教授
发布时间：2022-10-13
点击次数：
| 影响因子： | 1.813 |
| DOI码： | 10.1007/s11265-022-01816-w |
| 所属单位： | 长江大学计算机科学学院 |
| 发表刊物： | Journal of Signal Processing Systems |
| 刊物所在地： | Cham, Switzerland |
| 项目来源： | 国家自然科学基金（62276032）；中国高校产学研创新基金（2020ITA03012） |
| 关键字： | Deep learning · Semantic segmentation · Coordinate Attention · Draft mark detection · Waterline extraction · Image processing |
| 摘要： | Ship draft reading is an essential link to the draft survey. At present, manual observation is primarily used to determine a ship’s draft. However, manual observation is easily affected by complex situations such as Large waves on the water, Water obstacles, Water traces, Tilted draft characters, and Rusted draft characters. Traditional image-based methods of ship draft reading are difficult to adapt to these complex situations, and existing deep learning-based methods have disadvantages such as the poor robustness of ship draft reading in various complex situations. In this paper, we proposed a method that combines image processing and deep learning and is capable of adapting to a variety of complex situations, particularly in the presence of Large waves on the water and Water obstacles. We also propose a small U2-NetP neural network for semantic segmentation that incorporates Coordinate attention, hence enhancing the capture of information regarding spatial locations. Furthermore, its segmentation accuracy reached 96.47% compared with the original network. In addition, in consideration of the combination of lightweight and multitasking of the method, we use the lightweight Yolov5n network architecture to detect the ship draft characters, which achieves 98% of mAP_0.5 and effectively improves the lightweight of the draft reading. Experimental results on a real dataset encompassing many difficult situations illustrate the state-of-the-art performance of the suggested reading approach when compared to other existing deep learning methods. The average inaccuracy of the draft reading is less than ±0.005 m, and millimeter-level precision is achievable. It can serve as a valuable resource for manual reading. In addition, our work lays the groundwork for future research on the deployment of edge devices. |
| 合写作者： | Tao Han, Peiwen Wang, Hu Liu, Mengyuan Xiong, Shengbing Hong |
| 第一作者： | Weihao Li |
| 论文类型： | 期刊论文 |
| 通讯作者： | Wei Zhan |
| 论文编号： | 20260830-011 |
| 学科门类： | 工学 |
| 一级学科： | 计算机科学与技术 |
| 文献类型： | 期刊 |
| 卷号： | 95 |
| 期号： | 2-3 |
| 页面范围： | 177-195 |
| 字数： | 7499 |
| ISSN号： | 1939-8018; 1939-8115 |
| 发表时间： | 2023-03-01 |
