CoChat
Sign Up
Cited by 49Underwater Acoustics ResearchAdvanced Neural Network ApplicationsUnderwater Vehicles and Communication Systems

Deep gabor neural network for automatic detection of mine-like objects in sonar imagery

Open access
Thanh Le, Hoang, Phung, Son Lam, Chapple, Philip B., Bouzerdoum, Abdesselam, Ritz, Christian H., Tran, Le Chung·2020·IEEE Access
DOI 10.1109/access.2020.2995390
Open this paper in CoChat — free accountOpen PDF

Abstract

With the advances in sonar imaging technology, sonar imagery has increasingly been used for oceanographic studies in civilian and military applications. High-resolution imaging sonars can be mounted on various survey platforms, typically autonomous underwater vehicles, which provide enhanced speed and improved data quality with long-range support. This paper addresses the automatic detection of mine-like objects using sonar images. The proposed Gabor-based detector is designed as a feature pyramid network with a small number of trainable weights. Our approach combines both semantically weak and strong features to handle mine-like objects at multiple scales effectively. For feature extraction, we introduce a parameterized Gabor layer which improves the generalization capability and computational efficiency. The steerable Gabor filtering modules are embedded within the cascaded layers to enhance the scale and orientation decomposition of images. The entire deep Gabor neural network is trained in an end-to-end manner from input sonar images with annotated mine-like objects. An extensive experimental evaluation on a real sonar dataset shows that the proposed method achieves competitive performance compared to the existing approaches.

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.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1109/access.2020.2995390

Cites

7 of 7 references verified0 citation errors · 0 not checkable

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.

A few free checks a day · papers already verified open instantly · results are public

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?

  • Gabor-like spatial functions were applied as a bioinspired preprocessing stage for automatic detection of minelike objects from side-scan sonar images.

    “Gabor-like spatial functions were applied as a bioinspired preprocessing stage in [62] for the automatic detection of minelike objects (MLOs) from side-scan sonar images.”

    From A Survey of Underwater Acoustic Data Classification Methods Using Deep Learning for Shoreline Surveillance (2022) · cites this paper as [62]

    The Introduction describes the Gabor approach as biologically motivated, and the Sonar Data Acquisition and Annotation section identifies the images as side-scan sonar images used for MLO detection. However, the Gabor Layer and Network Architecture sections describe trainable Gabor feature extractors within an end-to-end network, not a separate preprocessing stage. (LLM, conf=0.85)

    Experimental — accuracy not yet measured

  • The Deep Gabor Neural Network relied on a large number of training data to be tuned.

    “The Deep Gabor Neural Network [62] (cited in Section 3.2 above) also relied on a large number of training data to be tuned.”

    From A Survey of Underwater Acoustic Data Classification Methods Using Deep Learning for Shoreline Surveillance (2022) · cites this paper as [62]

    The Introduction says that training a detection model requires sufficient labelled data. The Sonar Data Acquisition and Annotation section reports that the authors expanded each fold from 153 original training images to 1,683 through augmentation. This supports that the network was trained with additional data, but the paper does not establish that it specifically relied on a large dataset to be tuned. (LLM, conf=0.83)

    Experimental — accuracy not yet measured

  • Much work in underwater acoustic data classification is conducted using synthetic data only or on a limited set of real data augmented with semisynthetic examples for training.

    “Therefore, much work in this area is conducted using synthetic data only [124], or on a limited set of real data augmented with semisynthetic examples for training [62,128,139].”

    From A Survey of Underwater Acoustic Data Classification Methods Using Deep Learning for Shoreline Surveillance (2022) · cites this paper as [62]

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

Flagged first, then by citations

Research works printed but not linked: 40

  1. D. A. Mely and T. Serre, Towards a Theory of Computation in the Visual Cortex. Singapore: Springer, 2017, pp. 59-84.
  2. R. P. Scobey and A. J. Gabor, ''Orientation discrimination sensitivity of single units in cat primary visual cortex,'' Exp. Brain Res., vol. 77, no. 2, pp. 398-406, Sep. 1989.
  3. D. B. Hamilton, D. G. Albrecht, and W. S. Geisler, ''Visual cortical receptive fields in monkey and cat: Spatial and temporal phase transfer function,'' Vis. Res., vol. 29, no. 10, pp. 1285-1308, Jan. 1989.
  4. R. A. Frazor, D. G. Albrecht, W. S. Geisler, and A. M. Crane, ''Visual cortex neurons of monkeys and cats: Temporal dynamics of the spa- tial frequency response function,'' J. Neurophysiol., vol. 91, no. 6, pp. 2607-2627, Jun. 2004.
  5. G. M. Boynton and E. M. Finney, ''Orientation-specific adaptation in human visual cortex,'' J. Neurosci., vol. 23, no. 25, pp. 8781-8787, Sep. 2003.
  6. F. Fang, S. O. Murray, D. Kersten, and S. He, ''Orientation-tuned fMRI adaptation in human visual cortex,'' J. Neurophysiol., vol. 94, no. 6, pp. 4188-4195, Dec. 2005.
  7. I. J. Goodfellow, A. Courville, and Y. Bengio, ''Scaling up Spike-and- Slab models for unsupervised feature learning,'' IEEE Trans. Pattern Anal. Mach. Intell., vol. 35, no. 8, pp. 1902-1914, Aug. 2013.
  8. V. Nair and G. E. Hinton, ''Rectified linear units improve restricted Boltzmann machines,'' in Proc. 27th Int. Conf. Mach. Learn., 2010, pp. 807-814.
  9. C. Szegedy, W. Liu, Y. Jia, P. Sermanet, S. Reed, D. Anguelov, D. Erhan, V. Vanhoucke, and A. Rabinovich, ''Going deeper with convolutions,'' in Proc. IEEE Conf. Comput. Vis. Pattern Recognit. (CVPR), Jun. 2015, pp. 1-9.
  10. G. Huang, Z. Liu, L. Van Der Maaten, and K. Q. Weinberger, ''Densely connected convolutional networks,'' in Proc. IEEE Conf. Comput. Vis. Pattern Recognit. (CVPR), Jul. 2017, pp. 4700-4708.
  11. J. Redmon and A. Farhadi, ''YOLOv3: An incremental improvement,'' 2018, arXiv:1804.02767. [Online]. Available: http://arxiv.org/abs/1804. 02767 · arXiv
  12. E. Coiras, Y. Petillot, and D. M. Lane, ''Multiresolution 3-D reconstruction from side-scan sonar images,'' IEEE Trans. Image Process., vol. 16, no. 2, pp. 382-390, Feb. 2007.
  13. P. Hollesen, W. A. Connors, and T. Trappenberg, ''Comparison of learned versus engineered features for classification of mine like objects from raw sonar images,'' in Proc. Can. Conf. Artif. Intell., May 2011, pp. 174-185.
  14. J. Sawas and Y. Petillot, ''Cascade of boosted classifiers for automatic target recognition in synthetic aperture sonar imagery,'' in Proc. Meetings Acoust. ECUA, vol. 17, no. 1, 2012, Art. no. 070074.
  15. C. Barngrover, R. Kastner, and S. Belongie, ''Semisynthetic versus real- world sonar training data for the classification of mine-like objects,'' IEEE J. Ocean. Eng., vol. 40, no. 1, pp. 48-56, Jan. 2015.
  16. C. Barngrover, A. Althoff, P. DeGuzman, and R. Kastner, ''A brain- computer interface (BCI) for the detection of mine-like objects in sides- can sonar imagery,'' IEEE J. Ocean. Eng., vol. 41, no. 1, pp. 123-138, Jan. 2016.
  17. D. Gebhardt, K. Parikh, I. Dzieciuch, M. Walton, and N. A. V. Hoang, ''Hunting for naval mines with deep neural networks,'' in Proc. IEEE OCEANS Conf., Sep. 2017, pp. 1-5.
  18. J. McKay, I. Gerg, V. Monga, and R. G. Raj, ''What's mine is yours: Pretrained CNNs for limited training sonar ATR,'' in Proc. IEEE OCEANS Conf., Sep. 2017, pp. 1-7.
  19. K. Denos, M. Ravaut, A. Fagette, and H.-S. Lim, ''Deep learning applied to underwater mine warfare,'' in Proc. Oceans Aberdeen, Jun. 2017, pp. 1-7.
  20. R. Girshick, J. Donahue, T. Darrell, and J. Malik, ''Rich feature hierarchies for accurate object detection and semantic segmentation,'' in Proc. IEEE Conf. Comput. Vis. Pattern Recognit., Jun. 2014, pp. 580-587.
  21. R. Girshick, ''Fast R-CNN,'' in Proc. IEEE Int. Conf. Comput. Vis. (ICCV), Dec. 2015, pp. 1440-1448.
  22. W. Liu, D. Anguelov, D. Erhan, C. Szegedy, S. Reed, C. Y. Fu, and A. C. Berg, ''SSD: Single shot multibox detector,'' in Proc. Eur. Conf. Comput. Vis., Oct. 2016, pp. 21-37.
  23. J. Redmon and A. Farhadi, ''YOLO9000: Better, faster, stronger,'' in Proc. IEEE Conf. Comput. Vis. Pattern Recognit. (CVPR), Jul. 2017, pp. 6517-6525.
  24. P. Viola and M. Jones, ''Rapid object detection using a boosted cascade of simple features,'' in Proc. IEEE Comput. Soc. Conf. Comput. Vis. Pattern Recognit. CVPR, Dec. 2001, pp. 511-518.
  25. Z. Zhu, X. Xu, L. Yang, H. Yan, S. Peng, and J. Xu, ''A model-based sonar image ATR method based on SIFT features,'' in Proc. Oceans TAIPEI, Apr. 2014, pp. 1-4.
  26. P. Tueller, R. Kastner, and R. Diamant, ''A comparison of feature detectors for underwater sonar imagery,'' in Proc. Oceans MTS/IEEE Charleston, Oct. 2018, pp. 1-6.
  27. J. Dale, A. Galusha, J. Keller, and A. Zare, ''Evaluation of image features for discriminating targets from false positives in synthetic aperture sonar imagery,'' Proc. SPIE, vol. 11012, May 2019, Art. no. 110120A.
  28. X. Li, R. Jimmy, L. Ce, and J. Jiaya, ''Deep convolutional neural network for image deconvolution,'' in Proc. Adv. Neural Inf. Process. Syst., 2014, pp. 1790-1798.
  29. J. P. Jones and L. A. Palmer, ''The two-dimensional spatial structure of simple receptive fields in cat striate cortex,'' J. Neurophysiol., vol. 58, no. 6, pp. 1187-1211, Dec. 1987.
  30. K. He, X. Zhang, S. Ren, and J. Sun, ''Identity mappings in deep residual networks,'' in Proc. Eur. Conf. Comput. Vis., Oct. 2016, pp. 630-645.
  31. K. Simonyan and A. Zisserman, ''Very deep convolutional networks for large-scale image recognition,'' in Proc. Int. Conf. Learn. Represent., 2014, pp. 1-6.
  32. J.-K. Kamarainen, V. Kyrki, and H. Kalviainen, ''Invariance properties of Gabor filter-based features-overview and applications,'' IEEE Trans. Image Process., vol. 15, no. 5, pp. 1088-1099, May 2006.
  33. P. B. Chapple, ''Unsupervised detection of mine-like objects in seabed imagery from autonomous underwater vehicles,'' in Proc. Oceans, Oct. 2009, pp. 1-6.
  34. M. Everingham, L. Van Gool, C. K. I. Williams, J. Winn, and A. Zisserman, ''The pascal visual object classes (VOC) challenge,'' Int. J. Comput. Vis., vol. 88, no. 2, pp. 303-338, Jun. 2010.
  35. C. L. Zitnick and P. Dollar, ''Edge boxes: Locating object proposals from edges,'' in Proc. Eur. Conf. Comput. Vis., 2014, pp. 391-405.
  36. T.-Y. Lin, M. Maire, S. Belongie, J. Hays, P. Perona, D. Ramanan, P. Dollar, and C. L. Zitnick, ''Microsoft COCO: Common objects in context,'' in Proc. Eur. Conf. Comput. Vis., Sep. 2014, pp. 740-755.
  37. SON LAM PHUNG (Senior Member, IEEE) received the B.Eng. degree (Hons.) in computer engineering and the Ph.D. degree in computer engineering from Edith Cowan University, Aus- tralia, in 1999 and 2003, respectively. He is cur- rently an Associate Professor with the School of Electrical, Computer and Telecommunica- tions Engineering, University of Wollongong. His research interests include image and signal pro- cessing, neural networks, pattern recognition, and machine learning. He was awarded the University and Faculty Medals, in 2000.
  38. PHILIP B. CHAPPLE received the B.Sc. (Hons.) and Ph.D. degrees in experimental physics (mul- tiphoton laser spectroscopy) from the Australian National University and the Master of Mathemati- cal Sciences degree in signal and information pro- cessing. He is currently the Group Leader Littoral Systems Autonomy, Maritime Division of Defence Science and Technology (DST). His group con- ducts research in maritime autonomous systems to support naval capability development in mine hunting, hydrography, environmental assessment and blue-water naval oper- ations. He has worked for 20 years in the areas of maritime survey, maritime rapid environmental assessment, mine countermeasures, sonar image pro- cessing (automatic target recognition), and maritime autonomous systems. ABDESSELAM BOUZERDOUM (Senior Mem- ber, IEEE) graduated with M.S.E.E. and Ph.D. degrees in electrical engineering from the Uni- versity of Washington, Seattle, USA. He is cur- rently serving as the Head of Information and Computing Technology Division, College of Sci- ence and Engineering, Hamad Bin Khalifa Univer- sity, Qatar, and a Senior Professor of Computer Engineering from the University of Wollongong. His research interests include radar imaging and signal processing, image processing, vision, machine learning, and pattern recognition.
  39. CHRISTIAN H. RITZ (Senior Member, IEEE) received the B.Eng. degree in electrical engineer- ing, and the B.Math. and Ph.D. degrees from the University of Wollongong, in 1998 and 2003, respectively. He is currently a Professor with the School of Electrical, Computer, and Telecommu- nications Engineering. His current research inter- ests include signal and information processing for spatial audio, microphone arrays, sound classifica- tion, and sonar image classification.
  40. LE CHUNG TRAN (Senior Member, IEEE) received the B.E. degree (Hons.) from the Univer- sity of Transport and Communication, Vietnam, in 1997, the M.E. degree from the University of Science and Technology, Vietnam, in 2000, and the Ph.D. degree from the University of Wollongong, Australia, in 2006, all in telecommunications engi- neering. He is currently a Senior Lecturer with the School of Electrical, Computer, and Telecommu- nications Engineering, University of Wollongong. His research interests include 5G, MIMO, space-time-frequency processing, WBANs, the IoT, biomedical engineering, ultra-wideband, millimetre wave, cooperative and cognitive communications, software defined radio, network coding, and digital signal processing for communications.

Cited by

Sorted by citations

Authors
Thanh Le, H., Phung, S. L., Chapple, P. B., Bouzerdoum, A., Ritz, C. H., & Tran, L. C.
Affiliations
University of Wollongong (AU)
ORCID
Venue
IEEE Access
Publisher
Institute of Electrical and Electronics Engineers (IEEE)
Published
2020
Volume
8
Pages
94126-94139
ISSN
2169-3536 (ISSN-L)
DOI
10.1109/access.2020.2995390
OpenAlex
W3028411153
Cited by
49
References
7
Citations by year
2026: 5 · 2025: 9 · 2024: 10 · 2023: 10 · 2022: 10 · 2021: 4 · 2020: 1
FWCI
3.00
Type
journal article
Language
en
Open access
Licence
Locations
  • IEEE Access (publishedVersion, cc-by)
  • Research Online (University of Wollongong) (submittedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
  • Figshare (submittedVersion, cc-by)
Funders
  • NSW Defence Innovation Network and NSW State Government
  • Australian Research Council
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 › Engineering › Ocean Engineering › Underwater Vehicles and Communication Systems
Keywords
sonar images, feature pyramid network, autonomous underwater vehicles, underwater mine detection, end-to-end learning

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