研究目的
To define a methodology for rapid mapping of vegetated terraces using high-resolution airborne LiDAR in complex environments with canopy cover, focusing on the Liguria region in Italy.
研究成果
The proposed methodology successfully identified 448 ha of terraces in the Rupinaro basin, with 95% under canopy cover, demonstrating the superiority of LiDAR over photo-interpretation for mapping in vegetated areas. Careful planning of LiDAR survey parameters is crucial for accurate detection, and the approach is effective for rapid mapping but requires high data density and may need optimization for broader applications.
研究不足
The methodology requires high-density LiDAR data, which can be costly and data-intensive to handle. It may not perform well in areas with very dense vegetation or snow cover, and the geofilter relies on the availability and accuracy of external land use datasets. Validation was limited to accessible areas, and some densely vegetated terraces could not be ground-truthed.
1：Experimental Design and Method Selection:
The methodology involves a workflow starting with LiDAR survey planning to maximize ground visibility and detail, followed by point cloud processing using FUSION software for ground pulse extraction and conversion to ASCII raster DTMs. Morphometric queries (slope angle and relief) are applied with specific thresholds for terrace detection, and GIS-based postprocessing includes thematic filtering to exclude non-terraced flat areas.
2：Sample Selection and Data Sources:
The case study is the Rupinaro basin in Liguria, Italy, covering 1140 ha with mixed land cover including forest, olive orchards, cultivated areas, urban areas, and industrial zones. LiDAR data were acquired specifically for this area, and additional data sources include land use maps and technical maps from the Liguria region geoportal.
3：List of Experimental Equipment and Materials:
Airborne LiDAR sensor (RIEGL LMS-Q680i), helicopter (Eurocopter AS 350), GPS receiver (Leica 1200 with GSM modem and VRS correction), software (FUSION, R, QGIS), and datasets (LiDAR point cloud, orthoimages, land use maps).
4：Experimental Procedures and Operational Workflow:
LiDAR survey was conducted in foliage-free periods (January to April 2015) with high point density settings. Point cloud was filtered to ground pulses, converted to raster DTMs, and processed with morphometric algorithms. A geofilter was created from land use and technical maps to remove non-terraced areas, and validation was performed through manual mapping and GPS surveys.
5：Data Analysis Methods:
Statistical analysis involved comparing detected terraces with manual mappings and GPS surveys to assess accuracy. R scripting was used for batch processing of raster data, and QGIS for spatial analysis and map production.
独家科研数据包，助您复现前沿成果，加速创新突破
获取完整内容
