[1]袁红春,陈冠奇,张天蛟,等.基于全卷积网络的南太平洋长鳍金枪鱼渔场预报模型[J].江苏农业学报,2020,(02):423-429.[doi:doi:10.3969/j.issn.1000-4440.2020.02.024]
　YUAN Hong-chun,CHEN Guan-qi,ZHANG Tian-jiao,et al.Fishing ground forecast model of albacore tuna based on fully convolutional networks in the South Pacific[J].,2020,(02):423-429.[doi:doi:10.3969/j.issn.1000-4440.2020.02.024]
点击复制

基于全卷积网络的南太平洋长鳍金枪鱼渔场预报模型()
分享到：

江苏农业学报[ISSN:1006-6977/CN:61-1281/TN]

卷:
期数:
2020年02期
页码:
423-429
栏目:
畜牧兽医·水产养殖
出版日期:
2020-04-30

文章信息/Info

Title:
Fishing ground forecast model of albacore tuna based on fully convolutional networks in the South Pacific
作者:
袁红春1; 陈冠奇1; 张天蛟1; 宋利明2
(1.上海海洋大学信息学院，上海201306;2.上海海洋大学海洋科学学院，上海201306)
Author(s):
YUAN Hong-chun1; CHEN Guan-qi1; ZHANG Tian-jiao1; SONG Li-ming2
(1.College of Information Technology, Shanghai Ocean University, Shanghai 201306， China;2.College of Marine Sciences, Shanghai Ocean University, Shanghai 201306， China)
关键词:
全卷积网络; 三维独热编码; 渔场预报; 长鳍金枪鱼
Keywords:
fully convolutional networks; 3D one-hot encoding; fishing ground forecasting; albacore tuna
分类号:
S934
DOI:
doi:10.3969/j.issn.1000-4440.2020.02.024
文献标志码:
A
摘要:
长鳍金枪鱼（Thunnus alalunga)为南太平洋延绳钓的主要目标鱼种之一，精确预报其渔场对于提高捕捞效率和优化渔业资源管理具有重要意义。本研究依据2000-2015年南太平洋长鳍金枪鱼的延绳钓数据、渔场时空数据以及海表温度、叶绿素a浓度和海面高度3种环境因子，采用全卷积网络构建了一种以月为单位、空间分辨率为5°×5°的渔场预报模型。本研究提出三维独热编码技术将各月环境数据映射到三维矩阵的不同层上，并设计2种卷积结构和3种全卷积网络模型，利用2015年数据对研究模型进行验证，最佳模型总精准率达到72.0%。结果表明，全卷积网络在一定程度上解决了传统渔场预报方法在处理高维复杂海洋数据时准确率偏低的问题，为渔场预报提供了一种新方法。
Abstract:
Thunnus alalunga is one of main objects of longline fishing in the South Pacific. Accurate prediction of albacore tuna fisheries is of great significance for improving fishing efficiency and optimizing the management of fishery resources. Based on the historical catching data of albacore tuna, spatio-temporal data and three environmental data including sea surface temperature (SST), sea surface height (SSH) and chlorophyll-a concentration (Chla) form 2000 to 2015 in the South Pacific, a fishing ground prediction model with a monthly unit and spatial resolution of 5°×5° in the South Pacific was established using fully convolutional networks. This model mapped the environmental data to 3D array using 3D one-hot encoding, and designed two types of convolution kernel and three types of convolution network models. The prediction accuracy reached 72.0% based on the environmental data in 2015. The results show that the fully convolutional network solves the problem of low accuracy of the traditional prediction methods in processing high-dimensional complex ocean data to a certain extent, and provides a new idea for fishing ground prediction.

参考文献/References:

[1]NIKOLIC N, MORANDEAU G, HOARAU L, et al. Review of albacore tuna, Thunnus alalunga, biology, fisheries and management[J]. Reviews in Fish Biology and Fisheries, 2017, 27(4): 775-810.
[2]ZAGAGLIA C R, LORENZZETTI J A, STECH J L. Remote sensing data and longline catches of yellowfin tuna (Thunnus albacares) in the equatorial Atlantic[J]. Remote Sensing of Environment, 2004, 93(1/2):267-281.
[3]ZAINUDDIN M, SAITOH K, SAITOH S E I I. Albacore (Thunnus alalunga) fishing ground in relation to oceanographic conditions in the western North Pacific Ocean using remotely sensed satellite data[J]. Fisheries Oceanography, 2008, 17(2): 61-73.
[4]崔雪森,唐峰华,张衡,等.基于朴素贝叶斯的西北太平洋柔鱼渔场预报模型的建立[J].中国海洋大学学报(自然科学版),2015,45(2):37-43.
[5]宋利明,周建坤,沈智宾,等.基于支持向量机的库克群岛海域长鳍金枪鱼栖息环境综合指数[J].海洋通报,2017,36(2):195-208.
[6]HARRELL F E, LEE K L, MARK D B. Tutorial in biostatistics multivariable prognostic models: issues in developing models, evaluating assumptions and adequacy, and measuring and reducing errors[J]. Statistics in Medicine,1996，1(15):361-387.
[7]李思琴,林磊,孙承杰.基于卷积神经网络的搜索广告点击率预测[J].智能计算机与应用,2015,5(5):22-25，28.
[8]LIU Q, YU F, WU S, et al. A convolutional click prediction model[C]//Association for Computing Machinery. Proceedings of the 24th ACM international on conference on information and knowledge management. New York， USA: ACM, 2015: 1743-1746.
[9]WANG P, XU B, XU J, et al. Semantic expansion using word embedding clustering and convolutional neural network for improving short text classification [J]. Neurocomputing, 2016,174: 806-814.
[10]毛江美,陈新军,余景.基于神经网络的南太平洋长鳍金枪鱼渔场预报[J].海洋学报,2016,38(10):34-43.
[11]宋利明,谢凯,赵海龙,等.库克群岛海域海洋环境因子对长鳍金枪鱼渔获率的影响[J].海洋通报,2017,36(1):96-106.
[12]范永超,戴小杰,朱江峰,等.南太平洋长鳍金枪鱼延绳钓渔业CPUE标准化[J].海洋湖沼通报,2017(1):122-132.
[13]ZAINUDDIN M, SAITOH S, SAIROH K. Detection of potential fishing ground for albacore tuna using synoptic measurements of ocean color and thermal remote sensing in the northwestern North Pacific[J]. Geophysical Research Letters, 2004, 31(20)：183-213.
[14]ISMAIL A I, MORRISION E C, BURT B A, et al. Natural history of periodontal disease in adults: findings from the tecumseh periodontal disease study[J]. Journal of Dental Research, 1990, 69(2): 430-435.
[15]陈雪忠,樊伟,崔雪森,等.基于随机森林的印度洋长鳍金枪鱼渔场预报[J].海洋学报,2013,35(1):158-164.
[16]KRIZHEVSKY A,SUTSKEVER I,HINTON G E. Imagenet classification with deep convolutional neural networks[C]//NIPS. Advances in neural information processing systems. USA: Curran Associates Inc,2012: 1097-1105.
[17]SPTINGENBERG J T, DOSOVITSKIY A, BROX T, et al. Striving for simplicity: The all convolutional net[EB/OL].(2015-4-13)
[2019-7-15]. https://arxiv.org/abs/1412.6806.pdf.
[18]张衡,崔雪森,樊伟.基于遥感数据的智利竹筴鱼渔场预报系统[J].农业工程学报,2012,28(15):140-144.
[19]PRECHELT L. Automatic early stopping using cross validation: quantifying the criteria[J]. Neural Networks, 1998, 11(4): 761-767.

备注/Memo

备注/Memo:
收稿日期：2019-07-21基金项目：国家自然科学基金项目（41776142）；上海市青年科技英才扬帆计划项目（YF1407700）；上海海洋大学海洋科学研究院开放课题（A1-2006-00-601606）作者简介：袁红春（1971-），男，江苏海门人，博士，教授，博士生导师，主要从事专家系统、智能计算、智能信息处理等研究工作。（E-mail）hcyuan@shou.edu.cn
更新日期/Last Update: 2020-05-18