Under-canopy UAV laser scanning for accurate forest field measurements
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Surveying and robotic technologies are converging, offering great potential for robotic-assisted data collection and support for labour intensive surveying activities. From a forest monitoring perspective, there are several technological and operational aspects to address concerning under-canopy flying unmanned airborne vehicles (UAV). To demonstrate this emerging technology, we investigated tree detection and stem curve estimation using laser scanning data obtained with an under-canopy flying UAV. To this end, we mounted a Kaarta Stencil-1 laser scanner with an integrated simultaneous localization and mapping (SLAM) system on board an UAV that was manually piloted with the help of video goggles receiving a live video feed from the onboard camera of the UAV. Using the under-canopy flying UAV, we collected SLAM-corrected point cloud data in a boreal forest on two 32 m × 32 m test sites that were characterized as sparse (n = 42 trees) and obstructed (n = 43 trees), respectively. Novel data processing algorithms were applied for the point clouds in order to detect the stems of individual trees and to extract their stem curves and diameters at breast height (DBH). The estimated tree attributes were compared against highly accurate field reference data that was acquired semi-manually with a multi-scan terrestrial laser scanner (TLS). The proposed method succeeded in detecting 93% of the stems in the sparse plot and 84% of the stems in the obstructed plot. In the sparse plot, the DBH and stem curve estimates had a root-mean-squared error (RMSE) of 0.60 cm (2.2%) and 1.2 cm (5.0%), respectively, whereas the corresponding values for the obstructed plot were 0.92 cm (3.1%) and 1.4 cm (5.2%). By combining the stem curves extracted from the under-canopy UAV laser scanning data with tree heights derived from above-canopy UAV laser scanning data, we computed stem volumes for the detected trees with a relative RMSE of 10.1% in both plots. Thus, the combination of under-canopy and above-canopy UAV laser scanning allowed us to extract the stem volumes with an accuracy comparable to the past best studies based on TLS in boreal forest conditions. Since the stems of several spruces located on the test sites suffered from severe occlusion and could not be detected with the stem-based method, we developed a separate work flow capable of detecting trees with occluded stems. The proposed work flow enabled us to detect 98% of trees in the sparse plot and 93% of the trees in the obstructed plot with a 100% correction level in both plots. A key benefit provided by the under-canopy UAV laser scanner is the short period of time required for data collection, currently demonstrated to be much faster than the time required for field measurements and TLS. The quality of the measurements acquired with the under-canopy flying UAV combined with the demonstrated efficiency indicates operational potential for supporting fast and accurate forest resource inventories.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it