Hausdorff distance analysis allowed us to conclude that contour planting can be performed from data obtained via Remotely Piloted Aircraft, provided that vertical accuracy analysis controls the quality of the Digital Elevation Models.
Keywords