Application of Structure Tensor-Based Image Fusion Method in Marine Exploration
A single sensor is insufficient for effective marine detection. Infrared light and visible light have strong complementarity, and fusing them can generate high-quality images that enable more accurate and comprehensive detection of marine targets. However, existing fusion methods have not been appli...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | zho |
| Published: |
Science Press (China)
2025-02-01
|
| Series: | 水下无人系统学报 |
| Subjects: | |
| Online Access: | https://sxwrxtxb.xml-journal.net/cn/article/doi/10.11993/j.issn.2096-3920.2024-0066 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | A single sensor is insufficient for effective marine detection. Infrared light and visible light have strong complementarity, and fusing them can generate high-quality images that enable more accurate and comprehensive detection of marine targets. However, existing fusion methods have not been applied in marine detection and are not specifically developed for it, leading to poor fusion results. Additionally, there is a lack of deep learning datasets tailored for marine image fusion. To obtain high-quality color fusion images with a prominent performance in detecting targets and obtaining comprehensive information, the deep learning-based image fusion method using structure tensors was optimized based on the characteristics of marine targets. Multi-scale convolution was incorporated, and image fusion was performed according to channels. The collected data were used for comparative simulation experiments, with a variety of evaluation metrics applied. The results indicate that the improved image fusion method outperforms the original algorithm in six metrics, and its overall performance is better than the other ten commonly used image fusion algorithms. Furthermore, its generalization has been validated on other public datasets. The improved structure tensor-based image fusion method has an excellent performance in maritime situational awareness, with fusion results highlighting target features and surpassing the performance of other methods. |
|---|---|
| ISSN: | 2096-3920 |