Comparative Analysis of Ultra-Wideband and Mobile Laser Scanning Systems for Mapping Forest Trees under A Forest Canopy

Abstract. In this paper, we present an ultra-wideband (UWB)-based method for mapping forest trees under a forest canopy and compare its precision with that of three commercial mobile laser scanning systems (MLS): Zeb Horizon (GeoSLAM, UK), Hovermap (Emesent, Australia), and Deep Forestry (Deep Forestry, Sweden). The proposed method is simple to implement in forest environments, requiring reduced human efforts. The comparison was based on real-world datasets collected in a boreal forest in Finland, covering an area of approximately 50 m × 110 m, with tree locations obtained from a high-density airborne laser scanning (ALS) system as a reference. To our best knowledge, this is the first study to compare UWB and MLS for mapping forest trees in the literature. The experimental results show that the proposed method can accurately measure tree stem locations under the forest canopy with a root-mean-square-error (RMSE) of 14.44 cm and a mean-absolute-error (MAE) of 12.39 cm, providing accuracy comparable to that of the three tested MLSs. Therefore, the proposed method is sufficient for forest surveying and management.