﻿ 船舶尾流图像的数字化处理和特征描述技术
 舰船科学技术  2022, Vol. 44 Issue (20): 157-160    DOI: 10.3404/j.issn.1672-7649.2022.20.032 PDF

1. 吉林大学，吉林 长春 130015;
2. 长春信息技术职业学院，吉林 长春 130000

Digital processing and feature description technology of ship wake image
XU Ying1,2
1. Jilin University, Changchun 130015, China;
2. Changchun Information Technology College, Changchun 130000, China
Abstract: With the continuous advancement of military technology, the monitoring and tracking of ships at sea has become the central issue of various countries. The traditional sonar monitoring technology is mainly to install sonar equipment on underwater or surface ships to detect underwater sound waves. At present, the submarines of many countries in the world use sound attenuation technology. Monitoring is difficult because underwater sonar devices have difficulty collecting signals. However, boats on the surface and submarines underwater create water waves on the surface with consequences. Currently, countries use lasers, infrared, satellites and other technologies to conduct underwater searches. In this paper, the digital processing technology of ship stern image is studied in detail, focusing on the digital processing in ship stern image processing, and the corresponding feature description technology is adopted.
Key words: ship wake     feature description     digitization
0 引　言

1 船舶尾流形成与变化

 图 1 船舶尾流 Fig. 1 Ship wake

2 船舶尾流图像数字化处理 2.1 尾流图像处理

2.2 灰度变换

 ${{g}} = {{kf}} + {{b}} 。$

2.3 直方图

 图 2 船速为25 km/h尾流图像直方图 Fig. 2 Histogram of wake image with ship speed of 25 km/h
 $P（{rk}）={nk}/N \text{，}$

 图 3 船速为40 km/h尾流图像速度曲线 Fig. 3 Velocity curve of wake image with ship speed of 40 km/h
3 船舶尾流图像数字化与特征描述分割 3.1 Canny检测器梯度算子

 $\nabla f = \left[ {\frac{{{F_x}}}{{{F_y}}}} \right] = \left[ {\begin{array}{*{20}{c}} {\partial f/\partial x} \\ {\partial f/\partial y} \end{array}} \right] \text{。}$

 \begin{aligned} & \nabla F = {\rm{mag}}(\nabla f){\left[ {{F_x}^2 + {F_y}^2} \right]^{\frac{1}{2}}} = {\left[ {\left( {{\raise0.7ex\hbox{{\partial f}$} \mathord{\left/ {\vphantom {{\partial f} {\partial x}}}\right.} \lower0.7ex\hbox{${\partial x}$}}} \right) + {{\left( {{\raise0.7ex\hbox{${\partial f}$} \mathord{\left/ {\vphantom {{\partial f} {\partial y}}}\right.} \lower0.7ex\hbox{${\partial y}}}} \right)}^2}} \right]^{\frac{1}{2}}} \approx \\ & {F_x}^2 + {F_y}^2 \text{。}\end{aligned}

 $\frac{{\partial (\nabla f)}}{{\partial \theta }} = 0 \text{，}$
 $\theta (x,y) = \arctan \left( {\frac{{{F_y}}}{{{F_x}}}} \right) \text{。}$

3.2 船舶尾流图像数字化边缘分割

3.3 船舶尾流特征描述

 ${r_i} = {y_i} - {\hat y_i} \text{，}$
 $S = \sum\limits_{i = 1}^n {{r_i}^2} = {\sum\limits_{i = 1}^n {\left( {{y_i} - {{\hat y}_i}} \right)} ^2} \text{。}$

 $P = polyfit\left( {x,y,m} \right) \text{，}$

 图 4 傅里叶转变图 Fig. 4 Fourier transform diagram

 图 5 直流画尾流边缘分布图 Fig. 5 DC drawing wake edge distribution map
4 结　语

 [1] 万才超. 新媒体时代数字图像技术在木材科学中的应用——评《基于计算机数字图像处理技术木材表面纹理特征提取和分类识别方法》[J]. 科技管理研究, 2021, 41(21): 241. WAN Cai-chao. The application of digital Image technology in wood science in the new media era-Comment on "wood surface texture feature extraction and classification recognition method based on computer digital image processing technology"[J]. Science and Technology Management Research, 2021, 41(21): 241. DOI:10.3969/j.issn.1000-7695.2021.21.001 [2] 张瑞瑞. 基于频域理论的船舶尾流光学检测技术研究[J]. 舰船科学技术, 2021 ,43(22):208-210 ZHANG Rui-rui. Research on ship wake optical detection technology based on frequency domain theory[J]. Ship Science and Technology, 2021,43(22):208-210 [3] 邹宇. 基于图像处理技术的气泡特征提取系统设计[J]. 中国新技术新产品, 2022(7): 25-27. ZOU Yu. Design of bubble feature extraction system based on image processing technology[J]. China New Technology and New Products, 2022(7): 25-27. [4] 宫瑞邦, 程俊文, 王宇晨, 等. 基于频率响应分析技术和图像特征检测方法的变压器绕组故障诊断[J]. 电工电能新技术, 2022,41(1). GONG Rui-bang, CHENG Jun-wen, WANG Yu-chen, et al. Transformer winding fault diagnosis based on frequency response analysis technology and image feature detection method [J]. New Technology in Electrical Engineering, 2022,41(1). [5] 宋博文, 栾春娟, 梁丹妮. 机器学习视域下新兴技术主题识别研究——基于技术特征相似性[J]. 现代情报, 2022, 42(9): 49-57. SONG Bo-wen, LUAN Chun-juan, LIANG Dan-ni. Research on topic recognition of emerging technologies from the perspective of machine learning-based on similarity of technical Features[J]. Modern Intelligence, 2022, 42(9): 49-57. [6] 李萃. 计算机应用于图像处理的相关技术探讨[J]. 电脑编程技巧与维护, 2021(7): 159-160. LI Cui. Discussion on related technologies of computer application in image processing[J]. Computer Programming Skills and Maintenance, 2021(7): 159-160. DOI:10.3969/j.issn.1006-4052.2021.07.065