基于Sentinel-2 影像的淡水养殖水生动物类型识别研究

doi:10.15933/j.cnki.1004-3268.2022.04.018

摘要/Abstract

摘要： 为了利用遥感影像实现内陆淡水养殖空间分布的快速提取，以江苏省宜兴市为研究区域，基于Sentinel-2卫星影像数据，提出了一种结合卷积神经网络和随机森林算法的内陆淡水养殖池塘水产类型的识别方法。该方法以深度学习为基础，构建卷积神经网络模型进行养殖池塘语义分割，进而分析养殖区域斑块的归一化植被指数和归一化水体指数，最后采用随机森林算法区分养殖池塘的水产类型。结果表明，宜兴市2021 年淡水养殖池塘面积为121.25 km²，其中蟹塘面积74.48 km²，鱼塘面积46.77 km²，识别总体精度为88.33%，kappa系数为0.824 3。

关键词: 淡水养殖池塘, Sentinel-2遥感影像, 卷积神经网络, 随机森林, SE-Unet

Abstract: In order to realize the rapid extraction of the spatial distribution of inland freshwater aquaculture based on remote sensing images，this study chose Yixing City，Jiangsu Province as the study area and proposed a method to identify inland freshwater aquaculture ponds by combining the convolutional neural network and random forest algorithm based on the Sentinel‑2 satellite images.Firstly，the study built a convolutional neural network model for semantic segmentation of aquaculture ponds based on deep learning.Then，normalized difference vegetation index（NDVI） and normalized difference water index（NDWI）of the patches in the aquaculture areas were analyzed.Finally，the random forest algorithm was used to distinguish the types of aquaculture ponds.The results showed that there were 121.25 km² of freshwater aquaculture ponds in Yixing in 2021，and the areas of crab ponds and fish ponds were 74.48 km²and 46.77 km²，respectively.The overall accuracy of the method was 88.33%，and the kappa coefficient was 0.824 3.

Key words: Freshwater aquaculture ponds, Sentinel-2 imagery, Convolutional neural network, Random forest, SE-Unet

中图分类号:

TP79

金晶, 毛星, 张欣, 刘杨, 陆学文, 任妮 . 基于Sentinel-2 影像的淡水养殖水生动物类型识别研究[J]. 河南农业科学, 2022, 51(4): 160-170.

图/表 11

参考文献

［1］WANG Q，CHENG L，LIU J，et al.Freshwater aquaculture in PR China：Trends and prospects［J］.Reviews in Aquaculture，2015，7（4）：283‑302.
［2］WANG X，TJALE P M，YAO Q，et al.Comparison of the growth performance and nutritional qualities of Chinese
mitten crab（Eriocheir sinensis） with different stocking densities in rice‑crab culture systems［J］.Aquaculture Reports，2021，20：100761.
［3］胡金有，王靖杰，张小栓，等.水产养殖信息化关键技术研究现状与趋势［J］.农业机械学报，2015，46（7）：251‑263.
HU J Y，WANG J J，ZHANG X S，et al.Research status and development trends of information technologies in aquacultures［J］.Transactions of the Chinese Society for Agricultural Machinery，2015，46（7）：251‑263.
［4］OTTINGER M，CLAUSS K，KUENZER C.Large‑scale assessment of coastal aquaculture ponds with Sentinel‑1 time series data［J］.Remote Sensing，2017，9（5）：440.
［5］DAI Y，FENG L，HOU X，et al.Policy‑driven changes in enclosure fisheries of large lakes in the Yangtze Plain：Evidence from satellite imagery［J］.Science of the Total Environment，2019，688：1286‑1297.
［6］DUAN Y，LI X，ZHANG L，et al.Mapping national‑scale aquaculture ponds based on the Google Earth Engine in
the Chinese coastal zone［J］.Aquaculture，2020，520：734666.
［7］LIU Y，WANG Z，YANG X，et al.Satellite‑based monitoring and statistics for raft and cage aquaculture in China’s offshore waters［J］.International Journal of Applied Earth Observation and Geoinformation，2020，91：102118.
［8］DUAN Y，LI X，ZHANG L，et al.Detecting spatiotemporal changes of large‑scale aquaculture ponds regions over
1988—2018 in Jiangsu Province，China using Google Earth Engine［J］.Ocean & Coastal Management，2020，188：105144.
［9］LIU X，HUANG H，YANG X，et al.Method to extract raft‑cultivation area based on SPOT image［J］.Science of Surveying and Mapping，2013，38（2）：41‑43.
［10］蒋逸明，董迪，岳文，等.基于多源高分遥感影像的围海养殖动态监测［J］.海洋信息，2021，36（1）：40‑45.
JIANG Y M，DONG D，YUE W，et al.Dynamic monitoring of coastal aquaculture ponds with multi‑source high spatial resolution imagery［J］.Marine Information，2021，36（1）：40‑45.
［11］HU B，LI Q，HALL G B.A decision‑level fusion approach to tree species classification from multi‑source remotely sensed data［J］.ISPRS Open Journal of Photogrammetry and Remote Sensing，2021，1：100002.
［12］BELGIU M，DRĂGUŢ L.Random forest in remote sensing：A review of applications and future directions［J］.ISPRS Journal of Photogrammetry and Remote Sensing，2016，114：24‑31.
［13］HAN M，CONG R，LI X，et al.Joint spatial‑spectral hyperspectral image classification based on convolutional neural network［J］.Pattern Recognition Letters，2020，130：38‑45.
［14］OSCO L P，DE ARRUDA M D S，MARCATO JUNIOR J，et al.A convolutional neural network approach for counting and geolocating citrus‑trees in UAV multispectral imagery ［J］.ISPRS Journal of Photogrammetry and Remote Sensing，2020，160：97‑106.
［15］HE N，FANG L，LI S，et al.Skip‑connected covariance network for remote sensing scene classification［J］.IEEE Transactions on Neural Networks and Learning Systems，2020，31（5）：1461‑1474.
［16］NIE X，DUAN M，DING H，et al.Attention mask R‑CNN for ship detection and segmentation from remote sensing images［J］.IEEE Access，2020，8：9325‑9334.
［17］CUI B，FEI D，SHAO G，et al.Extracting raft aquaculture areas from remote sensing images via an improved U‑Net with a PSE structure［J］.Remote Sensing，2019，11（17）：2053.
［18］ZENG Z，WANG D，TAN W，et al.RCSANet：A full convolutional network for extracting inland aquaculture ponds from high‑spatial‑resolution images［J］.Remote Sensing，2021，13（1）：92.
［19］FU Y，YE Z，DENG J，et al.Finer resolution mapping of marine aquaculture areas using WorldView‑2 imagery and a hierarchical cascade convolutional neural network［J］.Remote Sensing，2019，11（14）：1678.
［20］李亚飞，董红斌.基于卷积神经网络的遥感图像分类研究［J］.智能系统学报，2018，13（4）：550‑556.
LI Y F，DONG H B.Classification of remote‑sensing image based on convolutional neural network［J］.CAAI Transactions on Intelligent Systems，2018，13（4）：550‑556.

［21］卢业伟，李强子，杜鑫，等.基于高分辨率影像的近海养殖区的一种自动提取方法［J］.遥感技术与应用，2015，30（3）：486‑494.

LU Y W，LI Q Z，DU X，et al.A method of coastal aquaculture area automatic extraction with high spatial resolution images［J］.Remote sensing Technology and Application，2015，30（3）：486‑494.
［22］黄帅，宋开宏，罗菊花，等.基于梯度变换的浅水湖泊围网区遥感提取算法［J］.湖泊科学，2017，29（2）：490‑497.
HUANG S，SONG K H，LUO J H，et al.A remote sensing extraction algorithm of enclosure culture area in shallow lakes based on gradient transform［J］.Journal of Lake Sciences，2017，29（2）：490‑497.
［23］REN C，WANG Z，ZHANG Y，et al.Rapid expansion of coastal aquaculture ponds in China from Landsat observations during 1984—2016［J］.International Journal of Applied Earth Observation and Geoinformation，2019，82：101902.
［24］XIA M，QIAN J，ZHANG X，et al.River segmentation based on separable attention residual network［J］.Journal of Applied Remote Sensing，2019，14（3）：32602.
［25］SHIH H，NG P.Diversity and biogeography of freshwatercrabs（Crustacea：Brachyura：Potamidae，Gecarcinucidae）from East Asia［J］.Systematics and Biodiversity，2011，9（1）：1‑16.
［26］宜兴市农业农村局. 宜兴市渔业养殖规划（2017—2025年）［EB/OL］.［2021‑10‑15］.http：//www.yixing.gov.cn/doc/2017/12/28/625493.shtml.
Yixing City Agricultural and Rural Bureau.Fishery breeding planning of Yixing City （2017—2025）［EB/OL］.［2021‑10‑15］.http：//www.yixing.gov.cn/doc/2017/12/28/625493.shtml.
［27］ZENG Q，JEPPESEN E，GU X，et al.Cannibalism and habitat selection of cultured Chinese mitten crab：Effects of submerged aquatic vegetation with different nutritional and refuge values［J］.Water，2018，10（11）：1542.
［28］LIU Y，FU W，JIN M，et al.Effects of different aquatic plants on water purification and greenhouse gas emission in crab pond in high temperature season［J］.Journal of Ecology and Rural Environment，2020，36（8）：1072‑1079.
［29］RONNEBERGER O，FISCHER P，BROX T.U‑Net：Convolutional networks for biomedical image segmentation［C］// International conference on medical image computing and computer‑Assisted Intervention.Switzerland：Springer International Publishing，2015.
［30］WANG Y，ZHANG Z，FENG L，et al.A new attention‑based CNN approach for crop mapping using time series Sentinel‑2 images［J］.Computers and Electronics in Agriculture，2021，184：106090.
［31］PASZKE A，GROSS S，MASSA F，et al.Pytorch：An imperative style，high‑performance deep learning library［J］.Advances in Neural Information Processing Systems，2019，32：8026‑8037.
［32］SHI L，JIN M，SHEN M，et al.Using Ipomoea aquatic as an environmental‑friendly alternative to Elodea nuttallii for the aquaculture of Chinese mitten crab［J］.PeerJ，2019，7：e6785.
［33］ZHANG B Y，LIU X N，LIU M L，et al.Detection of rice phenological variations under heavy metal stress by means of blended Landsat and MODIS image time series［J］.Remote Sensing，2019，11（1）：13.
［34］ASHOK A，RANI H P，JAYAKUMAR K V.Monitoring of dynamic wetland changes using NDVI and NDWI based landsat imagery［J］.Remote Sensing Applications：Society and Environment，2021，23：100547.
［35］GISLASON P O，BENEDIKTSSON J A，SVEINSSON J R.Random forests for land cover classification［J］.Pattern Recognition Letters，2006，27（4）：294‑300.
［36］LUO Y，HUANG D，LIU P，et al.An novel random forests and its application to the classification of mangroves remote sensing image［J］.Multimedia Tools and Applications，2016，75（16）：9707‑9722.
［37］DONG L，DU H，MAO F，et al.Very high resolution remote sensing imagery classification using a fusion of random forest and deep learning technique‑subtropical area for example［J］.IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing，2020，13：113‑128.
［38］MYINT S，LAM N，TYLER J.Wavelets for urban spatial feature discrimination：Comparisons with fractal， spatial autocorrelation， and spatial co‑occurrence approaches［J］.Photogrammetric Engineering and Remote Sensing ，2004，70：803‑812.
［39］DAI J，LI Y，HE K，et al.R‑FCN：Object detection via region‑based fully convolutional networks［C］//Advances in neural information processing systems.Barcelona，Spain：Curran Associates Inc，2016.
［40］TURCIOS A E，PAPENBROCK J.Sustainable treatment of aquaculture effluents：What can we learn from the past
for the future？［J］.Sustainability，2014，6（2）：836‑856.
［41］LIANG C，CHENG B，XIAO B，et al.Semi‑/weakly‑supervised semantic segmentation method and its application for coastal aquaculture areas based on multi‑source remote sensing images：Taking the Fujian coastal area（mainly Sanduo） as an example［J］.Remote Sensing，2021，13（6）：1083.

[1]	付博, 杨永锋, 刘向真, 赵森森, 刘茂林, 贾国涛, 牛洋洋, 张坤芳, 于建军, 彭桂新, 姬小明. 基于近红外光谱的烤烟香型分类模型研究[J]. 河南农业科学, 2023, 52(7): 163-171.
[2]	张科谦, 程钢, 吴微, 宋向阳, 张子谦, 姚顺, 吴帅. 融合主被动遥感影像的冬小麦种植面积提取研究[J]. 河南农业科学, 2023, 52(6): 160-171.
[3]	王焕鑫, 沈志豪, 刘泉, 刘金江. 基于改进MobileNetV2 模型的农作物叶片病害识别研究[J]. 河南农业科学, 2023, 52(4): 143-151.
[4]	徐夕博, 吕明荟, 王海会, 周忠科, 彭远新, 颜学文. 利用空间关联随机森林模型与遥感影像估算裸土期耕地土壤盐分含量的研究[J]. 河南农业科学, 2022, 51(5): 162-170.
[5]	冯晓, 李丹丹, 王文君, 郑国清, 刘海礁, 孙永胜, 梁山, 杨莹, 臧贺藏, 张辉. 基于轻量级卷积神经网络和迁移学习的小麦叶部病害图像识别[J]. 河南农业科学, 2021, 50(4): 174-180.
[6]	薛勇, 王立扬, 张瑜, 沈群. 基于卷积神经网络的蜜蜂采集花粉行为的识别方法[J]. 河南农业科学, 2020, 49(8): 162-172.
[7]	王利军, 郭燕, 王来刚, 贺佳, 张红利, 刘婷. GF6卫星红边波段对春季作物分类精度的影响[J]. 河南农业科学, 2020, 49(6): 165-173.
[8]	杨福芹，沙从术，冯海宽，韩瑞芳，徐平. 基于EFAST方法的苹果叶片叶绿素含量估算[J]. 河南农业科学, 2017, 46(5): 157-160.