Optimizing a rotating thermal-IR system to automatically detect marine mammals in Atlantic Canada
Bibliographic record
Abstract
We are interested in making a comparison of three methods to detect marine mammals at sea offshore of Atlantic Canada: marine mammal observers (MMOs) making visual observations, MMOs assisted by a thermal-IR (infrared) automatic detection system, and passive acoustic monitoring (PAM). Prior to making this comparison, the detection and classification algorithms for the thermal-IR system required optimization for use in the thermal regime offshore Atlantic Canada. In summer 2015, we made visual observations concurrent with the collection of thermal-IR data at a shore-based observation site at Cape Race, Newfoundland. A total of 1114 location fixes on marine mammals were made using a theodolite: humpback whales (n = 967), minke whales (n = 112), harbour porpoises (n = 10), unidentified baleen whales (n = 16), and unidentified whales (n = 9). Thermal imagery data were retrospectively scanned and thermal anomalies that could be identified as marine mammals were found to match 700 of the 1114 fixes. Of the remaining 414 fixes, 366 were not found in the images, and 48 were undetermined. This dataset was used to optimize the detection and classification algorithms prior to the 2016 field season. In general, ≥70% of marine mammal sightings made by MMOs within 3 km of the shore-based observation site were discernible in thermal-IR imagery during periods when the Beaufort wind force was ≤ 6, for all sighting cues (e.g., blow, body) and species combined. The three detection methods will be compared during a research cruise in summer 2017. The results should be useful in the development of best practice guidelines for marine mammal mitigation monitoring during seismic surveys in Canada.
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) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| 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".