Silvilaser 2019 - Poster Presentations »
The Impact of Point Density for the generation of LIDAR derived Terrain Models
The aim of this study is to evaluate the impact of point density in the retrieval of ground points for further generation of digital terrain models (DTM). Multiple Airborne LIDAR (Light Detection and Ranging) datasets acquired over the Amazon under the Sustainable Landscape project were selected. Two study areas were selected inside the Tapajós National Forest (TNF) where a mosaic of different successional forest stages can be found due to shifting agriculture and land abandonment. Land cover types were generated based on available RapidEye and/or PlanetDove images acquired close to the LIDAR acquisitions. The performance of different freeware and commercial software such as BCAL, MCC, FUSION and TIFFS were also evaluated. Results show that BCAL, MCC and FUSION software responded better for the retrieving of ground points according to the increasing of point density. The quality of the ground generated models also improved with the increasing of point density. TIFFS performed similarly for the datasets regardless of the point density of the input data. However, residuals were always noticed over selected land cover types identified by both RapidEye and PlanetDove images. This study was sponsored by FAPESC and CNPq. The Satellite images were provided by PlanetScope. ALS datasets were provided by the Sustainable Landscape Project.