A Deep Learning algorithm for accurate and fast identification of coral reef fishes in underwater videos
Bibliographic record
Abstract
Identifying and counting individual fish on videos is a crucial task to cost-effectively monitor marine biodiversity, but it remains a difficult and time-consuming task. In this paper, we present a method to assist the automated identification of fish species on underwater images, and we compare our algorithm performances to human ability in terms of speed and accuracy. We first tested the performance of a convolutional neural network trained with different photographic databases while accounting for different post-processing decision rules to identify 20 fish species. Finally, we compared the performance in species identification of our best model with human performances on a test database of 1197 pictures representing nine species. The best network was the one trained with 900 000 pictures of whole fish and of their parts and environment (e.g. reef bottom or water). The rate of correct identification of fish was 94.9%, greater than the rate of correct identifications by humans (89.3%). The network was also able to identify fish individuals partially hidden behind corals or behind other fish and was more effective than humans identification on smallest or blurry pictures while humans were better to recognize fish individuals in unusual positions (e.g. twisted body). On average, each identification by our best algorithm using a common hardware took 0.06 seconds. Deep Learning methods can thus perform efficient fish identification on underwater pictures which pave the way to new video-based protocols for monitoring fish biodiversity cheaply and effectively.
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 | 0.001 |
| 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) | 0.002 | 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".