Unsupervised landcover classification in a topographically diverse region of north-central British Columbia.
Bibliographic record
Abstract
Landcover mapping is an important tool for natural resource managers. The incorporation of remotely sensed data to produce landcover maps is becoming more common, as a result of lower cost and greater availability of image data. One factor that can negatively affect landcover classifications is topographic variation, particularly in mountainous areas. For this study, four unsupervised classification methods were compared using Landsat Thematic Mapper data (TM) to test the effect of topography on landcover classification. One scenario involved using only TM data, another combined TM data with an Incidence channel, a third used principal component analysis (PCA) and the last used selective principal component analysis (SPCA). Each classification was tested for two different classification legends. The first was based on habitat mapping while the other was a general landcover legend taken from a national mapping project. In each case, the imagery was pre-stratified using a normalized difference vegetation index and was clustered with the K-Means unsupervised classification method. Results showed that the combination of TM bands 3, 4, 5 and Incidence modestly increased the classification accuracy for both legends although accuracies were lower than 62%. The PCA and SPCA scenarios, which other research had shown held promise as a means of topographic compensation, did not produce higher accuracy without considerable class aggregation.
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 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.002 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.001 | 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".