Mangrove and Salt Marsh Detection in a Mangrove-saltmarsh Ecotone Using Segment Anything Model from Drone Imagery
Bibliographic record
Abstract
Mangroves and salt marshes coexist in the intertidal wetlands of many temperate and subtropical coastal regions, forming many mangrove-saltmarsh ecotones. They provide a wealth of ecological services, such as carbon sequestration, habitat provision, climate regulation and stabilization, water purification and conservation, flood protection, biodiversity, atmospheric maintenance, and etc. But the heterogeneous, fragmented and dynamic intertidal wetlands make it challenging for the detailed and precise monitoring of mangroves and salt marshes. In this paper, we combined Segment Anything Model (SAM), which is known for the exceptional generalization capabilities and zero-shot learning, and the red-green ratio index (RGRI) to detect mangroves and salt marshes from drone imagery in a representative mangrove-saltmarsh ecotone in Guangxi, China. The SAM was first used to segment the imagery into image segments, then the RGRI value was calculated and RGRI thresholds was used to discriminate mangroves and salt marshes. As the coastal background environment is complex, manual visual interpretation was last used to modify the mangrove and salt marsh detection results. By comparing the detection results with those based on multi-scale segmentation object-oriented classification method, we found that the combined SAM and RGRI method can produce more accurate boundary of the mangrove-saltmarsh ecotone, especially for the single mangrove trees, but might misidentify the small-area dense mangrove forests located among salt marshes. The detection accuracies of mangroves and salt marshes based on our method are 83.23% and 95.13%, respectively. The results reflect the potential of fine mapping of mangroves and salt marshes in complex mangrove-saltmarsh ecotones by SAM from super-high resolution drone imagery, contributing to the intelligent protection and management of the blue carbon ecosystems in China.
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.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 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 itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".