Abstract:
High accurate railway maps and terrain information (Digital Elevation Models) is a
major concern for future railway constructions and railway lines development.The
mapping of railway line using ground based surveying techniques istime
consuming and problematic.Unmanned Aerial Vehicles (UAV) technology has
revolutionized the aerial photogrammetric mapping due to its low cost and high
spatial resolution. It enables mapping the land use with greater accuracy in both 2D
and 3D. The “DJI Phantom 4“ drone was selected as the UAV platform to acquire
image data. In this study, we have developed a fully automated and highly accurate
engineering approach for detecting land use and railway line, which is based on
textural information from orthophoto and elevation information (Digital Surface
Models)obtained from the drone.The Pix4D software was used to develop the
orthophoto and a Digital Surface Model (DSM) and the DSM was validated by using
the ground control points.The rule sets knowledge-based classification method in
object oriented classification was used to classify the land use and railway with the
use of “eCognition“ software. Finally, the results were compared with digitize land
use layer to validate the results, and obtained overall accuracy of 90.15%.
