为提高舰船在航行过程中对周围环境的检测效果，保证舰船的安全航行，本文基于计算机视觉传达技术，提出舰船航行图形图像分析处理方法。经过图像亮度调整、色彩均衡、融合等过程，处理后的图像得到了明显改善。同时，对传统的FCTF-Net算法进行了改进，得到双通道双阶段图像去雨网络（DTDNet），在此基础上引入感知函数和TV Loss函数，实现对图像的去雾、去雨处理，并与其他几种算法进行了对比实验。实验结果表明，改进后的算法在信噪比和图像相似度上都具有明显的优势。
In order to improve the ship's detection effect on the surrounding environment during navigation and ensure the ship's safe navigation, this paper proposes an analysis and processing method of ship's navigation graphics and images based on computer vision communication technology. After image brightness adjustment, color balance, fusion and other processes, the processed images have been significantly improved. At the same time, the traditional FCTF-Net algorithm is improved to get a two-channel two-stage image rain removal network (DTDNet). On this basis, the perception function and TV Loss function are introduced to realize the image fog removal and rain removal processing, and the comparison experiment is conducted with other algorithms. The experimental results show that the improved algorithm has obvious advantages in signal-to-noise ratio and image similarity.
2023,45(23): 194-197 收稿日期：2023-04-20
DOI：10.3404/j.issn.1672-7649.2023.23.037
分类号：TP391
作者简介：冷桂丽(1984-),女,硕士,讲师,研究方向为艺术设计技术及艺术学
参考文献：
[1] 张哲卿, 朱志宇, 魏莱, 等. 复杂海面背景下船舶红外偏振图像融合方法[J]. 电光与控制, 2023, 30(7): 68-72.
[2] 段仕浩. 基于机器视觉技术的船舶航行危险区域自动识别方法[J]. 舰船科学技术, 2023, 45(3): 157-160.
[3] 吴勇, 初秀民, 刘兴龙, 田国昊. 船舶AIS与视频图像信息融合方法研究[J]. 武汉理工大学学报(交通科学与工程版), 2023, 47(3): 575-581.
[4] 解宇虹, 谢源, 陈亮, 李翠华, 曲延云. 真实有雾场景下的目标检测[J]. 计算机辅助设计与图形学学报, 2021, 33(5): 733-745.
[5] 王道累, 张天宇. 图像去雾算法的综述及分析[J]. 图学学报, 2020, 41(6): 861-870.
