Automatic cloud detection in GHGSat satellite imagery
Bibliographic record
Abstract
GHGSat operates a constellation of satellites that detect and quantify methane and carbon dioxide emissions from industrial facilities across the globe. With twelve satellites in orbit, each making around fifty observations per day, automatic data processing is required. A key step in the automation process is the detection of clouds. Identifying pixels that contain clouds or cloud shadow can improve the retrieval quality of cloudy observations and make it easier to detect greenhouse gas emissions. In this presentation, we discuss the ML/AI techniques used to detect and segment clouds in GHGSat imagery. We highlight some of the challenges encountered during the creation of training datasets and model training. A first guess at cloud masks is obtained with an unsupervised clustering approach to group pixels of similar intensity. Then, using a dataset of 1000 human-annotated observations, we compare the performance of U-NET and Mask2Former models trained for cloud segmentation. We discuss how the monitoring of training loss can help to identify problematic examples. Finally, we investigate the creation of cloud shadow masks using geometrical projections of the cloud masks, where cloud height is estimated through an intensity-based optimisation.
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 | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.002 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.001 |
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".