基于信道状态信息的船舶驾驶台人员检测及活跃度评价方法

张屹凡; 陈梦达; 王露; 陈聪; 刘克中; 陈默子

doi:10.3963/j.jssn.1674-4861.2023.04.010

基于信道状态信息的船舶驾驶台人员检测及活跃度评价方法

doi: 10.3963/j.jssn.1674-4861.2023.04.010

张屹凡^1,,
陈梦达¹,
王露²,
陈聪²,
刘克中^{1, 3, ,},
陈默子^{1, 3, 4}

1.
武汉理工大学航运学院武汉 430063
2.
中船邮轮科技发展有限公司上海 200137
3.
广东省内河港航产业研究有限公司广东韶关 512100
4.
武汉理工大学内河航运技术湖北省重点实验室武汉 430063

基金项目:

国家自然科学基金面上项目 51979216

湖北省自然科学基金创新群体项目 2021CFA001

湖北省自然科学基金青年项目 20221J0059

详细信息

作者简介:
张屹凡（1997—），硕士研究生.研究方向：智能航运、无线感知.E-mail：yifan.zhang@whut.edu.cn

通讯作者:
刘克中（1976—），博士，教授.研究方向：水上交通安全.E-mail: kzliu@whut.edu.cn

中图分类号: U691+.31
计量
- 文章访问数: 505
- HTML全文浏览量: 262
- PDF下载量: 14
- 被引次数: 0
出版历程
- 收稿日期: 2022-09-30
- 网络出版日期: 2023-11-23

A Method for Detecting Personnel at Vessel Bridge and Evaluating Level of Activities Based on Channel State Information

ZHANG Yifan^1
,,
CHEN Mengda¹,
WANG Lu²,
CHEN Cong²,
LIU Kezhong^{1, 3
, ,},
CHEN Mozi^{1, 3, 4}

1.
School of Navigation, Wuhan University of Technology, Wuhan 430063, China
2.
CSSC Cruise Technology Development Co., Ltd., Shanghai 200137, China
3.
Guangdong Inland Port and Shipping Industry Research Co., Ltd, Shaoguan 512000, Guangdong, China
4.
Hubei Provincial Key Laboratory of Inland Navigation Technology, Wuhan University of Technology, Wuhan 430063, China

摘要

摘要: 船舶驾驶台人员包括按照规定要求的常规值班人员和特殊情况下额外的瞭望人员或船长、引航员等，驾驶台人员活跃度是判断其工作状态的重要指标之一。传统的基于计算机视觉的人员检测方法在面对船舶驾驶台遮挡物多、夜间或恶劣天气下光线不足等问题时，精度明显降低。为解决该问题，提出了1种基于普通商用Wi-Fi设备的活跃度感知方法。由于船体材质、结构特点以及变化的运动状态导致动态多径多、信号噪声强，对Wi-Fi设备造成干扰，为此设计了值班高关联度数据（duty high correlation data，DHCD）选择模块及基于信道状态信息（channel state information，CSI）的多层级特征提取模块。DHCD选择模块分析驾驶台人员不同航行、值班情况下的CSI特点，对比0~5人在驾驶台内值班、工作时的信道变化，利用模糊C-means聚类算法提取CSI中对值班人员行为反应最灵敏的信道，去除对信号噪声反应敏感的信道信息；通过多层级特征提取模块计算去噪后CSI数据的幅值与相位离散度、多链路融合离散度、变异指数等多层特征，作为活跃度评价基础参数。依据驾驶台值班要求设计了驾驶台人员活跃度评价模块，采用支持向量机算法判断驾驶台人员数量，采用客观赋权法得到基础参数权重，结合人数信息与权重信息评价驾驶台人员活跃度。实验结果表明：使用DHCD选择模块和多层级模块处理后的多层级特征将驾驶台人员数量检测精度提升至89.6%，对比直接使用原始数据时检测精度提升7.1%。在夜间、雨雾天气等光照不足情况下，基于计算机视觉方法的检测精度会由光线充足时的96.2%降至60.3%，而该方法监测精度不会降低。因此，基于CSI的驾驶台人员活跃度检测方法丰富了驾驶台人员检测算法，能有效识别船舶驾驶台人员是否符合安全值班的基本要求。
- 航行安全 /
- 船舶驾驶台人员 /
- 活跃度评价模型 /
- 无线感知 /
- 信道状态信息
Abstract: The personnel on bridge consist of regularly scheduled officers on watch and additional person for look-out, captain, and pilots in specific circumstances. The activity level of the personnel on bridge is one of the crucial indicators to assess their work status. Traditional computer vision-based personnel detection methods show reduced accuracy when confronted with challenges such as multiple obstructions on the ship's bridge, insufficient light conditions during nighttime or adverse weather conditions. To address this issue, a detection and activity evaluation method based on ordinary commercial Wi-Fi devices is proposed. Due to the dynamic multipath and strong signal noise caused by the ship's material and structural characteristics and the changing motion states, the function of Wi-Fi devices is interfered. To mitigate these challenges, a duty high-correlation data (DHCD) selection module and a multi-layer feature extraction module based on channel state information (CSI) are designed. The DHCD selection module analyses the CIS characteristics in different navigation and duty situations and compares the channel variations when 0-5 people on the ship's bridge. The fuzzy C-means clustering algorithm is employed to extract the most responsive channel information to the behavior of personnel on bridge while eliminating the information sensitive to signal noise. The multi-layer feature extraction module calculates various features, including amplitude, phase dispersion, multi-link fusion dispersion, and variation index for denoised CSI data as the foundation for activity evaluation. The activity evaluation module is designed primarily based on the requirements for the on-duty personnel on bridge. The Support Vector Machine algorithm is utilized to determine the number of bridge personnel, while the Criteria Importance through the Intercriteria Correlation method is used to obtain the weight for basic parameters. Combining the headcount information and weight information, the activity level of bridge personnel is evaluated. The results indicate that the multi-layer features using the DHCD selection module and multi-layer module processing improve the accuracy of detecting the number of bridge personnel to 89.6%, representing a 7.1% increase compared to directly using raw data. In low-light conditions such as nighttime, rainy, or foggy weather, the accuracy of computer vision-based methods decreases from 96.2% under normal light to 60.3%. In contrast, the detection accuracy of proposed method remains stable. Therefore, the CSI-based detection and activity evaluation method enriches the detection algorithm for bridge personnel and can effectively identify whether the personnel meet the basic requirements for safe duty.
- navigation safety /
- personnel on bridge /
- activity evaluation model /
- wireless perception /
- channel state information

HTML全文

图 1 船舶驾驶室无线检测示意图

Figure 1. Schematic diagram of wireless detection on the ship's bridge

下载: 全尺寸图片幻灯片

图 2 室内不同人数信道状态信息相位图

Figure 2. Phase diagram of channel state information for different numbers of people indoors

下载: 全尺寸图片幻灯片

图 3 室内不同人数信道状态信息幅值图

Figure 3. Phase diagram of channel state information for different numbers of people indoors

下载: 全尺寸图片幻灯片

图 4 3人不同动作信道状态信息对比图

Figure 4. Comparison diagram of channel state information of different actions of a three persons

下载: 全尺寸图片幻灯片

图 5 原始数据与筛选后高关联度数据值班驾驶人信号幅值离散度对比图

Figure 5. Contrast diagram of amplitude dispersion of duty crew signal between raw data and CHCD

下载: 全尺寸图片幻灯片

图 6 原始数据与筛选后高关联度数据值班驾驶人相位离散度对比图

Figure 6. Contrast diagram of phase dispersion of duty crew signal between raw data and CHCD

下载: 全尺寸图片幻灯片

图 7 不同动作活跃度值对比拟合图

Figure 7. Graph of different actions activity

下载: 全尺寸图片幻灯片

图 8 船舶驾驶人活跃度检测流程

Figure 8. Process of OOW activity detection

下载: 全尺寸图片幻灯片

图 9 实船驾驶室实验场景图

Figure 9. Experimental scene of the cab of the real ship

下载: 全尺寸图片幻灯片

图 10 不同值班情况多人活跃度对比评价结果

Figure 10. Results of multi-person activity in different duty situations comparative evaluation

下载: 全尺寸图片幻灯片

图 11 一般数据与高关联度数据活跃度评价结果对比图

Figure 11. Result of comparison chart of activity evaluation between general data and high correlation data

下载: 全尺寸图片幻灯片

表 1 室内值班人数分类结果

Table 1. Result of indoor duty number classification

人数	准确率/%
0	96.3
1	93.2
2	91.5
3	89.2
4	86.7

下载: 导出CSV

表 2 船舶驾驶室值班人数要求

Table 2. Requirements for the number of people on duty in the ship bridge

航行时间/h	值班人数/人
航行时间/h	船舶吨位>3 000	船舶吨位>1 000~3 000	船舶吨位>600~1 000	船舶吨位>300~600	船舶吨位>100~300	船舶吨位≤ 100
< 10	4	3	1	1	1	1
10~16	4	3	2	2	2	2
>16	6	4	3	2	2	2

下载: 导出CSV

表 3 船舶驾驶室值班活跃度要求

Table 3. Requirements for the activity of people on duty in the ship bridge

航行时间/h	值班活跃度等级
航行时间/h	船舶吨位>3 000	船舶吨位>1 000~3 000	船舶吨位>600~1 000	船舶吨位>300~600	船舶吨位>100~300	船舶吨位≤ 100
< 10	高	中	低	低	低	低
10~16	高	中	中	中	中	中
>16	高	高	中	中	中	中

下载: 导出CSV

表 4 特征值组对准确率影响

Table 4. The influence of different feature accuracy

活跃度等级	准确率/%
活跃度等级	A	B	C
无	99.1	98.3	97.2
低	95.6	80.3	83.6
中	91.2	83.3	66.5
高	83.7	72.6	60.4

下载: 导出CSV

表 5 高关联度数据准确率的影响

Table 5. The influence of CHCD accuracy

驾驶室人数	准确率/%
驾驶室人数	一般数据	CHCD	准确率差/%
1	94.7	98.3	3.6
2	87	93.4	6.4
3	76.4	85.6	9.2
4	72.2	81.1	8.9

下载: 导出CSV

参考文献(21)

[1]	EMSA. Review of marine casualties and incidents[EB/OL]. (2019-4-23)[2022-9-23]. http://www.emsa.europa.eu/news-apress-ce
[2]	交通运输部海事局. 2018年水上交通安全形势分析[J]. 中国海事, 2019(5): 32-33 https://www.cnki.com.cn/Article/CJFDTOTAL-HSZG201905018.htm China MSA. Analysis on waterborne transport safety situation for year 2018[J]. China Maritime Safety, 2019(5): 32-33(in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-HSZG201905018.htm
[3]	DJANI M, ROBERT M, BARIC M. Deficiencies in learning COLREGs and new teaching methodology for nautical engineering students and seafarers in lifelong learning programs[J]. The Journal of Navigation, 2016, 69(4): 765-776. doi: 10.1017/S037346331500096X
[4]	BERGASA L M, NUEVO J, SOTELO M A, et al. Real-time system for monitoring driver vigilance[J]. IEEE Transactions on Intelligent Transportation Systems, 2006, 7(1): 63-77. doi: 10.1109/TITS.2006.869598
[5]	MARGARETA L, DUKIC T. Show me where you look and I'll tell you if you're safe: Eye tracking of maritime watchkeepers[C] Nordic Ergonomics Society annual conference, Lysekil, Sweden: Nordic Ergonomics Society, 2007.
[6]	YOUN I H, PARK D J, YIM J B. Analysis of lookout activity in a simulated environment to investigate maritime accidents caused by human error[J]. Applied Sciences, 2018, 9(1): 4-10. doi: 10.3390/app9010004
[7]	杨世超. 基于红外图像的驾驶人疲劳监控系统研究[D]. 沈阳: 东北大学, 2015. SHI C Y. Research of Driver Fatigue Detecting Based on Infrared Image[D]. Shengyang: Northeastern University, 2011(in Chinese)
[8]	赵晓华, 房瑞雪, 毛科俊. 基于生理信号的驾驶疲劳声音对策有效性实验[J]. 西南交通大学学报, 2010, 45(3): 457-463. doi: 10.3969/j.issn.0258-2724.2010.03.024 ZHAO X H, FANG R X, MAO K J. Test on the effectiveness of sound as a countermeasure against driving fatigue based on physiological signals[J]. Journal of Southwest Jiaotong University, 2010, 45(3): 457-463. (in Chinese) doi: 10.3969/j.issn.0258-2724.2010.03.024
[9]	钟铭恩, 吴平东, 彭军强, 等. 基于脑电信号的驾驶人情绪状态识别研究[J]. 中国安全科学学报, 2011(9): 64-69. ZHONG M E, WU P D, PENG J Q, et al. Study on an emotional state recognition technology based on drivers'EEGs[J]. China Safety Science Journal, 2011, 21(9): 64-69. (in Chinese)
[10]	周锋, 吴华锋, 孙志宽. 1种新型船舶驾驶台值班防疲劳监控系统设计[J]. 船海工程, 2017, 46(2): 170-174. doi: 10.3963/j.issn.1671-7953.2017.02.040 ZHOU F, WU H F, SUN Z K. Design of bridge navigation watch alarm system based on image processing[J]. Ship & Ocean Engineering, 2017, 46(2): 170-174(in Chinese) doi: 10.3963/j.issn.1671-7953.2017.02.040
[11]	陈家豪, 刘克中, 陈默子, 等. 基于信道状态信息的船舶敏感区域入侵检测方法[J]. 大连海事大学学报, 2019, 45(1): 89-95. https://www.cnki.com.cn/Article/CJFDTOTAL-DLHS201901011.htm CHEN J H, LIU K Z, CHEN M Z, et al. Intrusion detection method of ships sensitive regions based on channel state information[J]. Journal of Dalian Maritime University, 2019, 45(1): 89-95. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-DLHS201901011.htm
[12]	CHEN M Z, Ma J, ZENG X M. MD-alarm: A novel manpower detection method for ship bridge watchkeeping using Wi-Fi signals[J]. IEEE Transactions on Instrumentation and Measurement, 2022, 71(3): 1-13.
[13]	LIU K Z, WANG Y, CHEN M Z, et al. Deep learning based wireless human motion tracking for mobile ship environments[J]. IEEE Internet of Things Journal, 2022, 23(9): 24186-24198.
[14]	HALPRIN D, HU W, SHETH A, et al. Tool release: Gathering 802.11 n traces with channel state information[J]. ACM SIGCOMM Computer Communication Review, 2011, 41(1): 53-53. doi: 10.1145/1925861.1925870
[15]	陈伟炯. 船舶安全与管理[M]. 大连: 大连海事大学出版社, 1998. CHEN W J. Ship safety and management[M]. Dalian: Dalian Maritime University Press, 1998. (in Chinese)
[16]	黄琛, 陈德山, 吴兵, 严新平. 船舶航行交通事件实时检测技术研究现状与展望[J]. 交通信息与安全, 2022, 40(6): 1-11. doi: 10.3963/j.jssn.1674-4861.2022.06.001 HUANG C, CHEN D S, WU B, YAN X P. A real-time detection of nautical traffic events: A review and prospect[J]. Journal of Transport Information and Safety, 2022, 40(6): 1-11. doi: 10.3963/j.jssn.1674-4861.2022.06.001
[17]	SIMONE D D, SANCTIC D M, CIANCA E, et al. A trained-once crowd counting method using differential wifi channel state information[C]. A Campus-wide Testbed for Studying Mobile Physical Activities, New York, United States: ACM, 2016.
[18]	QIAN K, WU C S, YANG Z, et al. Widar2.0: Passive human tracking with a single Wi-Fi link[C] ACM International Conference on Mobile Systems, Applications, and Services, New York, United States: ACM, 2016.
[19]	YANG Z, WU D, XIONG J, et al. FarSense: Pushing the range limit of WiFi-based respiration sensing with a CSI ratio of two antennas[J]. International Joint Conference on Pervasive and Ubiquitous Computing, 2019, 3(3): 1-26.
[20]	DAVIES L, GATHER U, The identification of multiple outliers, Journal of the American Statistical Association[J]. Journal of the American Statistical Association, 1993, 88(423): 782-792.
[21]	PAIK B G, CHO S R, PARK B J, et al. Characteristics of wireless sensor network for full-scale ship application[J]. Journal of Marine Science and Technology, 2009, 14(1): 115-126.