FVID: Fishing Vessel Type Identification Based on VMS Trajectories

Vessel Monitoring System(VMS) provides a new opportunity for quantified fishing research. Many approaches have been proposed to recognize fishing activities with VMS trajectories based on the types of fishing vessels. However, one research problem is still calling for solutions, how to identify the fishing vessel type based on only VMS trajectories. This problem is important because it requires the fishing vessel type as a preliminary to recognize fishing activities from VMS trajectories. This paper proposes fishing vessel type identification scheme(FVID) based only on VMS trajectories. FVID exploits feature engineering and machine learning schemes of XGBoost as its two key blocks and classifies fishing vessels into nine types. The dataset contains all the fishing vessel trajectories in the East China Sea in March 2017, including 10031 pre-registered fishing vessels and 1350 unregistered vessels of unknown types. In order to verify type identification accuracy, we first conduct a 4-fold cross-validation on the trajectories of registered fishing vessels. The classification accuracy is 95.42%. We then apply FVID to the unregistered fishing vessels to identify their types. After classifying the unregistered fishing vessel types, their fishing activities are further recognized based upon their types. At last, we calculate and compare the fishing density distribution in the East China Sea before and after applying the unregistered fishing vessels, confirming the importance of type identification of unregistered fishing vessels.

Analyses of Trawling Track and Fishing Activity Based on the Data of Vessel Monitoring System(VMS): A Case Study of the Single Otter Trawl Vessels in the Zhoushan Fishing Ground

The original purpose of Vessel Monitoring System(VMS) is for enforcement and control of vessel sailing. With the application of VMS in fishing vessels, more and more population dynamic studies have used VMS data to improve the accuracy of fisheries stock assessment. In this paper, we simulated the trawl trajectory under different time intervals using the cubic Hermite spline(c Hs) interpolation method based on the VMS data of 8 single otter trawl vessels(totally 36000 data items) fishing in Zhoushan fishing ground from September 2012 to December 2012, and selected the appropriate time interval. We then determined vessels' activities(fishing or non-fishing) by comparing VMS speed data with the corresponding speeds from logbooks. The results showed that the error of simulated trajectory greatly increased with the increase of time intervals of VMS data when they were longer than 30 minutes. Comparing the speeds from VMS with those from the corresponding logbooks, we found that the vessels' speeds were between 2.5 kn and 5.0 kn in fishing. The c Hs interpolation method is a new choice for improving the accuracy of estimation of sailing trajectory, and the VMS can be used to determine the vessels' activities with the analysis of their trajectories and speeds. Therefore, when the fishery information is limited, VMS can be one of the important data sources for fisheries stock assessment, and more attention should be paid to its construction and application to fisheries stock assessment and management.

Study on economic and social benefits of the FRP ocean fishing vessels

Ocean fishing is one of the most important parts of the fishery economy in China. After more than 20 years of development, China has developed into one of the major ocean fishing countries in the world. However, due to the influence of natural laws at home and abroad, China＇s deep-sea fisheries have been faced with difficulties such as rising costs, increasing competition pressure on fishery resources and backward fishery equipment. In recent years, the development of ocean fishing has faced many difficulties. Therefore, this paper analyzes the economic and social benefits of the steel ocean fishing vessel and the glass and steel ocean fishing vessel. It has reached the aim of providing theoretical support for the new development direction of ocean fishing in China. Among them, the economic benefits are mainly compared and analyzed from four aspects of the maintenance fee, depreciation, fuel consumption and operational efficiency. Social benefits are mainly embodied in the aspects of energy saving, emission reduction, increasing consumption and reducing consumption, protecting resources and environment, ensuring the safety of fishermen at sea, optimizing the structure of fishery industry, and promoting standardization of fishing vessels. Through the analysis of these two aspects, the advantage and the obvious gain effect of the FRP ocean fishing vessel are fully proved. The FRP ocean fishing vessels can be realized more quickly by the enterprises built and used. Finally, it can promote the development of ocean FRP Ocean fishing vessels in China.

Contribution of Human Factors to Fishing Vessel Accidents and Near Misses in the UK

The research paper in hand presents a thorough exploration of the fishing vessel accidents and near misses in the UK fishing industry as well as the underlying human element factors and sub-factors contributing to them. In this respect, the regulatory regime in the fishing industry both at a national and international level is initially examined while also complemented by the investigation of past research efforts to address these issues. Furthermore, the analysis of the fishing vessels accidents and near misses as recorded in the UK MAIB （Marine Accident Investigation Branch） database for a period of 19 years is performed in order to derive the very causal factors leading to the fishing vessel accidents. It is initially shown that the fatalities and injuries taking place due to fishing vessels＇ accidents have alarmingly remained unchanged over the last 15-20 years. Another key finding is that the number of accidents and near misses per day and night shifis is quite similar while most accidents take place in coastal waters. Furthermore, human factors are related to the vast majority of fishing vessels accidents with the principal ones referring to ＂non-compliance＇, ＂equipment misuse or poorly designed＂, ＂training＂ and ＂competence＂. Finally, remedial measures are also suggested in order to address the main accident causes identified.

Study on Net Hauler Safety Device for Fishing Vessels

[Objective] To design a safety device set for fishing vessel net hauler, in order to reduce the severe potential safety hazards in offshore operation. [Method] In view of the operation and structure of net hauler on the fishing trawler, this paper investigated the existing safety protection devices and designed a new set of purely mechanical structure without changing the original structures of the vessel significantly. [Result] Using the easily-adjustable functions of the main shaft of CNC lathe, and simulating the operation performance of different hauling line speeds and testing the safety device, the results showed that the new device could protect the crew from safety accidents caused by net hauler. [Conclusion] The result provides references for the design and application of mechanical protection devices for net haulers on other fishing trawlers.

Identification of navigation characteristics of single otter trawl vessel using four machine learning models

Fishing logbook records the fishing behaviors and other information of fishing vessels.However,the accuracy of the recorded information is often difficult to guarantee due to the misreport and concealment.The fishing vessel monitoring system(VMS)can monitor and record the navigation information of fishing vessels in real time,and it may be used to improve the accuracy of identifying the state of fishing vessels.If the VMS data and fishing logbook are combined to establish their relationships,then the navigation characteristics and fishing behavior of fishing vessels can be more accurately identified.Therefore,first,a method for determining the state of VMS data points using fishing log data was proposed.Secondly,the relationship between VMS data and the different states of fishing vessels was further explored.Thirdly,the state of the fishing vessel was predicted using VMS data by building machine learning models.The speed,heading,longitude,latitude,and time as features from the VMS data were extracted by matching the VMS and logbook data of three single otter trawl vessels from September 2012 to January 2013,and four machine learning models were established,i.e.,Random Forest(RF),Adaptive Boosting(AdaBoost),K-Nearest Neighbor(KNN),and Gradient Boosting Decision Tree(GBDT)to predict the behavior of fishing vessels.The prediction performances of the models were evaluated by using normalized confusion matrix and receiver operator characteristic curve.Results show that the importance rankings of spatial(longitude and latitude)and time features were higher than those of speed and heading.The prediction performances of the RF and AdaBoost models were higher than those of the KNN and GBDT models.RF model showed the highest prediction performance for fishing state.Meanwhile,AdaBoost model exhibited the highest prediction performance for non-fishing state.This study offered a technical basis for judging the navigation characteristics of fishing vessels,which improved the algorithm for judging the behavior of fishing vessels based on VMS data,enhanced the prediction accuracy,and upgraded the fishery management being more scientific and efficient.

波浪作用下小尺度多船并排锚泊横摇运动试验研究

采用物理模型试验方法研究双船、三船和四船并排锚泊时各船的运动量,并与单船艏艉锚泊情况进行了对比,重点分析了横浪作用下各船的横摇运动规律,探究了波浪周期和并排锚泊船数对多船并排锚泊时横摇运动的影响。结果表明,多船并排锚泊时,横摇运动响应随着波浪周期的增大而逐渐减小。双船并排锚泊时,横摇运动响应同单船相比减少了40%~50%;三船和四船并排锚泊时,横摇运动响应同单船相比减少50%~60%,四船并排锚泊时横摇运动响应减小的幅度与三船并排时相比差别不大。

考虑多影响因素的渔船单船捕捞能力评估方法

针对以往海洋机动渔船捕捞能力评估研究中考虑影响因素较少,对各影响因素的主观权重值和客观权重值综合分析不足,不能给出渔船单船捕捞能力量化值等问题,该研究提出了考虑多影响因素的渔船单船捕捞能力评估模型。首先,考虑多种影响渔船捕捞能力的因素,建立单船捕捞能力评估指标体系,将其分为可量化指标和不可量化指标,并初步制定了不可量化指标评估标准。其次,利用专家、渔民调查问卷中的数据,运用层次分析法计算各指标主观权重值,结果表明各指标权重依次为:捕捞装备(0.108)、功率(0.094)、拖网(0.074)、作业时长(0.071)、总吨(0.049)、探鱼仪器(0.047)、刺网(0.040)、船长(0.032)、张网(0.028)、围网(0.024)、钓业(0.021)、作业环境(0.019)、钢质材质(0.019)、罩网(0.017)、玻璃钢材质(0.016)、船龄(0.013)、木制材质(0.012);渔具相关指标权重依次为:网具主尺寸(0.402)、网具结构(0.149)、装配技术(0.093)、制造材料(0.051)。最后,为兼顾主客观权重优点,基于加法合成法、博弈论法、最小鉴别信息法对各指标权重值进行组合计算,以获得目标权重值,并根据某省渔船数据对建立的模型进行验证分析,通过Spearman等级相关系数法计算捕捞能力评估值与实际渔获量之间的等级相关系数,结果表明基于博弈论法得到的结果相关性最高,相关系数为0.937,证实了组合权重值和评估模型的科学性和合理性。研究结果可为渔业管理部门制定科学合理的渔业资源调控政策提供依据。

机器学习算法在渔船数据挖掘技术中的应用

本文基于PCA_Word2Vec的渔船数据挖掘技术展开研究,构建了PCA_Word2Vec渔船数据挖掘模型并对渔船轨迹数据进行挖掘,在此基础之上通过将LightGBM模型与其他经典模型分类效果进行比较,以及对PCA_Word2Vec数据挖掘模型挖掘前后的数据展开对比,验证了渔船数据挖掘技术的有效性和实用性。

船型网箱运动-波浪场扰动-网衣受力耦合动力响应特性

船型网箱是一种由大型浮体、钢架、网衣系统及系泊系统构成的新型海洋养殖结构.大型浮体在波浪中产生的绕射和辐射作用会对波浪场产生扰动,进而改变网衣受力,这种相互作用此前还未被研究过.本文基于势流理论求得绕射波和辐射波在网衣周围波浪场的速度传递函数,同时采用莫里森方程求解计及波浪场扰动的网衣水动力载荷;对网衣水动力载荷与浮体运动不断迭代计算,实现计及网箱运动-波浪场扰动-网衣受力耦合效应的船型网箱动力响应分析;最后,研究了不同幅值的规则波下这种耦合效应对网箱运动、网衣张力、容积损失和连接器载荷的影响规律.结果发现:波浪场扰动对网箱运动和容积损失的影响不明显,但是会导致网衣张力和连接器载荷明显增大,这能够为船型网箱的结构强度校核及安全设计提供指导.

基于BWM-DS证据理论的大风浪环境下渔船航行安全评估

为预判渔船在大风浪环境下航行时的安全状况,从人、船、环境和管理等4方面进行分析,结合因果分析图法建立渔船在大风浪环境下航行的安全评估指标体系。运用最优最劣法(BWM)确定各指标的权重,采用D-S证据理论的合成标准融合各评价指标的权重,通过基本信任函数赋值的方法得到基本可信度值,确定其风险,得到基于BWM-DS证据理论的渔船安全评估新模型。运用该模型对渔船“粤阳江渔07128”在大风浪环境下翻沉的事故进行评估,验证该模型的有效性。评估结果表明,采用该模型评估所得结果与事故实际情况和专家主观评估结果相符,证明该模型能有效评价渔船在大风浪环境下的安全航行状况。

基于改进YOLOv8的渔港船舶进出港目标检测与统计方法

为减轻渔港执法人员在渔船进出港统计工作中的压力,提出了一种基于改进YOLOv8模型的渔港船舶进出港目标检测与统计模型YOLOv8n-Bi-FPN-EMA,该模型采用加权双向特征金字塔网络结构(Bi-FPN),引入高效多尺度注意模块(EMA)以提高模型的目标检测准确性和稳定性,并结合BoT-SORT跟踪算法实现对不同类型船舶进出渔港的分类统计。结果表明:与YOLOv8n模型相比,YOLOv8n-Bi-FPN-EMA模型的准确率提高了5.0%,mAP@0.5提高了0.2%,对执法船、拖船和渔船的识别准确率分别提高了9.9%、4.9%、0.3%。其中,对渔船的识别准确率为99.1%,召回率为84.4%,mAP@0.5为97.8%。研究表明,YOLOv8n-Bi-FPN-EMA模型在渔船识别上有较好的表现,可用于渔船识别与进出港统计的辅助工作。

基于渔船轨迹数据的进出港区域识别方法

针对当前渔船进出港区域获取方法成本高、更新周期长等问题,提出了一种基于渔船轨迹数据的渔船进出港区域识别方法。首先,提出基于多特征融合下轨迹点间相似性的轨迹划分算法,将渔船轨迹划分为不同渔船行为的轨迹段;然后,提出特征距离加权-K均值聚类算法(Feature Distance Weighted-K-means clustering algorithm,FDW-K-means),将上一步得到的轨迹段特征作为聚类对象,实现渔船进出港轨迹段的提取。最后,综合运用DBSCAN(Density-Based Spatial Clustering of Applications with Noise)聚类算法和Del-Alpha-Shape算法对聚集的渔船进出港轨迹段轨迹点集进行边界提取获得渔船进出港区域。以椒江渔港和博贺渔港2021年3月的渔船轨迹数据为例,识别到椒江渔港和博贺渔港的渔船进出港区域的正确率分别为94.2%和95.8%。与使用K-means聚类算法或传统基于对各特征设定约束条件思想提取轨迹段的方法相比,该方法识别到的渔港渔船进出港区域正确率分别提高了10.7%,8.7%和9.5%,6.6%。实验结果表明所提方法能够有效识别渔船进出港区域,其结果能为渔船进出港监管提供科学参考。

基于组合权重的渔船安全灰色综合评价模型研究

海洋渔业是国际公认的最危险行业之一,为了解中国渔船安全情况,依据国内外关于渔船安全指标的研究和中国渔业互保协会相关统计研究报告,选择千船全损数、千船部损数、万人死亡数、万人伤残数和万人受伤数5个指标构成渔船安全评价指标体系,运用层次分析法(AHP)与CRITIC权重法确认各评价指标的主客观权重,最后通过博弈论将主客观权重结合,以灰色关联度为基础建立了渔船安全灰色综合评价模型。为验证该模型的科学性与合理性,以中国渔业互保协会承保渔船、渔民事故数据为样本进行了评价研究。评价结果与该协会统计数据反映的渔船、渔民安全状况基本一致,并且与该协会专家对承保渔船安全走势总体判断也基本一致,表明模型可以应用于渔船安全评价实践。[中国渔业质量与标准,2024,14(2):43-50]

河北沿海商渔船碰撞成因规律及防范对策分析

通过对河北沿海近年来商渔船碰撞事故的统计,探寻商渔船碰撞事故发生的源头因素,并从加强自身管理、行业管理、从业者素质以及运用现代化手段等几方面提出避免商渔船碰撞的对策建议,以达到减少事故发生的目的。

基于深度学习和时空特征融合的海洋渔船密度预测方法

为了从海量渔船轨迹数据中挖掘隐含的信息和知识,进而为渔业行政主管部门的决策提供科学依据,本研究以AIS渔船轨迹数据为研究对象,提出了一种基于深度学习和面向时空特征融合的海洋渔船密度预测方法:首先,利用渔船轨迹数据集对渔船行驶区域进行网格划分;其次,筛选出渔船高密度区域进行研究,避免数据稀疏性问题;再次,根据渔船轨迹数据的时空分析,构建三维时空融合矩阵;最后,通过卷积循环神经网络模型捕获渔船分布的时间和空间特征,并利用卷积神经网络的堆叠加强对空间特征的学习。实验通过东海海域渔船真实轨迹数据进行具体测试,结果表明渔船密度预测值与真实值非常接近,平均绝对误差为4×10^(-4),模型较好地拟合了渔船密度分布特征,有效地提高了渔船捕捞热点预测的准确性和鲁棒性。

小庙洪航道渔船航路及警戒区设置探讨

根据南通沿海港区与渔港相关基础资料,分析了小庙洪航道商渔船通航安全风险,并探讨了渔船航路及警戒区布置方案。结果表明,与东灶渔港渔船习惯性航路重合的三夹沙南航道转弯段横流大、航行条件复杂,通航安全风险较大,可考虑沿东侧天然汊道布置渔船航路或东灶渔港搬迁方案。吕四中心渔港和蒿枝港渔船习惯性航路与小庙洪航道交叉,存在一定的通航安全风险,可在商渔船交会处设置警戒区,引导通航秩序和改善通航环境。

民国时期日本对上海渔业侵略的历史探究

民国时期日本对我国渔业资源的侵略不仅有渔轮大肆侵捕,倾销到上海鱼市场的直接侵渔,还有辗转浏河等其他港口运入上海,由鱼行代售的隐蔽侵渔方式。面对日本对我国渔业的侵略和破坏,国民政府采取与日本公使外交交涉、划定领海界线以及制定航行禁令限制日本渔轮等措施,民间也自发成立上海渔业仇货检查委员会等组织来抵制日本侵渔,然而国民政府采取的措施无法从根本上杜绝日本侵渔,各项政策执行的效果大打折扣,民间自发打击日本侵渔的行动亦收效甚微,日本侵渔变本加厉。

深圳西部海域渔船通航风险分析及商渔船防碰撞对策研究

通过对深圳西部海域渔船通航风险的调查研究,揭示渔业船舶在深圳西部海域的生产作业规律及其生产活动对商船通航的影响,分析渔船监管难点,提出商渔船防碰撞对策建议。

渔船安全作业的光伏智能监管装置研发与示范应用

针对2023年度广东省国库渔船AIS通导设备在线率不足50%的问题,本文研制了渔船安全作业的光伏智能监管装置,支持5、9.5、12、24 V 4种规格电源输出和剩余电量、运行状态显示,具有SOP、SOH和SOC等电池安全保护策略、故障诊断、光伏充电均衡、北斗短报文通信功能;同时开发渔船“光伏电碳中和”一张图管理平台,具备船载光伏智能监管装置定位和运行状态监控、渔船动态轨迹监控、船只回港测算、台风预报、渔船档案信息管理多种功能,实现渔船的实时监控和安全管理。设备及其平台已在惠州市惠阳区大亚湾渔港、汕尾市陆丰市甲子渔港、深汕特别合作区鲘门渔港、江门市台山市横山渔港、珠海市香洲区洪湾渔港开展示范应用,获得相关渔政部门和用户一致好评,有效提升了渔船安全作业水平。