Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2018 Dec 17;13(12):e0204713.
doi: 10.1371/journal.pone.0204713. eCollection 2018.

Automatic interpretation of otoliths using deep learning

Endre Moen  1 , Nils Olav Handegard  1 , Vaneeda Allken  1 , Ole Thomas Albert  1 , Alf Harbitz  1 , Ketil Malde  1   2
Affiliations

Automatic interpretation of otoliths using deep learning

Endre Moen et al. PLoS One. .

Abstract

The age structure of a fish population has important implications for recruitment processes and population fluctuations, and is a key input to fisheries-assessment models. The current method of determining age structure relies on manually reading age from otoliths, and the process is labor intensive and dependent on specialist expertise. Recent advances in machine learning have provided methods that have been remarkably successful in a variety of settings, with potential to automate analysis that previously required manual curation. Machine learning models have previously been successfully applied to object recognition and similar image analysis tasks. Here we investigate whether deep learning models can also be used for estimating the age of otoliths from images. We adapt a pre-trained convolutional neural network designed for object recognition, to estimate the age of fish from otolith images. The model is trained and validated on a large collection of images of Greenland halibut otoliths. We show that the model works well, and that its precision is comparable to documented precision obtained by human experts. Automating this analysis may help to improve consistency, lower cost, and increase the extent of age estimation. Given that adequate data are available, this method could also be used to estimate age of other species using images of otoliths or fish scales.

PubMed Disclaimer

Conflict of interest statement

The authors have declared that no competing interests exist.

Figures

Fig 1
Fig 1. Example of otolith images.
Otoliths can have loose fragments (A), vary in size (B), or be broken (C).
Fig 2
Fig 2. Age distribution of all 8875 images.
Fig 3
Fig 3. A pair of otoliths from 2014 with an estimated age of 13 years.
Due to the size difference between the otoliths, the image was split with a substantial offset from the middle (A). There was also a small horizontal overlap causing a fragment of the right otolith to remain in the left image. Resizing causes stretching of the images (B), which is particularly evident in the image of the left otolith.
Fig 4
Fig 4. Age predictions.
Predictions are shown using single otoliths (A) and using the average prediction of each pair (B), compared to the age estimated by a human reader.
Fig 5
Fig 5. Age predictions.
Predictions are shown for the right (A) and left (B) otoliths compared to the age as estimated by a human reader.
Fig 6
Fig 6. Examples of images where the network failed to correctly predict age.
A) Dark images of otoliths with deep lobes are read as 12 years, and predicted as 15.7 (left), and 15.6 (right). B) Lighter otoliths below are read as 21 years, and predicted as 15.6 (left and right).
See this image and copyright information in PMC

References

    1. Panfili J, De Pontual H, Troadec H, Wrigh PJ. Manual of fish sclerochronology. 2002;.
    1. Hilborn R, Walters CJ. Quantitative fisheries stock assessment: choice, dynamics and uncertainty. Reviews in Fish Biology and Fisheries. 1992;2(2):177–178. 10.1007/BF00042883 - DOI
    1. Aanes S, Vølstad JH. Gradient-based learning applied to document recognition. Canadian Journal of Fisheries and Aquatic Sciences. 2015;72:938–953.
    1. Talman S, Krusic-Golub K, Roberson S, Green C. Age estimation of deepwater fish species from the eastern north Atlantic. Final Report to the Bord Iascaigh Mhara (Irish Sea Fisheries Board)(Marine and Freshwater Resources Institute, Department of Primary Industries: Queenscliff, Vic, Australia). 2003;.
    1. Campana S. Accuracy, precision and quality control in age determination, including a review of the use and abuse of age validation methods. Journal of fish biology. 2001;59(2):197–242. 10.1111/j.1095-8649.2001.tb00127.x - DOI

Publication types

MeSH terms