Paper Overview
- Field: Computer Vision / Remote Sensing
- Authors: Natsuki Yuda, Shinichiro Shirota
- Posted: 2026-11-06
- arXiv: 2611.09854
- In steep forests of western Japan, the authors super-resolved existing 10 m-resolution DTMs to 1 m. Improving their accuracy required fusion with a global 30 m-resolution DTM from a different source.
- Using the resulting DTM, they first translated GEDI observations along the ground track as a whole, then corrected per individual shot.
- For most ground tracks, the translation nearly matched the airborne LiDAR DTM result — likely because DTM errors cancel out across many observations.
- Per-shot corrections, however, were inconsistent: errors do not cancel on a single shot, are too large for correction purposes, and mainly originate from the input DTM.
What the Paper Does
The Global Ecosystem Dynamics Investigation (GEDI) mission measures three-dimensional forest structure with a LiDAR aboard the International Space Station. GEDI observations have geolocation errors of about 10 m, which cause misalignment with other remote sensing data and reduce the accuracy of downstream tasks. Matching GEDI ground elevations against high-resolution digital terrain models (DTMs) from airborne LiDAR is one way to correct these errors, but such DTMs exist in few regions.
To widen this method's applicability, the authors super-resolved a more widely available low-resolution DTM and used it for the matching. The central question: does a super-resolved DTM yield the same correction as a 1 m-resolution airborne LiDAR DTM — and if not, why?
Findings
Conclusion
Even without a 1 m-resolution DTM, super-resolving a ~10 m-resolution DTM fused with a global DTM to 1 m should produce trajectory-level GEDI geolocation corrections nearly identical to those from a 1 m-resolution airborne LiDAR DTM.
--- *Auto-collected 2026-11-10*