Use of seabed classification methodology to obtain species information from fisheries acoustic data
Bibliographic record
Abstract
No abstracts are to be cited without prior reference to the author.Recent developments and success in acoustic seabed classification has prompted us to use ideas from this field to explore the important and limiting question of species discrimination in fisheries acoustic data. Many parallels exist between backscattered signals from the seabed and from fish/plankton especially when we view echograms as echo intensity maps analogous to multibeam seabed images. Our approach is similar to that used for multibeam seabed classification. We divide the echogram into cells, each containing echo intensity information from a small depth interval and a small number of pings. Image processing techniques then capture features that characterize the acoustic structure of each cell. PCA is then used to find the most effective feature combinations. This is followed by K-means partition for unsupervised classification. Classes reflect the acoustic diversity of the cells, a prerequisite for species classification. The process is useful to find features that will yield a high level of acoustic diversity and to explore the amount of species information that is available. We tested a subset of this approach on data that were collected with a 38 kHz Simrad EK500 echo sounder during a Pacific hake (Merluccius productus) assessment survey in the Strait of Georgia, British Columbia, Canada. The analysis focused on acoustic returns from hake, rockfish, and plankton aggregations that were 'ground-truthed' with catch information. More than 50 cells with different species, densities and depths were selected. Features were obtained from two multi-fractal analysis methods, which were applied to backscatter (SV) data without, and with thresholds of -80 and -70 dB. Each method generated a set of 33 features. Sets were submitted individually and in combination to PCA. The first three principal components contained 90% or more of the variance. Feature classification with K-means partitioning generated classes that showed significant acoustic diversity and good correspondence between acoustic classes and species. These results are encouraging and tests on a larger data set of single and multi-frequency data and with a wider range of different features are planned. For multi-frequency data, dB differencing could provide additional features.
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 | 0.004 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.005 | 0.004 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.003 | 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".