Applying deep learning to imaging sonar for the automated detection, classification and counting of untagged fish in fish passages
Bibliographic record
Abstract
The Ocean Aware project, led by Innovasea and funded through Canada’s Ocean Supercluster, is developing the next generation of underwater observation systems to transform fishing, aquaculture, and marine energy. One facet of this project is developing innovative methods for real-time tracking of untagged fish and species at risk around man-made infrastructures, such as hydropower dams, that present barriers to fish passage. The system is based on applying modern Deep Learning techniques to automatically detect and classify fish and their species from high resolution imaging sonar. In this paper, we present our results from applying adaptations of the widely used YOLO machine learning model to detect and classify multiple species of fish from a public dataset containing eight distinct species of fish from the Ocqueoc River captured by a high resolution DIDSON imaging sonar. Our results demonstrate that deep learning models can indeed be used to detect, classify species, and track fish using high resolution imaging sonar. Although there has been extensive research in the literature identifying particular fish, such as eel versus non-eel and seal versus fish, to our knowledge this is the first successful application of deep learning for classifying multiple fish species with high resolution imaging sonar.
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 | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| 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".