Pre-trained Models for Sonar Images
Open accessAbstract
Machine learning and neural networks are now ubiquitous in sonar perception, but it lags behind the computer vision field due to the lack of data and pre-trained models specifically for sonar images. In this paper we present the Marine Debris Turntable dataset and produce pre-trained neural networks trained on this dataset, meant to fill the gap of missing pre-trained models for sonar images. We train Resnet 20, MobileNets, DenseNet121, SqueezeNet, MiniXception, and an Autoencoder, over several input image sizes, from $32\times 32$ to $96\times 96$, on the Marine Debris turntable dataset. We evaluate these models using transfer learning for low-shot classification in the Marine Debris Watertank and another dataset captured using a Gemini 720i sonar. Our results show that in both datasets the pre-trained models produce good features that allow good classification accuracy with low samples (10-30 samples per class). The Gemini dataset validates that the features transfer to other kinds of sonar sensors. We expect that the community benefits from the public release of our pre-trained models and the turntable dataset.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
AUSCC was introduced to provide an unbiased evaluation of ML models for classification of underwater acoustic signals, taking into account sample sizes per class.
“The notion of area under the sample complexity curve (AUSCC) was recently introduced in [88] to provide an unbiased evaluation of ML models for the classification of underwater acoustic signals, taking into account the sample sizes per class.”
From A Survey of Underwater Acoustic Data Classification Methods Using Deep Learning for Shoreline Surveillance (2022) · cites this paper as [88]
In “Transfer Learning on Watertank Dataset” (p. 5), the paper introduces AUSCC to compare model accuracy across different numbers of training samples per class. It evaluates classification of sonar images, but does not describe AUSCC as an unbiased evaluation or establish its use for underwater acoustic signals generally. (LLM, conf=0.86)
Experimental — accuracy not yet measuredResults support that transfer learning using pretrained models can be used successfully to obtain high accuracy in the classification of sonar data using fewer samples to retrain the models.
“The results presented support the authors’ initial hypothesis that transfer learning using pretrained models can be used successfully to obtain high accuracy in the classification of sonar data, using fewer samples to retrain the models.”
From A Survey of Underwater Acoustic Data Classification Methods Using Deep Learning for Shoreline Surveillance (2022) · cites this paper as [88]
The Abstract (p. 1) reports good sonar classification accuracy with 10–30 samples per class, and the Discussion (p. 7) says the pre-trained models can reduce sample complexity. However, the Watertank methods (p. 5) describe extracting features from those models and training a support vector machine, not retraining the pre-trained models themselves. (LLM, conf=0.85)
Experimental — accuracy not yet measuredPretrained neural networks for sonar images have been produced to provide a basis for improving sonar classification results by means of transfer learning.
“Following from the finding that TL is potentially beneficial, a set of pretrained neural networks for sonar images have been produced which are aimed at providing a basis for improving sonar classification results by means of transfer learning [88].”
From A Survey of Underwater Acoustic Data Classification Methods Using Deep Learning for Shoreline Surveillance (2022) · cites this paper as [88]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
Deep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.90 · Cited by 235,474 · see citing works
Densely Connected Convolutional Networks
Huang, G., Liu, Z., Van Der Maaten, L., & Weinberger, K. Q. (2017) · 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2017.243 · Cited by 48,566 · see citing works
ImageNet Large Scale Visual Recognition Challenge
Russakovsky, O., Deng, J., Su, H., Krause, J., Satheesh, S., Ma, S., Huang, Z., Karpathy, A., Khosla, A., Bernstein, M., et al. (2015) · International Journal of Computer Vision · doi:10.1007/s11263-015-0816-y · Cited by 39,380 · see citing works
MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications
Howard, A. G., Zhu, M., Chen, B., Kalenichenko, D., Wang, W., Weyand, T., Andreetto, M., & Adam, H. (2017) · arXiv · doi:10.48550/arxiv.1704.04861 · Cited by 19,015 · see citing works
SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB model size
Iandola, F. N., Han, S., Moskewicz, M. W., Ashraf, K., Dally, W. J., & Keutzer, K. (2016) · arXiv · doi:10.48550/arxiv.1602.07360 · Cited by 6,201 · see citing works
CNN Features Off-the-Shelf: An Astounding Baseline for Recognition
Razavian, A. S., Azizpour, H., Sullivan, J., & Carlsson, S. (2014) · 2014 IEEE Conference on Computer Vision and Pattern Recognition Workshops · doi:10.1109/cvprw.2014.131 · Cited by 4,451 · see citing works
What makes ImageNet good for transfer learning?Preprint
Huh, M., Agrawal, P., & Efros, A. A. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1608.08614 · Cited by 591 · see citing works
Object Recognition in Forward Looking Sonar Images using Transfer Learning
Fuchs, L. R., Gallstrom, A., & Folkesson, J. (2018) · 2018 IEEE/OES Autonomous Underwater Vehicle Workshop (AUV) · doi:10.1109/auv.2018.8729686 · Cited by 40 · see citing works
Deep Neural Networks for Marine Debris Detection in Sonar ImagesPreprint
Valdenegro-Toro, M. (2019) · arXiv.org · doi:10.48550/arxiv.1905.05241 · Cited by 31 · see citing works
Self-citationBest practices in convolutional networks for forward-looking sonar image recognition
Valdenegro-Toro, M. (2017) · OCEANS 2017 - Aberdeen · doi:10.1109/oceanse.2017.8084987 · Cited by 30 · see citing works
Self-citationAutomatic Target Recognition in Forward-Looking Sonar Images using Transfer Learning
Lou, G., Zheng, R., Liu, M., & Zhang, S. (2020) · Global Oceans 2020: Singapore – U.S. Gulf Coast · doi:10.1109/ieeeconf38699.2020.9389217
Flagged first, then by citations
Research works printed but not linked: 1
- [3][3] A. Kuznetsova, H. Rom, N. Alldrin, J. Uijlings, I. Krasin, J. Pont-Tuset, S. Kamali, S. Popov, M. Malloci, A. Kolesnikov, et al. , 'The open images dataset v4,' International Journal of Computer Vision , vol. 128, no. 7, pp. 1956-1981, 2020. -
Cited by
Sorted by citations
Domingos, L. C. F., Santos, P. E., Skelton, P. S. M., Brinkworth, R. S. A., & Sammut, K. (2022) · Sensors · doi:10.3390/s22062181 · Cited by 93 · see citing works
“Analogous to the work described in [88], a CNN was pretrained with images from the ImageNet dataset [51] and fine-tuned with a small sonar dataset for the classification of forward-looking sonar (FLS) data [129].”
BackgroundMethodInfluential
- Authors
- Valdenegro-Toro, M., Preciado-Grijalva, A., & Wehbe, B.
- Affiliations
- German Research Center for Artificial Intelligence
- German Research Centre for Artificial Intelligence (DE)
- ORCID
- Wehbe, Bilal
- Venue
- OCEANS 2021: San Diego – Porto
- Publisher
- IEEE
- Published
- 2021-09
- Pages
- 1-8
- OpenAlex
- W3188448245
- Cited by
- 21
- References
- 11
- Citations by year
- 2026: 2 · 2025: 9 · 2024: 4 · 2023: 1 · 2022: 5
- FWCI
- 9.46
- Type
- conference paper
- Language
- en
- Open access
- green
- arxiv.org
- Locations
- MTS/IEEE OCEANS Conference (OCEANS) (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- Publication Server of Bonn-Rhein-Sieg University of Applied Sciences (Bonn-Rhein-Sieg University of Applied Sciences) (submittedVersion, cc-by)
- arXiv (Cornell University) (cc-by)
- Funders
- Heriot-Watt University
- European Commission (H2020-ICT, 101016958, HORIZON2020, H2020)
- Horizon 2020 Framework Programme
- Topics
- Physical Sciences › Earth and Planetary Sciences › Oceanography › Underwater Acoustics Research
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- Physical Sciences › Computer Science › Artificial Intelligence › Domain Adaptation and Few-Shot Learning
- Keywords
- sonar images, pre-trained models, transfer learning, convolutional neural networks, autoencoder, underwater target recognition
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
