journal article·2024-12-20Open access

Automating an Encoder–Decoder Incorporated Ensemble Model: Semantic Segmentation Workflow on Low-Contrast Underwater Images

Jale Bektaş

Applied Sciences· Volume 14· Issue 24

Abstract

Numerous methods have been proposed for semantic segmentation and the state-of-the-art part is likely to be incorporated by deep learning-based methods which show a salient performance. This study addresses the challenge of semantic segmentation in low-contrast imbalanced underwater images. Moreover, it employs nine model fusions as a downstream workflow task using encoder–decoder architectures with Dice Loss and Focal Loss training focusing on the imbalance data. Afterwards, the most effective two encoder–decoder fusion models, Res34+Unet and VGG19+FPN, by 0.592%, 0.590% mIoU on average and by 0.510%, 0.491% F1-score yielded better performance, respectively, than other models. Using a weight-optimization algorithm, the ensemble model with recreated IoU results improves the accuracy for both the Res34+Unet and the VGG19+FPN models, by 0.652% mIoU on average which is 6%. The ensemble model combines the model performances of independent models by considering their superior inference accuracy on a per-class basis separately and improves the model performances by emphasizing the better one on a per-class basis.
Source: crossref · Quality: baseline

Research context

OpenAlex topics: Image Enhancement Techniques · Advanced Neural Network Applications · Domain Adaptation and Few-Shot Learning

SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.

Affiliations

Department of Computer Engineering, Mersin University, 33110 Mersin, Türkiye

References (35)

Show source-provided references
1. HookNet: Multi-resolution convolutional neural networks for semantic segmentation in histopathology whole-slide images10.1016/j.media.2020.101890 ↗
2. Schutera, M., Rettenberger, L., Pylatiuk, C., and Reischl, M. (2022). Methods for the frugal labeler: Multi-class semantic segmentation on heterogeneous labels. PLoS ONE, 17.10.1371/journal.pone.0263656 ↗
3. Coordinate-Aware Mask R-CNN with Group Normalization: A underwater marine animal instance segmentation framework10.1016/j.neucom.2024.127488 ↗
4. Underwater image enhancement using contrast limited adaptive histogram equalization and layered difference representation10.1007/s11042-020-10426-2 ↗
5. Md Jahidul, I., Chelsey, E., Yuyang, X., Peigen, L., Muntaqim, M., Christopher, M., Sadman Sakib, E., and Junaed, S. (2020, January 25–29). Semantic Segmentation of Underwater Imagery: Dataset and Benchmark. Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Las Vegas, NV, USA.
6. A semantic segmentation algorithm for fashion images based on modified mask RCNN10.1007/s11042-023-14958-1 ↗
7. Tackling background ambiguities in multi-class few-shot point cloud semantic segmentation10.1016/j.knosys.2022.109508 ↗
8. Effect of label noise on multi-class semantic segmentation: A case study on Bangladesh marine region10.1080/08839514.2022.2039348 ↗
9. Approaches to training multiclass semantic image segmentation of damage in concrete10.1111/jmi.12906 ↗
10. Semantic segmentation and detection of satellite objects using U-Net model of deep learning10.1007/s11042-022-12892-2 ↗
11. Efficient Cloud Removal Network for Satellite Images Using SAR-optical Image Fusion
12. Nunes, A., and Matos, A. (2023). Improving Semantic Segmentation Performance in Underwater Images. J. Mar. Sci. Eng., 11.10.20944/preprints202310.1767.v1 ↗
13. Semantic segmentation of breast cancer images using DenseNet with proposed PSPNet10.1007/s11042-023-17411-5 ↗
14. Enhancing Underwater Image Segmentation: A Semantic Approach to Segment Objects in Challenging Aquatic Environment10.1016/j.procs.2024.04.036 ↗
15. DatUS: Data-driven Unsupervised Semantic Segmentation with Pre-trained Self-supervised Vision Transformer10.1109/TCDS.2024.3383952 ↗
16. Semantic segmentation of fish and underwater environments using deep convolutional neural networks and learned active contours10.1109/ACCESS.2023.3262649 ↗
17. DIResUNet: Architecture for multiclass semantic segmentation of high resolution remote sensing imagery data10.1007/s10489-022-03310-z ↗
18. Pergeorelis, M., Bazik, M., Saponaro, P., Kim, J., and Kambhamettu, C. (2022, January 17–20). Synthetic data for semantic segmentation in underwater imagery. Proceedings of the OCEANS 2022, Hampton Roads, VA, USA.10.1109/OCEANS47191.2022.9976962 ↗
19. Kabir, I., Shaurya, S., Maigur, V., Thakurdesai, N., Latnekar, M., Raunak, M., and Reza, M.A. (2023, January 1–5). Few-Shot Segmentation and Semantic Segmentation for Underwater Imagery. Proceedings of the 2023 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Detroit, MI, USA.10.1109/IROS55552.2023.10342227 ↗
20. DrsNet: Dual-resolution semantic segmentation with rare class-oriented superpixel prior10.1007/s11042-020-09691-y ↗
21. Recent progress in semantic image segmentation10.1007/s10462-018-9641-3 ↗
22. UISS-Net: Underwater Image Semantic Segmentation Network for improving boundary segmentation accuracy of underwater images10.1007/s10499-024-01439-x ↗
23. Semantic segmentation using Vision Transformers: A survey10.1016/j.engappai.2023.106669 ↗
24. An efficient approach to detect and segment underwater images using Swin Transformer10.1016/j.rineng.2024.102460 ↗
25. Ds-transunet: Dual swin transformer u-net for medical image segmentation
26. Sun YBi, F., Gao, Y., Chen, L., and Feng, S. (2022). A multi-attention UNet for semantic segmentation in remote sensing images. Symmetry, 14.10.3390/sym14050906 ↗
27. Attention-augmented U-Net (AA-U-Net) for semantic segmentation10.1007/s11760-022-02302-3 ↗
28. Simonyan, K., and Zisserman, A. (2014). Very deep convolutional networks for large-scale image recognition. arXiv.
29. He, K., Zhang, X., Ren, S., and Sun, J. (2016, January 27–30). Deep residual learning for image recognition. Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, NV, USA.10.1109/CVPR.2016.90 ↗
30. Chiu, Y.C., Tsai, C.Y., Ruan, M.D., Shen, G.Y., and Lee, T.T. (September, January 31). Mobilenet-SSDv2: An improved object detection model for embedded systems. Proceedings of the 2020 International Conference on System Science and Engineering (ICSSE), Kagawa, Japan.10.1109/ICSSE50014.2020.9219319 ↗
31. Ronneberger, O., Fischer, P., and Brox, T. (2015, January 5–9). U-net: Convolutional networks for biomedical image segmentation. Proceedings of the International Conference on Medical Image Computing and Computer-Assisted Intervention, Munich, Germany.10.1007/978-3-319-24574-4_28 ↗
32. Chaurasia, A., and Culurciello, E. (2017, January 10–13). LinkNet: Exploiting The encoder Representations for Efficient Semantic Segmentation. Proceedings of the IEEE Visual Communications and Image Processing (VCIP), St. Petersburg, FL, USA.10.1109/VCIP.2017.8305148 ↗
33. Jin, L., and Liu, G. (2021). An approach on image processing of deep learning based on improved SSD. Symmetry, 13.10.3390/sym13030495 ↗
34. Zhao, R., Buyue, Q., Xianli, Z., Yang, L., Rong, W., Yang, L., and Yinggang, P. (2020, January 20). Rethinking dice loss for medical image segmentation. Proceedings of the 2020 IEEE International Conference on Data Mining (ICDM), Sorrento, Italy.10.1109/ICDM50108.2020.00094 ↗
35. Kingma, D.P., and Ba, J. (2020). Adam: A method for stochastic optimization. arXiv.