Please wait a minute...
img

官方微信

遥感技术与应用  2022, Vol. 37 Issue (3): 564-570    DOI: 10.11873/j.issn.1004-0323.2022.3.0564
农业遥感专栏     
基于WorldView-2影像和语义分割模型的小麦分类提取
董秀春(),刘忠友,蒋怡,郭涛,李宗南()
四川省农业科学院遥感与数字农业研究所,四川 成都 610066
Winter Wheat Extraction of WorldView-2 Image based on Semantic Segmentation Method
Xiuchun Dong(),Zhongyou Liu,Yi Jiang,Tao Guo,Zongnan Li()
Institute of Remote Sensing and Digital Agriculture, Sichuan Academy of Agricultural Sciences, Chengdu 610066,China
 全文: PDF(2710 KB)   HTML
摘要:

为使用高分辨率遥感影像和深度学习语义分割模型实现快速准确的小麦种植空间信息提取,以WorldView-2遥感影像为数据源,制作尺度分别为128×128、256×256、512×512的样本数据集,对U-net和DeepLab3+语义分割模型的参数进行训练,建立小麦遥感分类模型;通过与极大似然和随机森林方法比较,检验深度学习分类效果。结果显示:①不同尺度样本训练得到的模型总体精度、Kappa系数分别在94%和0.82以上,模型精度稳定,样本尺度大小对小麦分类提取模型影响较小;②深度学习方法的小麦分类总精度和Kappa系数分别在94%和0.89以上,极大似然和随机森林则在92%和0.85以下,表明该研究建立的小麦遥感分类模型优于传统分类方法。研究结果可为高分辨率遥感影像作物种植信息的深度学习方法提取提供参考。

关键词: 高分辨率影像U?netDeepLabv3+小麦信息提取    
Abstract:

In order to realize fast and accurate extraction of winter wheat planting spatial information by using high-resolution remote sensing image and deep learning semantic segmentation model, worldView-2 remote sensing image was used as the data source to produce the sample data sets with the scales of 128×128, 256×256 and 512×512, which trained the parameters of U-net and DeepLabv3+ semantic segmentation model to establish remote sensing classification model of winter wheat. The classification effects of deep learning was tested by comparing with maximum likelihood and random forest methods. The results showed that: (1) the overall accuracy and Kappa coefficient of the models obtained by training samples of different scales were more than 94% and 0.82, and the model accuracy was stable, which indicated that the sample sizes have little influence on the semantic segmentation model of winter wheat classification. (2) The overall classification accuracy and Kappa coefficient of the deep learning methods were above 94% and 0.89, while the maximum likelihood and random forest were below 92% and 0.85, respectively. This results suggested that the remote sensing classification model of winter wheat established in this study was superior to the traditional classification methods. The results can provide the references for the deep learning methods of crop planting information extraction with high resolution remote sensing image.

Key words: High-resolution remote sensing images    U-net    DeepLabv3+    Winter wheat    Information extraction
收稿日期: 2021-01-22 出版日期: 2022-08-25
ZTFLH:  S127  
基金资助: 四川省科技计划项目”农业大数据资产化管理及智能分析应用系统“(2021YFG0028);成都市重点研发支撑计划技术创新研发项目“互联网+机器学习下的农情遥感监测方法与大数据平台”(2019-YF05-01368-SN)
通讯作者: 李宗南     E-mail: 642721838@qq.com;li_zongnan@foxmail.com
作者简介: 董秀春(1987-),女,四川南部人,硕士,助理研究员,主要从事农业遥感方面的研究。E?mail:642721838@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  
董秀春
刘忠友
蒋怡
郭涛
李宗南

引用本文:

董秀春,刘忠友,蒋怡,郭涛,李宗南. 基于WorldView-2影像和语义分割模型的小麦分类提取[J]. 遥感技术与应用, 2022, 37(3): 564-570.

Xiuchun Dong,Zhongyou Liu,Yi Jiang,Tao Guo,Zongnan Li. Winter Wheat Extraction of WorldView-2 Image based on Semantic Segmentation Method. Remote Sensing Technology and Application, 2022, 37(3): 564-570.

链接本文:

http://www.rsta.ac.cn/CN/10.11873/j.issn.1004-0323.2022.3.0564        http://www.rsta.ac.cn/CN/Y2022/V37/I3/564

图1  影像覆盖区域及主要解译标志
波段名空间分辨率/m半值波宽/nm中心波长/nm
全色0.5336.9632.2
海岸蓝2.051.8427.3
蓝色2.060.8477.9
绿色2.069.8546.2
黄色2.038.5607.8
红色2.059.3658.8
红边2.039.8723.7
近红外12.0117.8832.5
近红外22.092.5908.0
表1  WorldView-2数据光谱信息
模型

样本

尺度

总精度/%KappaIOU

耗时

/min

U-net128×12896.300.871 50.882 2148
256×25695.470.852 90.866 7129
512×51294.680.835 20.851 4144
DeepLabv3+128×12895.150.833 10.850 5169
256×25694.650.817 90.838 4163
512×51294.620.831 50.848 4162
表2  U-net和DeepLabv3+模型在不同尺度样本下的评价指标对比
分类方法错分误差漏分误差总精度/%Kappa
U-net5.303.1295.630.912 4
DeepLabv3+5.694.3594.810.896 2
随机森林9.564.3692.590.851 5
极大似然10.884.7591.610.831 9
表3  不同分类方法小麦提取精度
图2  研究区4个典型区域小麦提取结果对比
1 Liu Jia, Wang Limin, Teng Fei, et al. Crop area ground sample survey using Google Earth image-aided[J]. Transactions of the Chinese Society of Agricultural Engineering. 2015, 31(24): 157-162.刘佳, 王利民, 滕飞, 等. Google_Earth影像辅助的农作物面积地面样方调查[J]. 农业工程学报, 2015, 31(24): 157-162.
2 Zhang Peng, Hu Shougeng. Fine crop classification by remote sensing in complex planting areas based on field parcel[J]. Transactions of the Chinese Society of Agricultural Engineering, 2019, 35(20): 125-134.
2 张鹏, 胡守庚. 地块尺度的复杂种植区作物遥感精细分类[J]. 农业工程学报, 2019, 35(20): 125-134.
3 Li Zongnan, Chen zhongxin, Ren Guoye, et al. Estimation of maize lodging area based on Worldview-2 image[J]. Transactions of the Chinese Society of Agricultural Engineering. 2016, 32(2): 1-5.李宗南, 陈仲新, 任国业, 等. 基于Worldview-2影像的玉米倒伏面积估算[J]. 农业工程学报, 2016, 32(2): 1-5.
4 Li Deren, Tong Qingxi, Li Rongxin, et al. Current issues in high-resolution earth observation technology[J]. China Science: Earth Sciences, 2012, 2012, 42(6): 805-813.李德仁, 童庆禧, 李荣兴, 等. 高分辨率对地观测的若干前沿科学问题[J]. 中国科学:地球科学, 2012, 42(6): 805-813.
5 Luo Huan, Li Weiguo, Jing Yuanshu, et al. Remote sensing extraction of winter wheat planting area based on SVM[J]. Journal of Triticeae Crop, 2019, 39(4): 455-46.
5 罗桓, 李卫国, 景元书, 等. 基于SVM的县域冬小麦种植面积遥感提取[J]. 麦类作物学报, 2019, 39(4): 455-462.
6 Wang Limin, Liu Jia, Yang Lingbo, et al. Application of fandom forest method in maize-soybean accurate identification[J]. Acta Agronomica Sinica, 2018, 44(4): 101-112.
6 王利民, 刘佳, 杨玲波, 等. 随机森林方法在玉米-大豆精细识别中的应用[J]. 作物学报, 2018, 44(4): 101-112.
7 Zhang Bing. Remote sensed big data era and intelligent information extraction[J]. Geomatics and Information Science of Wuhan University, 2018, 43(12): 108-118.
7 张兵. 遥感大数据时代与智能信息提取[J]. 武汉大学学报(信息科学版), 2018, 43(12): 108-118.
8 Yang Jianyu, Zhou Zhenxu, Du Zhenrong, et al. Rural construction land extraction from high spatial resolution remote sensing image based on SegNet semantic segmentation model[J]. Transactions of the Chinese Society of Agricultural Engineering, 2019, 35(5): 259-266.
8 杨建宇, 周振旭, 杜贞容, 等. 基于SegNet语义模型的高分辨率遥感影像农村建设用地提取[J]. 农业工程学报, 2019, 35(5): 259-266.
9 Yang Q C, Liu M, Zhang Z T, et al. Mapping plastic mulched farmland for high resolution images of unmanned aerial vehicle using deep semantic segmentation[J]. Remote Sensing, 2019, 11(17): 2008. DOI:10.3390/rs11172008 .
doi: 10.3390/rs11172008
10 Guo X, Chen Y, Liu X, et al. Extraction of snow cover from high-resolution remote sensing imagery using deep learning on a small dataset[J]. Remote Sensing Letters, 2020, 11(1): 66-75. DOI:10.1080/2150704X.2019.1686548 .
doi: 10.1080/2150704X.2019.1686548
11 Gou Jiesong, Jiang Yi, Li Zongnan, et al. Aquaculture water body information extraction in the Chengdu plain based on Deeplabv3+ model[J]. Journal of Agriculture Mechanization, 2021,42(3): 105-112.
11 苟杰松, 蒋怡, 李宗南,等 .基于Deeplabv3+模型的成都平原水产养殖水体信息提取[J]. 中国农机化学报, 2021,42(3): 105-112.
12 Osco L P, Nogueira K, Marques ramos AP, et al. Semantic segmentation of citrus-orchard using deep neural networks and multispectral UAV-based imagery[J]. Precision Agriculture, 2021,22(4):1171-1188.DOI:10.1007/s11119-020-09777-5 .
doi: 10.1007/s11119-020-09777-5
13 Dong Xiuchun, Jiang Yi, Wang Si, et al. Extraction of sugarcane from Google Earth image based on the U-Net model[J]. China Agricultural Informatics, 2019, 31(6): 29-34.
13 董秀春, 蒋怡, 王思, 等. 基于U-Net的甘蔗提取方法[J]. 中国农业信息, 2019, 31(6): 29-34.
14 He Hongshu, Huang Xiaoxia, Li Hongga, et al. Water body extraction of high resolution remote sensing image based on improved U-Net Network[J]. Journal of Geo-information Science, 2020, 22(10): 2010-2022.
14 何红术, 黄晓霞, 李红旮, 等. 基于改进U-Net网络的高分遥感影像水体提取[J]. 地球信息科学学报, 2020, 22(10): 2010-2022.
15 Zheng Lei, He Zhimeng, Ding Haiyong. Research on the sparse plastic shed extraction from high resolution images using Envinet 5 Deep Learning Method[J].Remote Sensing Technology and Application,2021,36(4):908-915.
15 郑磊, 何直蒙, 丁海勇. 基于Envinet 5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.
16 Chen Qian, Zheng Lijuan, Li Xiaojuan, et al. Water body extraction from high-resolution satellite remote sensing images based on deep learning[J]. Geography and Geo-Information Science, 2019, 35(4): 43-49.
16 陈前, 郑利娟, 李小娟, 等. 基于深度学习的高分遥感影像水体提取模型研究[J]. 地理与地理信息科学, 2019, 35(4): 43-49.
17 Liu Dawei, Han Ling, Han Xiaoyon. High spatial fesolution remote sensing image classification based on deep learning[J]. Acta Optica Sinica, 2016, 36(4): 306-314.
17 刘大伟, 韩玲, 韩晓勇. 基于深度学习的高分辨率遥感影像分类研究[J]. 光学学报, 2016, 36(4): 306-314.
18 Yang Rui, Qi Yuan, Su Yang. U-Net Neural networks and its application in high resolution satellite image classification[J]. Remote Sensing Technology and Application, 2020, 35(4): 767-774.
18 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.
19 Zhang C M, Gao S, Yang X X, et al. Convolutional neural network-based remote sensing images segmentation method for extracting winter wheat spatial distribution[J]. Applied Sciences-Basel, 2018, 10(8): 1981. DOI:10.3390/app8101981 .
doi: 10.3390/app8101981
20 Du P, Samat A, Waske B, et al. Random forest and rotation forest for fully polarized sar image classification using polarimetric and spatial features[J]. ISPRS Journal of Photogrammetry and Remote Sensing,2015,105:38-53. DOI:10.1016/ j.isprsjprs.2015.03. 002 .
doi: 10.1016/ j.isprsjprs.2015.03. 002
21 Tang Wenli. Multi-scale semantic segmentation of remote sensing image based on deep residential network[D]. Wuhan: Wuhan University, 2018.
21 唐文莉. 基于深度残差网络的多尺度遥感影像语义分割研究[D]. 武汉:武汉大学, 2018.
22 Huang Dongmei, Li Minghui, Song Wei, et al. Performance of convolutional neural network and deep belief network in sea ice-water classification using SAR imagery[J]. Journal of Image and Graphics,2018,23(11) : 1720-1732.
22 黄冬梅, 李明慧, 宋巍, 等. 卷积神经网络和深度置信网络在SAR影像冰水分类的性能评估[J]. 中国图象图形学报,2018,23(11) : 1720-1732.
23 Ronneberger O, Fischer P, Brox T. U-Net:convolutional networks for fiomedical image segmentation[C]∥ International Conference on Medical Image Computing and Computer-Assisted Intervention. Springer, Cham, 2015.
24 Su Jianmin, Yang Lanxin, Jing Weipeng. U-Net based semantic segmentation method for high resolution remote sensing image[J]. Computer Engineering and Applications, 2019, 55(7): 207-213.
24 苏健民,杨岚心, 景维鹏. 基于U-Net的高分辨率遥感图像语义分割方法[J]. 计算机工程与应用, 2019, 55(7): 207-213.
25 Chen L C, Zhu Y, Papandreou G. Encoder-Decoder with atrous separable convolution for semantic image segmentation[C]∥ Computer Vision-ECCV 2018,Springer,Cham, 2018: 833-851.
26 Yang Yingpin, Wu Zhifeng, Luo Jiancheng, et al. Parcel-based crop distribution extraction using the spatiotemporal collaboration of remote sensing data[J]. Transactions of the Chinese Society of Agricultural Engineering 2021,37(7):166-174.杨颖频,吴志峰,骆剑承,等. 时空协同的地块尺度作物分布遥感提取[J]. 农业工程学报, 2021, 37(7): 166-174.
27 Yang Shuqin, Song Zhishuang, Yin Hanping, et al. Crop classification method of UVA multispectral remote sensing based on deep semantic segmentation[J]. Transactions of the Chinese Society for Agricultural Machinery, 2021,52(3): 1-13.
27 杨蜀秦,宋志双,尹瀚平,等. 基于深度语义分割的无人机多光谱遥感作物分类方法[J]. 农业机械学报, 2021,52(3): 1-13.
28 Wu Zhifeng, Luo Jiancheng, Sun Yingwei, et al. Research on precision agricultural based on the spatial- temporal remote sensing collaboration[J]. Journal of Geo-information Science, 2020, 22(4): 731-742.
28 吴志峰,骆剑承,孙营伟,等. 时空协同的精准农业遥感研究[J]. 地球信息科学学报, 2020, 22(4): 731-742.
[1] 张飞舟,刘华亮,张立福,岑奕,孙雪剑,张红明. 高分辨率光学遥感影像变化检测算法在地震灾情调查中的应用[J]. 遥感技术与应用, 2022, 37(2): 306-318.
[2] 郭擎,朱丽娅,李安,顾铃燕. 基于NDVI变化检测的滑坡遥感精细识别[J]. 遥感技术与应用, 2022, 37(1): 17-23.
[3] 李业钦,王常颖,隋毅,初佳兰,李劲华. 基于自动阈值分类的高分辨率影像海陆分界线自动提取[J]. 遥感技术与应用, 2021, 36(6): 1379-1387.
[4] 朱金明,李利伟,程钢,高连如,张兵. 基于Sentinel-2和全卷积网络的北京六环内高层建筑区提取与分析[J]. 遥感技术与应用, 2021, 36(6): 1436-1445.
[5] 梁爽,宫兆宁,赵文吉,关鸿亮,梁亚囡,陆丽,赵雪. 基于多季相Sentinel-2影像的白洋淀湿地信息提取[J]. 遥感技术与应用, 2021, 36(4): 777-790.
[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] 尹捷,周雷雷,李利伟,张雅琼,黄文江,张赫林,王岩,郑诗军,范海生,纪婵,陈俊杰,彭代亮. 多源遥感数据小麦识别及长势监测比较研究[J]. 遥感技术与应用, 2021, 36(2): 332-341.
[8] 金航,竞霞,高媛,刘良云. 遥感探测小麦条锈病严重度的GBRT模型研究[J]. 遥感技术与应用, 2021, 36(2): 411-419.
[9] 孙奇,关琳琳,焦全军,刘新杰,戴华阳. 基于植被指数融合的冬小麦生物量反演研究[J]. 遥感技术与应用, 2021, 36(2): 391-399.
[10] 夏炎,黄亮,王枭轩,陈朋弟. 基于无人机影像的烟草精细提取[J]. 遥感技术与应用, 2020, 35(5): 1158-1166.
[11] 李萌,年雁云,边瑞,白艳萍,马金辉. 基于多源遥感影像的青海云杉和祁连圆柏分类[J]. 遥感技术与应用, 2020, 35(4): 855-863.
[12] 宋德娟,魏青迪,张承明,李峰,韩颖娟,范克琦. 利用RefineNet模型提取冬小麦种植信息的方法[J]. 遥感技术与应用, 2019, 34(4): 720-726.
[13] 陈思媛, 竞霞, 董莹莹, 刘良云. 基于日光诱导叶绿素荧光与反射率光谱的小麦条锈病探测研究[J]. 遥感技术与应用, 2019, 34(3): 511-520.
[14] 王凯,赵军,朱国锋. 基于GF-1遥感数据决策树与混合像元分解模型的冬小麦种植面积早期估算[J]. 遥感技术与应用, 2018, 33(1): 158-167.
[15] 方雨晨,王培燕,田庆久. 不同覆盖度下小麦农田土壤对NDVI影响模拟分析[J]. 遥感技术与应用, 2017, 32(4): 660-666.