Abstract
In recent years, the recognition technology for ship targets has become a research hotspot. Ship targets have a variety of detection forms, where optical imaging and radar imaging are two important means. However, the quality of acquired multi-source images can be easily affected by external conditions and their own sensor limitations in practice. It is worth considering how to simultaneously and fully utilize the characteristics of optical and radar images to improve the recognition rate of ships. To solve these problems, this paper proposes a new radar-to-optical image conversion and fusion algorithm for ships based on U-GAT-IT by utilizing the strong correlation between ISAR images and optical images. The proposed algorithm obtains a higher-quality target-generated image, which effectively improves the ship recognition rate. Experimental results verify the effectiveness of the proposed algorithm.
| Original language | English |
|---|---|
| Pages (from-to) | 7808-7812 |
| Number of pages | 5 |
| Journal | International Geoscience and Remote Sensing Symposium (IGARSS) |
| DOIs | |
| State | Published - 2025 |
| Externally published | Yes |
| Event | 2025 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2025 - Brisbane, Australia Duration: 3 Aug 2025 → 8 Aug 2025 |
Keywords
- Multi-source image fusion
- Radar images
- Ship target recognition
- U-GAT-IT
Fingerprint
Dive into the research topics of 'A NOVEL SHIP RECOGNITION METHOD BASED ON MULTI-SOURCE IMAGE FUSION'. Together they form a unique fingerprint.Cite this
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver