Low altitude photogrammetry for urban road mapping / Shahrul Nizan Abd Mukti and Khairul Nizam Tahar

Road connection has high impact on the city development. It helps boost the economic environment along the road. Therefore, it is important to maintain and divide by their traffic flow and road reserve well to determine the privilege of maintenance and budget contribution for every year. Road opens up the relation of intercity and urban as it gives the impact of development along the road. To manage road over the country, geometry data of road is needed for decision making and project well management. The primary data is usually contributed by field technical support persons, such as surveyor, engineer, and others for conventional method of survey, image along the road, computer aid drawing (cad data) as built drawing or topographical plan, and others. This study proposes an urban road mapping with optimal flight parameter and flying low for detail texture acquisition of feature. It ensures the high efficiency, low cost, short cycle, strong maneuverability, convenient operation, and others of a product. The objective of the project is to determine the optimal flight parameter in mapping out a road feature inside the road reserve with detailed digital orthophoto model (DOM) and digital elevation model (DEM). The flight parameters of unmanned aerial vehicle (UAV) and requirements for focal length effectiveness, flight planning preparation, image lap percentage, UAV altitude, and ground control point (GCP) distribution setup were outlined. The study investigated the effect of different focal length effects, GCP shape-based network (pyramid square-, square-, and linear-based networks), UAV altitude (90m, 65m, and 35m), and end lap percentage of image (90%, 80%, and 70%) on the photogrammetry-derived product. The 95m and 65m altitudes gave the lowest root mean square error (RMSE) value (±5cm horizontal and ±8cm vertical). In addition, 80% consistently showed the lowest RMSE for all end lap percentage options. Meanwhile, the pyramid square-based network recovered a total of 40% accuracy higher than square- and linear-based networks. This study could help the local authorities to implement smart road maintenance within their region.