Weak Edge Target Segmentation Network Based on Dual Attention Mechanism
Source abstract
Abstract
Segmentation of weak edge targets such as glass and plastic poses a challenge in the field of target segmentation. The detection process is susceptible to background interference and various external factors due to the transparent nature of these materials. To address this issue, this paper introduces a segmentation network for weak edge target objects (WETS-Net). To effectively extract edge information of such objects and eliminate redundant information during feature extraction, a dual-attention mechanism is employed, including the Edge Attention Extraction Module (EAEM) and the Multi-Scale Information Fusion Module (MIFM). Specifically, the EAEM combines improved edge feature extraction kernels to selectively enhance the importance of edge features, aiding in more precise target region extraction. The MIFM utilizes spatial attention mechanisms to fuse multi-scale features, reducing background and external interference. These innovations enhance the performance of WETS-Net, offering a new direction for weak edge target segmentation research. Finally, through ablation experiments, the effectiveness of each module is effectively validated. Moreover, the proposed algorithm achieves an average detection accuracy of 95.83% and 96.13% on the dataset and a self-made dataset, respectively, outperforming similar U-Net-improved networks.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Infrared Target Detection Methodologies · Advanced Neural Network Applications · Medical Image Segmentation Techniques
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
School of Automation and Intelligence, Beijing Jiaotong University, Beijing 100044, China
Funding
Natural Science Lateral Project of Beijing Jiaotong University
References (36)
Show source-provided references
1. Visual Semantic Segmentation Based on Few/Zero-Shot Learning: An Overview10.1109/JAS.2023.123207 ↗
2. Clinical feasibility of a myocardial signal intensity threshold-based semi-automated cardiac magnetic resonance segmentation method
3. Inward-region-growing-based accurate partitioning of closely stacked objects for bin-picking10.1088/1361-6501/aba283 ↗
4. Indoor Point Cloud Segmentation Using a Modified Region Growing Algorithm and Accurate Normal Estimation10.1109/ACCESS.2023.3270709 ↗
5. Coal gangue image segmentation method based on edge detection theory of star algorithm10.1080/19392699.2021.2024173 ↗
6. Image segmentation method of mine pass soil and ore based on the fusion of the confidence edge detection algorithm and mean shift algorithm
7. Deep learning techniques for tumor segmentation: A review10.1007/s11227-021-03901-6 ↗
8. Intelligent Image Semantic Segmentation: A Review Through Deep Learning Techniques for Remote Sensing Image Analysis10.1007/s12524-022-01496-w ↗
9. Deep learning based brain tumor segmentation: A survey10.1007/s40747-022-00815-5 ↗
10. Transfer learning in medical image segmentation: New insights from analysis of the dynamics of model parameters and learned representations10.1016/j.artmed.2021.102078 ↗
11. U-Net and Its Variants for Medical Image Segmentation: A Review of Theory and Applications10.1109/ACCESS.2021.3086020 ↗
12. Sariturk, B., and Seker, D.Z. (2022). A Residual-Inception U-Net (RIU-Net) Approach and Comparisons with U-Shaped CNN and Trans-former Models for Building Segmentation from High-Resolution Satellite Images. Sensors, 22.10.3390/s22197624 ↗
13. A real-time efficient object segmentation system based on U-Net using aerial drone images10.1007/s11554-021-01166-z ↗
14. An improved U-Net method for the semantic segmentation of remote sensing images10.1007/s10489-021-02542-9 ↗
15. Ahsan, R., Shahzadi, I., Najeeb, F., and Omer, H. (2024). Brain tumor detection and segmentation using deep learning. Magn. Reson. Mater. Phys. Biol. Med., 1–10.10.1007/s10334-024-01203-5 ↗
16. Double U-Nets for Image Segmentation by Integrating the Region and Boundary Information10.1109/ACCESS.2021.3075294 ↗
17. An attention-based U-Net for detecting deforestation within satellite sensor imagery10.1016/j.jag.2022.102685 ↗
18. An Attention-Based Improved U-Net Neural Network Model for Semantic Segmentation of Moving Objects10.1109/ACCESS.2024.3391249 ↗
19. Progressive Glass Segmentation10.1109/TIP.2022.3162709 ↗
20. DAGNet: Depth-aware Glass-like objects segmentation via cross-modal attention10.1016/j.jvcir.2024.104121 ↗
21. Hu, X., Gao, R., Yang, S., and Cho, K. (2023). TGSNet: Multi-Field Feature Fusion for Glass Region Segmentation Using Transformers. Mathematics, 11.10.3390/math11040843 ↗
22. UAV imagery-based railroad station building inspection using hybrid learning architecture10.1088/1361-6501/ad4ab5 ↗
23. Ronneberger, O., Fischer, P., and Brox, T. (2015, January 5–9). U-Net: Convolutional Networks for Biomedical Image Segmentation. Proceedings of the 18th International Conference, Munich, Germany.10.1007/978-3-319-24574-4_28 ↗
24. Jumutc, V., Bļizņuks, D., and Lihachev, A. (2022). Multi-Path U-Net Architecture for Cell and Colony-Forming Unit Image Segmentation. Sensors, 22.10.3390/s22030990 ↗
25. ELU-Net: An Efficient and Lightweight U-Net for Medical Image Segmentation10.1109/ACCESS.2022.3163711 ↗
26. A Densely Connected Network Based on U-Net for Medical Image Segmentation
27. Three-dimensional deep neural network for automatic delineation of cervical cancer in planning computed tomography images10.1002/acm2.13566 ↗
28. Inner Cascaded U2-Net: An Improvement to Plain Cascaded U-Net
29. Wang, D., Xiang, S., Zhou, Y., Mu, J., Zhou, H., and Irampaye, R. (2022). Multiple-Attention Mechanism Network for Semantic Segmentation. Sensors, 22.10.3390/s22124477 ↗
30. Dual Adversarial Attention Mechanism for Unsuper-vised Domain Adaptive Medical Image Segmentation10.1109/TMI.2022.3186698 ↗
31. Duan, S., Zhao, J., Huang, X., and Zhao, S. (2024). Semantic Segmentation of Remote Sensing Data Based on Channel Attention and Feature Information Entropy. Sensors, 24.10.3390/s24041324 ↗
32. SeNet: Structured Edge Network for Sea–Land Segmentation10.1109/LGRS.2016.2637439 ↗
33. Swin-CDSA: The Semantic Segmentation of Remote Sensing Images Based on Cascaded Depthwise Convolution and Spatial Attention Mechanism10.1109/LGRS.2024.3431638 ↗
34. Research on Image Semantic Segmentation Based on Hybrid Cascade Feature Fusion and Detailed Attention Mechanism10.1109/ACCESS.2024.3395541 ↗
35. Attention based multi-scale nested network for biomedical image segmentation10.1016/j.heliyon.2024.e33892 ↗
36. GLSANet: Global-Local Self-Attention Network for Remote Sensing Image Semantic Segmentation10.1109/LGRS.2023.3235117 ↗