﻿ 基于图像处理的船舶舱室设计缺陷检测系统
 舰船科学技术  2023, Vol. 45 Issue (12): 148-151    DOI: 10.3404/j.issn.1672-7619.2023.12.029 PDF

Design defect detection system of ship cabin based on image processing
SONG Xiao-jing
Henan Provincial Research Center of Wisdom Education and Intelligent Technology Application Engineering Technology, Zhengzhou 450000, China
Abstract: The detection of cabin design defects is a necessary step in the ship manufacturing process. The traditional cabin design defect detection relies on the experience of inspectors and the accuracy of inspection equipment, and the detection efficiency is low, and the large size of ships has put forward new requirements for design defect detection. Based on the full study of image processing technology, this paper proposes a naval cabin design defect detection system, designs the overall structure of the system, and tests the system. The test results show that the system can detect the basic structural parameters of the cabin, and at the same time can automatically identify and determine the wall cracks, which greatly improves the efficiency and accuracy of cabin design defect detection.
Key words: Image processing     image segmentation     defect detection     ship
0 引　言

1 图像处理技术

1.1 直方图均衡

 ${P_r}({r_k}) = \frac{{{N_k}}}{N} 。$ (1)

 ${M}_{k}={\displaystyle \sum _{i=0}^{k}{P}_{\text{r}}}({r}_{\text{i}}) ，$ (2)

 图 1 处理前后灰度直方图对比 Fig. 1 Comparison of gray histogram before and after processing

 图 2 直方图均衡图片处理效果对比 Fig. 2 Histogram equalization image processing effect comparison
1.2 图像分割

 图 3 拉普拉斯算子图像分割原理 Fig. 3 Laplace operator image segmentation principle

1）若中心像素点为负数

 $g(x,y) = f(x,y) - {\nabla ^2}f(x,y),$ (3)

2）若中心像素点为正数

 $g(x,y) = f(x,y) + {\nabla ^2}f(x,y) 。$ (4)

2 船舶舱室设计缺陷检测系统设计 2.1 系统整体结构设计

1）图像采集设备

2）数字化设备

3）存储设备

4）图像预处理

5）图像均衡和分割

6）缺陷检测

 图 4 系统结构框图 Fig. 4 System structure block diagram
2.2 系统测试

1）舱室设计长宽高的检测实现

2）舱室墙壁检测

3 结　语

1）分析船舶舱室设计缺陷检测系统中涉及到的图像处理关键技术，包括直方图均衡和图像分割技术，并对处理前后的图像进行效果对比；

2）对系统的整体结构进行设计，并对系统进行测试。测试结果表明，系统可以实现对舱室结构参数的基本检测，同时对墙壁裂缝做到自动识别和判定，发现一些不明显的裂缝，大幅度提升舱室设计缺陷检测的工作效率和准确率。本文建立的船舶舱室设计缺陷检测系统可以应用于各类船舶舱室，同时通过不断积累数据，未来可以结合云计算和神经网络系统来实现对舱室设计缺陷的自动智能检测。

 [1] 赵崇林, 朱江, 胡永进, 等. 融合注意力和多尺度特征的航空发动机缺陷检测[J]. 北京航空航天大学学报: 1–14. ZHAO Chong-lin, ZHU Jiang, HU Yong-jin, et al. Fusion of attention and multi-scale features for aero-engine defect detection [J]. Journal of Beijing University of Aeronautics and Astronsutics : 1–14. [2] 郭慧文, 刘晓鸣, 陈浩天, 等. 基于机器视觉的镍基高温合金表面缺陷检测系统开发[J]. 制造业自动化, 2023, 45(4): 81-87. GUO Hui-wen, LIU Xiao-ming, CHEN Hao-tian, et al. Development of Surface Defect detection System for Nickel-based Superalloy based on Machine vision[J]. Manufacturing Automation, 2023, 45(4): 81-87. DOI:10.3969/j.issn.1009-0134.2023.04.018 [3] 杨硕, 孙林. 基于图像处理和深度学习的膜片缺陷检测[J]. 激光杂志, 2023, 44(4): 103-106. YANG Shuo, SUN Lin. Patch defect detection based on image processing and deep learning[J]. Laser Journal, 2023, 44(4): 103-106. DOI:10.14016/j.cnki.jgzz.2023.04.103 [4] 王丹, 高炜欣. 基于稀疏描述的X射线焊缝图像缺陷检测研究[J]. 制造技术与机床, 2023(4): 126-131. WANG Dan, GAO Wei-xin. Research on Defect Detection of X-ray Weld Image Based on Sparse Description[J]. Manufacturing Technology & Machine Tool, 2023(4): 126-131. DOI:10.19287/j.mtmt.1005-2402.2023.04.020 [5] 刘杰. 探究钢铁表面缺陷检测的研究与实现[J]. 中国设备工程, 2023(6): 172-174. LIU Jie. Research and Implementation of Steel Surface Defect Detection[J]. China Equipment Engineering, 2023(6): 172-174. [6] 郭庆. 基于图像处理的铁轨表面缺陷检测研究[D]. 南京: 南京邮电大学, 2022. [7] 张东岭. 基于图像处理的机械材料缺陷检测技术研究[J]. 粘接, 2002, 49(6): 39-43. ZHANG Dong-ling. Research on Defect Detection Technology of Mechanical Materials Based on Image Processing[J]. Journal of Bonding, 2002, 49(6): 39-43. [8] 王昱东. 基于3D技术的舰船舱室设计缺陷可视化方法研究[J]. 舰船科学技术, 2019, 41(24): 13-15. WANG Yu-dong. Research on Visualization Method of Ship Cabin Design Defects Based on 3D Technology[J]. Ship Science and Technology, 2019, 41(24): 13-15.