Best practices for the verification and validation of 3D absolute and predicted accuracy of LIDAR data utilizes both point and area ground control surveys. A comparison of both approaches indicates that the legacy (DGPS point survey) provides a rigid, well defined estimate of the measured total accuracy that can be directly compared with predicted accuracy models using the Generic Point-cloud Model (GPM 1.1) construct for modeling sensor measurement error. In contrast, the area ground control approach, often using terrestrial or drone based LIDAR sensors can provide a much higher density survey of the test site, but with a fundamentally different perspective on the absolute accuracy of the site under measurement and consequently the predicted accuracy of the survey. Comparisons of both approaches highlight the strengths and weaknesses of both approaches, especially in the characterization of non-traditional features including building corners, building centers, and building sides.
|