Please wait a minute...
img

官方微信

遥感技术与应用  2021, Vol. 36 Issue (4): 908-915    DOI: 10.11873/j.issn.1004-0323.2021.4.0908
遥感应用     
基于ENVINet5的高分辨率遥感影像稀疏塑料大棚提取研究
郑磊1(),何直蒙2,丁海勇2()
1.浙江省桐庐县气象局,浙江 杭州 311500
2.南京信息工程大学 遥感与测绘工程学院,江苏 南京 210044
Research on the Sparse Plastic Shed Extraction from High Resolution Images Using ENVINet 5 Deep Learning Method
Lei Zheng1(),Zhimeng He2,Haiyong Ding2()
1.Tonglu County Meteorological Bureau,Hangzhou 311500,China
2.School of Remote Sensing and Geomatics Engineering,Nanjing University of Information Science and Technology,Nanjing 210044,China
 全文: PDF(2505 KB)   HTML
摘要:

随着设施农业管理要求的提高,需要提取高分辨率遥感影像中大范围、低密度的塑料大棚空间分布信息作为农业管理和资源分配的依据。以浙江省桐庐县为研究区域,利用高分辨率遥感影像数据,对比分析不同机器学习方法提取塑料大棚的效果。ENVINet 5深度学习架构可以克服标签较少的困难,通过语义学习进行塑料大棚提取和面积估算,总体精度和Kappa系数达到97.84%和0.81;U-net深度学习网络的提取结果中,总体精度和Kappa系数为96.22%和0.79,两种深度学习方法均优于利用支持向量机进行塑料大棚提取的结果。研究表明通过深度学习方法提取高分辨率遥感影像中稀疏分布的塑料大棚有很好的效果,可以为农业经济作物管理、规划和气象保障提供支持。

关键词: 深度学习高分辨率遥感塑料大棚卷积神经网络    
Abstract:

With the increasing requirements of facility agriculture management, it is necessary to extract the spatial distribution information of plastic greenhouses with large range and low density in high-resolution remote sensing images as the basis for agricultural management and resource allocation. This study takes Tonglu County, Zhejiang Province as the study area, and uses high-resolution remote sensing images to compare and analyze the effect of extracting plastic sheds using different machine learning methods. It was found that the ENVINet5 deep learning architecture could perform plastic shed extraction and area estimation by small-sample semantic learning, and the overall accuracy and kappa coefficient reached 97.84% and 0.81; in the extraction results of U-net deep learning network, the overall accuracy and kappa coefficient were 96.22% and 0.79, which were better than the plastic shed extraction using support vector machine results. This study shows that the extraction of sparsely distributed plastic sheds in high-resolution remote sensing images by deep learning has good results and can provide support for agricultural cash crop management, planning, and weather assurance.

Key words: Deep Learning    High resolution remote sensing image    Plastic shed    CNN
收稿日期: 2020-05-27 出版日期: 2021-09-26
ZTFLH:  TP79  
基金资助: 浙江省杭州市气象局2019气象科技计划项目“基于遥感技术的桐庐县设施农业分布研究”(QX201907);国家自然科学基金项目(41571350)
通讯作者: 丁海勇     E-mail: 453034201@qq.com;hyongd@163.com
作者简介: 郑磊(1986-),男,山东荣成人,工程师,主要从事地理信息系统应用研究。 E?mail : 453034201@qq.com
服务  
把本文推荐给朋友
加入引用管理器
E-mail Alert
RSS
作者相关文章  
郑磊
何直蒙
丁海勇

引用本文:

郑磊,何直蒙,丁海勇. 基于ENVINet5的高分辨率遥感影像稀疏塑料大棚提取研究[J]. 遥感技术与应用, 2021, 36(4): 908-915.

Lei Zheng,Zhimeng He,Haiyong Ding. Research on the Sparse Plastic Shed Extraction from High Resolution Images Using ENVINet 5 Deep Learning Method. Remote Sensing Technology and Application, 2021, 36(4): 908-915.

链接本文:

http://www.rsta.ac.cn/CN/10.11873/j.issn.1004-0323.2021.4.0908        http://www.rsta.ac.cn/CN/Y2021/V36/I4/908

图1  ENVINet5架构数据处理流程图
图2  浙江省桐庐县
图3  大棚提取技术路线
参数数值
Patch Size572
Number of Epoch30
Number of Patches per Epoch400
Number of Patches per Batch16
Patch Sampling Rate5
Solid Distance
Blur DistanceMin:1,Max :3
表1  参数预设置
图4  参数测试结果
参数数值
Patch Size572
Number of Epoch30
Number of Patches per Epoch500
Number of Patches per Batch5
Patch Sampling Rate16
Solid Distance6
Blur DistanceMin:1,Max :5
表2  最优参数
参数数值
Batch Size2
Number of Epoch30
Number of Steps per Epoch1000
表3  U-net网络的训练参数
参数数值
Kernel TypeRadial Basis Function
Gamma in Kernel Function0.333
Number of Steps per Epoch Penalty Parameter100
Pyramid Levels0
Classification Probability Threshola0
表4  SVM的训练参数
图5  U-net、ENVINet5和SVM分类结果对比
图6  影像分类局部示意图
图7  后处理结果图
1 Tan Yaohua, Wang Changwei. Application of chinese High-resolution remote sensing data in urban forest resources monitoring[J]. Bulletin of Surveying and Mapping,2019(5): 113-115,154.
1 谭耀华,王长委.国产高分辨率遥感数据在城市森林资源监测中的应用[J].测绘通报,2019(5):113-115,154.
2 Qu Yang,Yuan Zhanliang,Zhao Wenzhi,et al. Crop classification based on bulti-temporal features and convolutional neural network[J]. Remote Sensing Technology and Application, 2021, 36(2): 304-313.
2 屈炀,袁占良,赵文智,等. 基于多时序特征和卷积神经网络的农作物分类[J]. 遥感技术与应用, 2021, 36(2): 304-313.
3 Zhou Jie, Fan Xiwei, Liu Yaohui.Research.on the method of UAV remote sensing in plastic greenhouse recognition[J].China Agricultural Information,2019,31(1):95-111.
3 周洁,范熙伟,刘耀辉.无人机遥感在塑料大棚识别中的方法研究[J].中国农业信息,2019,31(1):95-111.
4 Zhao Lu, Ren Hongyan, Yang Linsheng. Retrieval of agriculture greenhouse based on GF-2 remote sensing images[J].Remote Sensing Technology and Application, 2019,34(3):677-684.
4 赵璐,任红艳,杨林生.基于高分二号遥感数据的农业阳光大棚提取[J].遥感技术与应用,2019,34(3):677-684.
5 Chen Jun, Shen Runping, Li Bolun, et al. The development of plastic greenhouse index based on Logistic regression analysis[J].Remote Sensing for Land and Resources, 2019,31(3):43-50.
5 陈俊,沈润平,李博伦,等.基于Logistic回归分析的塑料大棚遥感指数构建[J].国土资源遥感,2019,31(3):43-50.
6 Shi zepeng, Ma Youhua, Wang Yujia, et al. Review on the classification methods of land use/cover based on remote sensing image[J]Chinese Agricultural Science Bulletin,2012,28(12):273-278.
6 史泽鹏,马友华,王玉佳,等.遥感影像土地利用/覆盖分类方法研究进展[J].中国农学通报,2012,28(12):273-278.
7 Wang Jue, Shi Chunyi. Machine learning research[J]. Journal of Guangxi Normal University (Natural Science Edition),2003,21(2):1-15.
7 王珏,石纯一.机器学习研究[J].广西师范大学学报(自然科学版),2003,21(2):1-15.
8 Wang Junmin, Li Yan. Research on data clustering and image segmentation based on K-means algorithm [J]. Journal of Pingdingshan University,2014,29(2):43-45.
8 王军敏,李艳.基于K均值算法的数据聚类和图像分割研究[J].平顶山学院学报,2014,29(2):43-45.
9 Yang Wei, Fang Tao, Xu Gang. Semi-supervised learning remote sensing image classification based on Naive Bayesian[J].Computer Engineering, 2010,36(20):167-169.
9 杨伟,方涛,许刚.基于朴素贝叶斯的半监督学习遥感影像分类[J].计算机工程,2010,36(20):167-169.
10 Liu Ying. Land cover remote sensing classification method based on semi supervised ensemble support vector machine[D].Beijing: Northeast Institute of Geography and Agroecology(Chinese Academy of Sciences),2013.
10 刘颖. 基于半监督集成支持向量机的土地覆盖遥感分类方法研究[D].北京:中国科学院研究生院(东北地理与农业生态研究所),2013.
11 Qi Wang Yue, Hu Hong Xiang, Xia Ping, et al. Multi-resolution remote sensing images classification and comparison analysis based on improved BP neural network[J]. Journal of Anhui Agricultural University, 2019,46(4):737-744.
11 戚王月,胡宏祥,夏萍,等.基于改进BP神经网络的多分辨率遥感图像分类及对比分析[J].安徽农业大学学报,2019,46(4):737-744.
12 Hu Naixun,Chen Tao,Zhen Na,et al. Object-oriented open pit extraction based on convolutional neural network[J]. Remote Sensing Technology and Application, 2021, 36(2): 265-274.
12 胡乃勋,陈涛,甄娜,等. 基于卷积神经网络的面向对象露天采场提取[J]. 遥感技术与应用, 2021, 36(2): 265-274.
13 Lin Na,Feng Lirong,Zhang Xiaoqing. Aircraft detection in remote sensing image based on optimized Faster-RCNN[J]. Remote Sensing Technology and Application, 2021, 36(2): 275-284.
13 林娜,冯丽蓉,张小青. 基于优化Faster-RCNN的遥感影像飞机检测[J]. 遥感技术与应用, 2021, 36(2): 275-284.
14 Yang Yanqing, Chai Xurong. The research on remote sensing image classification based on Artificial Neural Network[J]. Journal of Shanxi Normal University (Natural Science Edition),2017,31(1):94-98.
14 杨艳青,柴旭荣.基于人工神经网络法的遥感影像分类研究[J].山西师范大学学报(自然科学版),2017,31(1):94-98.
15 Zhao Jianpeng. Research on extraction method of agricultural greenhouse based on GF-2 remote sensing image[D].Langfang:North China Institute of Aerospace Engineering,2019.
15 赵建鹏. 基于GF-2遥感影像提取农业大棚方法研究[D].廊坊:北华航天工业学院,2019.
16 Li Qing,Chen Junjie,Li Qingting,et al. Detection of tailings pond in Beijing—Tianjin—Hebei region based on SSD model[J]. Remote Sensing Technology and Application, 2021, 36(2): 293-303.
16 李庆,陈俊杰,李庆亭,等. 基于SSD模型的京津冀地区尾矿库检测[J]. 遥感技术与应用, 2021, 36(2): 293-303.
17 Kussul N, Lavreniuk M, Skakun S,et al. Deep learning classification of land cover and crop types using remote sensing data[J].IEEE Geoscience and Remote Sensing Letters,2017(99):1-5. DOI:10.1109/LGRS.2017.2681128.
doi: 10.1109/LGRS.2017.2681128
18 Kavita, Bhosle, Vijaya, et al. Evaluation of deep learning CNN model for land use land cover classification and crop identification using Hyperspectral remote sensing images[J]. Journal of the Indian Society of Remote Sensing, 2019, 47(11):1949–1958.
19 Chen Min,Pan Jiawei,Li Jiangjie,et al. High resolution remote sensing image construction land detection combined with VGGNet and Mask R-CNN[J]. Remote Sensing Technology and Application,2021,36(2):256-264.
19 陈敏,潘佳威,李江杰,等. 结合VGGNet与Mask R-CNN的高分辨率遥感影像建设用地检测[J]. 遥感技术与应用, 2021, 36(2): 256-264.
20 Chen Ni,Ying Feng,Wang Jing,et al. Research on land use information extraction based on U-Net[J]. Remote Sensing Technology and Application, 2021, 36(2): 285-292.
20 陈妮,应丰,王静,李健. 基于U-Net的高分辨率遥感图像土地利用信息提取[J].遥感技术与应用,2021,36(2):285-292.
21 Zhang Minmin, Xu Heping, Wang Xiaojie, et al. Application of google tensorflow machine learning framework[J].Microcomputer Its Application,2017,36(10):58-60.
21 章敏敏,徐和平,王晓洁等.谷歌TensorFlow机器学习框架及应用[J].微型机与应用,2017,36(10):58-60.
22 RRonneberger O, Fischer P, Brox T. U-Net: convolutional networks for biomedical image segmentation[C]∥ Proceedings of 2015 IEEE International Conference on Computer Vision (ICCV),2015. Santiago,Chile: IEEE,2015:1520-1528.
23 Wu Guangming, Chen Qi, Shibasaki Ryosuke, et al. High precision building detection from aerial imagery using a U-Net like convolutional architecture[J]. Acta Geodaetica et Cartographica Sinica, 2018,47 (6): 864-872.
23 伍广明,陈奇, Shibasaki Ryosuke,等,基于U型卷积神经网络的航空影像建筑物检测[J].测绘学报,2018,47(6) :864-872.
24 Long J, Shelhamer E, Darrell T. Fully convolutional networks for semantic segmentation[C]∥ Proceedings of the IEEE conference on computer vision and pattern recognition. 2015: 3431-3440.
25 Chen L C,Papandreou G,Kokkinos I,et al. DeepLab: semantic image segmentation with deep convolutional nets, atrous convolution, and fully connected CRFs[J]. IEEE Transactions on Pattern Analysis & Machine Intelligence, 2016, 40(4):834-848.
26 Yu F, Koltun V. Multi-scale context aggregation by dilated convolutions[C]∥ The International Conference on Learning Representations (ICLR), 2016, arXiv:.
27 Song Can, Wu Jin, Zhu Lei, et al. Greenhouse segmentation of remote sensing images based on Deep Learning[J]. Microelectronics t &Computer, 2021,38(1):51-56.
27 宋灿,吴谨,朱磊,等,针对遥感图像大棚提取的深度学习模型研究[J].微电子学与计算机,2021,38(1) :51-56.
[1] 路春燕,雷依凡,苏颖,黄雨菲,刘明月,贾明明. 基于面向对象—深度学习的闽东南低海拔海岸带地区湿地动态遥感分析[J]. 遥感技术与应用, 2021, 36(4): 713-727.
[2] 高子为,孙伟伟,程朋根,杨刚,孟祥超. 融合高分辨率遥感影像和POI数据的多特征潜在语义信息用于识别城市功能区[J]. 遥感技术与应用, 2021, 36(3): 618-626.
[3] 刘钊,赵桐,廖斐凡. 基于卷积神经网络预测结果的缝隙修复算法研究[J]. 遥感技术与应用, 2021, 36(2): 247-255.
[4] 陈敏,潘佳威,李江杰,徐璐,刘加敏,韩健,陈奕云. 结合VGGNet与Mask R-CNN的高分辨率遥感影像建设用地检测[J]. 遥感技术与应用, 2021, 36(2): 256-264.
[5] 胡乃勋,陈涛,甄娜,牛瑞卿. 基于卷积神经网络的面向对象露天采场提取[J]. 遥感技术与应用, 2021, 36(2): 265-274.
[6] 陈妮,应丰,王静,李健. 基于U-Net的高分辨率遥感图像土地利用信息提取[J]. 遥感技术与应用, 2021, 36(2): 285-292.
[7] 林娜,冯丽蓉,张小青. 基于优化Faster-RCNN的遥感影像飞机检测[J]. 遥感技术与应用, 2021, 36(2): 275-284.
[8] 李庆,陈俊杰,李庆亭,李柏鹏,卢凯旋,昝露洋,陈正超. 基于SSD模型的京津冀地区尾矿库检测[J]. 遥感技术与应用, 2021, 36(2): 293-303.
[9] 屈炀,袁占良,赵文智,陈学泓,陈家阁. 基于多时序特征和卷积神经网络的农作物分类[J]. 遥感技术与应用, 2021, 36(2): 304-313.
[10] 徐瑾昊,冯敏,王建邦,冉有华,祁元,杨联安,李新. 基于高分遥感数据和深度学习的石冰川自动提取研究[J]. 遥感技术与应用, 2020, 35(6): 1329-1336.
[11] 张永宏,曹海啸,阚希. 基于FY-4A/AGRI时空特征融合的新疆地区积雪判识[J]. 遥感技术与应用, 2020, 35(6): 1337-1347.
[12] 杨瑞,祁元,苏阳. 深度学习U-Net方法及其在高分辨卫星影像分类中的应用[J]. 遥感技术与应用, 2020, 35(4): 767-774.
[13] 王卓,闫浩文,禄小敏,冯天文,李亚珍. 一种改进U-Net的高分辨率遥感影像道路提取方法[J]. 遥感技术与应用, 2020, 35(4): 741-748.
[14] 连喜红,祁元,王宏伟,张金龙,杨瑞. 基于面向对象的青海湖环湖区居民地信息自动化提取[J]. 遥感技术与应用, 2020, 35(4): 775-785.
[15] 张峰极,吴艳兰,姚雪东,梁泽毓. 基于改进DenseNet网络的多源遥感影像露天开采区智能提取方法[J]. 遥感技术与应用, 2020, 35(3): 673-684.