Skip to main content
SpringerLink
Log in

Different treatments of pixels in unlabeled images for semi- supervised sonar image segmentation

  • Original Article
  • Published:
International Journal of Machine Learning and Cybernetics Aims and scope Submit manuscript

A Correction to this article was published on 17 January 2024

This article has been updated

Abstract

Pseudo-labeling is an effective semi-supervised segmentation method. Most pseudo-labeling works were based on a common assumption: lower entropy means lower uncertainty. Hence many high entropy pseudo-labels are discarded and not involved in the training. Inadequate labeled data limit performance of model segmentation. In order to expand the labeled data capacity, we propose a new semi-supervised segmentation method, namely, different treatments of pixels in unlabeled images (DTP). Our DTP consists of three main components: labeled images segmentation, certain pixels segmentation, uncertain pixels segmentation. A sonar image with two different parallel segmentation networks will produce two one-hot segmentation maps. If the predictions of a pixel in the sonar image are consistent on two one-hot segmentation maps, this pixel is regarded as reliable in this segmentation and delineated as the certain pixel. On the contrary, if the prediction results are different, the pixel is delineated as the uncertain pixel. Then uncertain pixels are necessary to choose an advanced semi-supervised framework for label assignment to minimize the possible error propagation. Meanwhile, certain pixels are assessed in two extreme ways—radical segmentation and conservative segmentation. Compared with other methods, our method is novel in (1) indicating certain/uncertain pixels to expand the labeled data capacity (2) introducing other advanced semi-supervised methods for segmenting uncertain pixels to improve the segmentation performance. Experimental results show that our method achieves advanced semi-supervised segmentation performance in sonar dataset.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5

Similar content being viewed by others

Change history

References

  1. Chuang MC, Hwang JN, Ye JH, Huang SC, Williams K (2017) Underwater fish tracking for moving cameras based on deformable multiple kernels. IEEE Trans Syst Man Cybern Syst 47(9):2467–2477

    Google Scholar 

  2. Karoui I, Quidu I, Legris M (2015) Automatic sea-surface obstacle detection and tracking in forward-looking sonar image sequences. IEEE Trans Geosci Remote Sens 53(11):6315–6315

    Article  Google Scholar 

  3. Renga A et al (2014) SAR bathymetry in the Tyrrhenian Sea by COSMO-SkyMed data: a novel approach. IEEE J Sel Top Appl Earth Obs Remote Sens 7(7):2834–2847

    Article  Google Scholar 

  4. Kostylev VE, Courtney RC, Robert G, Todd BJ (2003) Stock evaluation of giant scallop (Placopecten magellanicus) using high-resolution acoustics for seabed mapping. Fish Res 60(2–3):479–492

    Article  Google Scholar 

  5. Brice G (1996) Salvage and the underwater cultural heritage. Mar Policy 20(4):337–342

    Article  Google Scholar 

  6. Boulinguez D, Quinquis A (2002) 3-D underwater object recognition. IEEE J Ocean Eng 27(4):814–829

    Article  Google Scholar 

  7. Kong WZ et al (2020) YOLOv3-DPFIN: a dual-path feature fusion neural network for robust real-time sonar target detection. IEEE Sens J 20(7):3745–3756

    Article  Google Scholar 

  8. Bhosale YH, Patnaik KS (2022) Application of deep learning techniques in diagnosis of Covid-19 (Coronavirus): a systematic review. Neural Process Lett

  9. Bhosale YH, Patnaik KS (2023) PulDi-COVID: chronic obstructive pulmonary (lung) diseases with COVID-19 classification using ensemble deep convolutional neural network from chest X-ray images to minimize severity and mortality rates,". Biomed Signal Process Control 81:104445

    Article  Google Scholar 

  10. Krizhevsky A, Sutskever I, Hinton GE (2017) ImageNet classification with deep convolutional neural networks. Commun ACM 60(6):84–90

    Article  Google Scholar 

  11. Simonyan K, Zisserman A (2015) Very deep convolutional networks for large-scale image recognition. arXiv preprint

  12. Szegedy C et al (2015) Going deeper with convolutions. In: Presented at the 2015 IEEE conference on computer vision and pattern recognition (CVPR)

  13. Ronneberger O, Fischer P, Brox T (2015) U-Net: convolutional networks for biomedical image segmentation. In: Presented at the medical image computing and computer-assisted intervention, PT III

  14. He KM, Zhang XY, Ren SQ, Sun J, IEEE (2016) Deep residual learning for image recognition. In: Presented at the 2016 IEEE conference on computer vision and pattern recognition (CVPR)

  15. Chen LC, Papandreou G, Kokkinos I, Murphy K, Yuille AL (2018) DeepLab: semantic image segmentation with deep convolutional nets, atrous convolution, and fully connected CRFs. IEEE Trans Pattern Anal Mach Intell 40(4):834–848

    Article  Google Scholar 

  16. Badrinarayanan V, Kendall A, Cipolla R (2017) SegNet: a deep convolutional encoder-decoder architecture for image segmentation. IEEE Trans Pattern Anal Mach Intell 39(12):2481–2495

    Article  Google Scholar 

  17. Sarker MMK et al (2018) SLSDeep: skin lesion segmentation based on dilated residual and pyramid pooling networks. In: Presented at the medical image computing and computer assisted intervention—MICCAI 2018, PT II

  18. Milletari F, Navab N, Ahmadi SA, IEEE (2016) V-Net: fully convolutional neural networks for volumetric medical image segmentation. In: Presented at the proceedings of 2016 fourth international conference on 3D vision (3DV)

  19. Isensee F, Jaeger PF, Kohl SAA, Petersen J, Maier-Hein KH (2021) nnU-Net: a self-configuring method for deep learning-based biomedical image segmentation. Nat Methods 18(2):203

    Article  Google Scholar 

  20. Banan A, Nasiri A, Taheri-Garavand A (2020) Deep learning-based appearance features extraction for automated carp species identification. Aquac Eng 89:102053

    Article  Google Scholar 

  21. Wang WC, Du YJ, Chau KW, Xu DM, Liu CJ, Ma Q (2021) An ensemble hybrid forecasting model for annual runoff based on sample entropy, secondary decomposition, and long short-term memory neural network. Water Resour Manag 35(14):4695–4726

    Article  Google Scholar 

  22. Fan YJ, Xu KK, Wu H, Zheng Y, Tao B (2020) Spatiotemporal modeling for nonlinear distributed thermal processes based on KL decomposition, MLP and LSTM network. IEEE Access 8:25111–25121

    Article  Google Scholar 

  23. Chen WB, Sharifrazi D, Liang GX, Band SS, Chau KW, Mosavi A (2022) Accurate discharge coefficient prediction of streamlined weirs by coupling linear regression and deep convolutional gated recurrent unit. Eng Appl Comput Fluid Mech 16(1):965–976

    Google Scholar 

  24. Afan HA et al (2021) Modeling the fluctuations of groundwater level by employing ensemble deep learning techniques. Eng Appl Comput Fluid Mech 15(1):1420–1439

    Google Scholar 

  25. Chen CC et al (2022) Forecast of rainfall distribution based on fixed sliding window long short-term memory. Eng Appl Comput Fluid Mech 16(1):248–261

    Google Scholar 

  26. Zhu PP, Isaacs J, Fu B, Ferrari S, IEEE (2017) Deep learning feature extraction for target recognition and classification in underwater sonar images. In: Presented at the 2017 IEEE 56th annual conference on decision and control (CDC)

  27. Valdenegro-Toro M, IEEE (2017) Best practices in convolutional networks for forward-looking sonar image recognition. In: Presented at the OCEANS 2017—Aberdeen

  28. Kim J, Cho H, Pyo J, Kim B, Yu SC, IEEE (2016) The convolution neural network based agent vehicle detection using forward-looking sonar image. In: Presented at the oceans 2016 MTS/IEEE Monterey

  29. Song Y, He B, Liu P (2021) Real-time object detection for AUVs using self-cascaded convolutional neural networks. IEEE J Ocean Eng 46(1):56–67

    Article  Google Scholar 

  30. Dzieciuch I, Gebhardt D, Barngrover C, Parikh K (2017) Non-linear convolutional neural network for automatic detection of mine-like objects in sonar imagery. In: Presented at the proceedings of the 4th international conference on applications in nonlinear dynamics (ICAND 2016)

  31. Song Y et al (2017) Side scan sonar segmentation using deep convolutional neural network. In: Presented at the Oceans 2017—Anchorage

  32. Wang Z, Guo JX, Huang WZ, Zhang SW (2022) Side-scan sonar image segmentation based on multi-channel fusion convolution neural networks. IEEE Sens J 22(6):5911–5928

    Article  Google Scholar 

  33. Cao WP, Gao JZ, Ming Z, Cai SB, Shan ZG (2018) Fuzziness-based online sequential extreme learning machine for classification problems. Soft Comput 22(11):3487–3494

    Article  Google Scholar 

  34. Cao WP et al (2021) An ensemble fuzziness-based online sequential learning approach and its application. In: Presented at the knowledge science, engineering and management, PT I

  35. Cao WP, Yang PF, Ming Z, Cai SB, Zhang JY, IEEE (2020) An improved fuzziness based random vector functional link network for liver disease detection. Presented at the 2020 IEEE 6th int conference on big data security on cloud (bigdatasecurity)/6th IEEE int conference on high performance and smart computing, (HPSC)/5th IEEE int conference on intelligent data and security (IDS)

  36. Laine S, Aila T (2017) Temporal ensembling for semi-supervised learning. arXiv preprint

  37. Tarvainen A, Valpola H (2017) Mean teachers are better role models: Weight-averaged consistency targets improve semi-supervised deep learning results. Presented at the advances in neural information processing systems 30 (NIPS 2017)

  38. Verma V et al (2022) Interpolation consistency training for semi-supervised learning. Neural Netw 145:90–106

    Article  Google Scholar 

  39. Li X, Yu L, Chen H, Fu C-W, Xing L, Heng P-A (2020) Transformation consistent self-ensembling model for semi-supervised medical image segmentation. arXiv preprint

  40. Yu LQ, Wang SJ, Li XM, Fu CW, Heng PA (2019) Uncertainty-aware self-ensembling model for semi-supervised 3D left atrium segmentation. In: Presented at the medical image computing and computer assisted intervention—Miccai 2019, PT II

  41. Shi YH et al (2022) Inconsistency-aware uncertainty estimation for semi-supervised medical image segmentation. IEEE Trans Med Imaging 41(3):608–620

    Article  Google Scholar 

  42. Zhang Y, Yang L, Chen J, Fredericksen M, Hughes DP, Chen DZ (2017) Deep adversarial networks for biomedical image segmentation utilizing unannotated images. In: Presented at the international conference on medical image computing and computer-assisted intervention

  43. Wu YC, Xu MF, Ge ZY, Cai JF, Zhang L (2021) Semi-supervised left atrium segmentation with mutual consistency training. In: Presented at the medical image computing and computer assisted intervention—Miccai 2021, PT II

  44. Sajjadi M, Javanmardi M, Tasdizen T, IEEE (2016) Mutual exclusivity loss for semi-supervised deep learning. In: Presented at the 2016 IEEE international conference on image processing (ICIP)

  45. Cubuk ED, Zoph B, Mane D, Vasudevan V, Le QV, Soc IC (2019) AutoAugment: learning augmentation strategies from data. In: Presented at the 2019 IEEE/CVF conference on computer vision and pattern recognition (CVPR 2019)

  46. Cubuk ED, Zoph B, Shlens J, Le QV, Ieee Comp SOC (2020) Randaugment: Practical automated data augmentation with a reduced search space. In: Presented at the 2020 IEEE/CVF conference on computer vision and pattern recognition workshops (CVPRW 2020)

  47. Yun S et al (2019) CutMix: regularization strategy to train strong classifiers with localizable features. In: Presented at the 2019 IEEE/CVF international conference on computer vision (ICCV 2019)

  48. Berthelot D, Carlini N, Goodfellow I, Oliver A, Papernot N, Raffel C (2019) MixMatch: a holistic approach to semi-supervised learning. In: Presented at the advances in neural information processing systems 32 (NIPS 2019)

  49. Zheng H et al (2019) Semi-supervised segmentation of liver using adversarial learning with deep atlas prior. In: Presented at the medical image computing and computer assisted intervention—MICCAI 2019, PT VI

Download references

Author information

Authors and Affiliations

  1. School of Marine Electrical Engineering, Dalian Maritime University, Dalian, China

    Huipu Xu, Pengfei Tong & Yongzhi Li

Authors
  1. Huipu Xu
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Pengfei Tong
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Yongzhi Li
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Huipu Xu.

Ethics declarations

Conflict of interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Xu, H., Tong, P. & Li, Y. Different treatments of pixels in unlabeled images for semi- supervised sonar image segmentation. Int. J. Mach. Learn. & Cyber. 15, 637–646 (2024). https://doi.org/10.1007/s13042-023-01930-6

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s13042-023-01930-6

Keywords

Search

Navigation