卷积神经网络（CNN）作为智能图像视觉中一个深层神经网络，能够对船舱内部进行特征提取，本文基于卷积神经网络的智能图像视觉，对船舱内部多角度监控方法展开研究，解决人工巡检船舱内部周期较长的问题。同时实时对船舱的运行状态进行监控，实现全天候对船舱内部设备信息的感知和实时预警，加强船舱相关设备的监控工作，提升船舶的使用寿命及工作效率。
Convolutional neural network (CNN), as a deep neural network in intelligent image vision, can extract features from the interior of the cabin. This paper will introduce the intelligent image vision based on convolutional neural network. The multi-angle monitoring method of the cabin interior is studied to solve the problem of a long period of manual inspection of the cabin interior. At the same time, it can also monitor the operation status of the cabin in real time, realize all-weather perception and real-time warning of equipment information inside the cabin, strengthen the monitoring work of related equipment in the cabin, and improve the service life and work efficiency of the ship.
2023,45(15): 127-130 收稿日期：2023-03-16
DOI：10.3404/j.issn.1672-7649.2023.15.025
分类号：U667.65
作者简介：李谷雨(1982-),女,硕士,讲师,研究方向为视觉图形图像、计算机三维设计及建筑装饰设计等
参考文献：
[1] 万梓然, 刘斌, 赵岩, 等. 融合图像与环境参数的农作物生长远程监控平台[J]. 物联网技术, 2023, 13(6): 3–7+10
[2] 张贻飞, 何勇, 李莹莹, 等. 基于风光储的输电线路远程图像监控装置融冰系统研究[J]. 电动工具, 2023, (2): 26-30+33.
[3] 薛迪杰. 基于无线网络的船舶机舱自动化监控系统[J]. 舰船科学技术, 2022, 44(22): 166–169
XUE Di-jie. Wireless network-based ship engine room automation monitoring system[J]. Ship Science and Technology, 2022, 44(22): 166–169
[4] 甘超莹, 甘建文, 顾野. 基于图像识别技术的城市轨道交通车站智能视频监控系统的设计与应用[J]. 城市轨道交通研究, 2023, 26(2): 76–80+85
[5] 张伟, 王宇, 陈新怡, 等. 基于图像块码本模型的监控视频背景参考帧生成方法[J]. 通信学报, 2023, 44(1): 129–141
ZHANG Wei, WANG Yu, CHEN Xin-yi, et al. Surveillance video background reference frame generation method based on image block codebook model[J]. Journal of Communications, 2023, 44(1): 129–141
