Please wait a minute...
img

官方微信

遥感技术与应用  2019, Vol. 34 Issue (4): 720-726    DOI: 10.11873/j.issn.1004-0323.2019.4.0720
CNN 专栏     
利用RefineNet模型提取冬小麦种植信息的方法
宋德娟1,2(),魏青迪1,2,张承明1,2(),李峰3,韩颖娟4,范克琦1
1. 山东农业大学 信息科学与工程学院, 山东 泰安 271018
2. 山东省数字农业工程技术研究中心, 山东 泰安 271018
3. 山东省气候中心, 山东 济南 250001
4. 中国气象局旱区特色农业气象灾害监测预警与风险管理重点实验室,宁夏 银川 750002
Extraction Method for Winter Wheat Planting Area based on RefineNet
Dejuan Song1,2(),Qingdi Wei1,2,Chengming Zhang1,2(),Feng Li3,Yingjuan Han4,Keqi Fan1
1. School of Information Science & Engineering, Shandong Agricultural University, Tai’an 271018, China
2. Shandong Technology and Engineering Center for Digital Agriculture, Tai’an 271018, China
3. Shandong Provincal Climate Center, Jinan 250001, China
4. Key Laboratory for Meteorological Disaster Monitoring and Early Warning and Risk Management of Characteristic Agriculturein in Arid Regions, CMA, Yinchun 750002, China
 全文: PDF(4107 KB)   HTML
摘要:

冬小麦是我国主要的粮食作物,获取精细的冬小麦种植信息对于指导农业生产具有重要的意义。通过对RefineNet模型进行扩展,形成了适宜提取冬小麦种植信息的Ex-RefineNet(Extend-RefineNet)模型,Ex-RefineNet模型由两个子模型组成,Ex-RefineNet-Edge子模型用于提取冬小麦种植区域的边缘像素,Ex-RefineNet-Inner子模型用于提取冬小麦种植区域的内部像素,使用贝叶斯模型对两个子模型的提取结果进行合并处理,形成最终提取结果。利用山东省济南市和泰安市的16幅高分2号遥感影像进行实验,将每幅影像的2/3作为训练数据,其他数据作为测试数据,选择平均精度、查全率和Kappa系数作为对比指标,Ex-RefineNet模型的结果分别为0.93、0.92、0.91,而RefineNet模型的结果分别为0.86、0.84、0.83,说明本文给出的方法在提取冬小麦种植信息方面具有较明显的优势。

关键词: 影像分割GF-2RefineNet模型贝叶斯模型冬小麦    
Abstract:

Winter wheat is the main food crop in Shandong area. It is of great significance to obtain accurate information of winter wheat planting structure for the study of food security. By expanding the RefinNet model, an Ex-RefineNet(Extend-RefineNet) suitable for extracting the information of winter wheat planting structure was formed. Ex-RefineNet consists of two submodels, the Ex-RefineNet-Edge submodel used to extract the edge pixels of the winter wheat growing area, Ex-RefineNet-Innner submodel is used to extract the inner pixels of winter wheat growing area. Finally, using Bayesian model the extraction results of the sub-model are merged to form the final extraction results. A total of 16 GF-2 images were used for comparative experiments in Jinan City and Tai'an City, Shandong Province, and 2/3 of each image was used as training data and other data were used as test data. In terms of average accuracy, total search rate, and Kapp-coefficient, results of the Ex-RefineNet model were 0.93, 0.92, and 0.91, respectively, while results of the RefineNet model were 0.86, 0.84, and 0.83, respectively. The extraction effect of the Ex-RefineNet model is significantly higher than that of the RefineNet model. Results showed that the Ex-RefineNet is advantageous to extract the structure of winter wheat.

Key words: Image Segmentation    GF-2    RefineNet Model    Bayesian Model    Winter wheat
收稿日期: 2018-08-23 出版日期: 2019-10-16
ZTFLH:  TP753  
基金资助: 国家重点研发计划项目(2017YFA0603004);国家自然科学基金项目(41471299);山东省自然科学基金项目(ZR2017MD018);中国气象局旱区特色农业气象灾害监测预警与风险管理重点实验室开放研究项目(CAMF?201701)
通讯作者: 张承明     E-mail: 15650099558@163.com;chming@sdau.edu.cn
作者简介: 宋德娟(1994— ),女,山东曲阜人,硕士研究生, 主要从事遥感信息提取研究。E?mail:15650099558@163.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  
宋德娟
魏青迪
张承明
李峰
韩颖娟
范克琦

引用本文:

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.

Dejuan Song,Qingdi Wei,Chengming Zhang,Feng Li,Yingjuan Han,Keqi Fan. Extraction Method for Winter Wheat Planting Area based on RefineNet. Remote Sensing Technology and Application, 2019, 34(4): 720-726.

链接本文:

http://www.rsta.ac.cn/CN/10.11873/j.issn.1004-0323.2019.4.0720        http://www.rsta.ac.cn/CN/Y2019/V34/I4/720

图1  模型结构图
图2  样本制作流程图
图3  山东省影像图
方法 正确识别的冬小麦占整体冬小麦的比例/% 正常识别的背景占整体背景的比例/% 漏分的冬小麦占整体冬小麦地比例/% 错分的冬小麦占整体背景地比例/% 总体分类精度/%

查全率

/%

Kapp系数
RefineNet 85 83 15 17 84 84 0.83
Ex-RefineNet 94 92 8 6 93 92 0.91
表1  实验结果评价
图4  部分实验结果图
图5  结果统计
1 Wang Limin , Liu Jia , Yao Baomin , et al . Area Change Monitoring of Winter Wheat based on Relationship Analysis of GF-1 NDVI among Different Years [J]. Transactions of the Chinese Society of Agricultural Engineer,2018,34(8):184 -191.
1 王利民, 刘佳, 姚保民,等 . 基于GF-1影像NDVI年度间相关分析的冬小麦面积变化监测[J]. 农业工程学报, 2018, 34(8):184 -191.
2 Zhang J H , Feng L L , Yao F M . Improved Maize Cultivated area Estimation over a Large Scale Combining MODIS-EVI Time Series Data and Crop Phonological Information[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2014, 94: 102-113.
3 Wu M Q , Wang C Y , Niu Z . Mapping Paddy Field in Large areas, based on Time Series Multi-sensors Data[J].Transactions of the Chinese Society of Agricultural Engineering, 2010, 26(7): 240-244.
4 Xu Qingyun , Yang Guijun , Long Huiling , et al . Crop Information Identification based on MODIS NDVI Time-series Data[J]. Transactions of the Chinese Society of Agricultural Engineering, 2014, 30(11): 134-144.
4 许青云,杨贵军,龙慧灵 .基于MODIS NDVI多年时序数据的农作物种植识别[J]. 农业工程学报,2014,30(11):134-144.
5 Mkhabela M S , Bullock P , Raj S , et al . Crop Yield Forecasting on the Canadian Prairies Using MODIS NDVI Data[J]. Agricultural Forest Meteorology, 2011, 151(3):385-393.
6 Pringle M J , Denham R J , Devadas R . Identification of Cropping Activity in Central and Southern Queensland,Australia, with the aid of MODIS MOD13Q1 Imagery[J].International Journal of Applied Earth Observation and Geoinformation, 2012, 19: 276-285.
7 Yang Y Z , Zhao P X , Hao H K , et al .Spatiotemporal Variation of Vegetation in Northern Shaanxi of Northwest China based on SPOT-VGT NDVI[J]. Chinese Journal of Applied Ecology, 2012,23(7): 1897-1903.
8 Hao W P , Mei X R , Cai X L , et al . Crop Planting Extraction based on Multi-temporal Remote Sensing Data in Northeast China[J]. Transactions of the Chinese Society of Agricultural Engineering, 2011, 27(1): 201-207.
9 Jha A , Nain A S , Ranjan R . Wheat Acreage Estimation Using Remote Sensing in Tarai Region of Uttarakhand[J]. Vegetos,2013, 26(2): 105-111.
10 Wu M Q , Yang L C , Yu B , et al . Mapping Crops Acreages based on Remote Sensing and Sampling Investigation by Multivariate Probability Proportional to Size[J]. Transactions of the Chinese Society of Agricultural Engineering, 2014, 30(2): 146-152.
11 You Jiong , Pei Zhiyuan , Wang Fei , et al . Area Extraction of Winter Wheat at County Scale based on Modified Multivariate Texture and GF-1 Satellite Images[J]. Transactions of the Chinese Society of Agricultural Engineering, 2016, 32(13): 131-139.
11 游炯, 裴志远, 王飞,等 . 基于改进多元纹理信息模型和GF-1影像的县域冬小麦面积提取[J]. 农业工程学报,2016, 32(13):131-139.
12 Ma S J , Yi X S , You J , et al . Winter Wheat Cultivated Area Estimation and Implementation Evaluation of Grain Direct Subsidy Policy based on GF-1 Imagery[J]. Transactions of the Chinese Society of Agricultural Engineering, 2016, 32(18): 169-174.
13 Wang Limin , Liu Jia , Yang Fugang ,et al .Early Recognition of Winter Wheat Area based on GF-1 Satellite[J].Transactions of the Chinese Socirty of Agricultural Engineering,2015,31(11): 194-201.
13 王利民,刘佳,杨福刚,等 .基于GF-1卫星遥感的冬小麦面积早期识别[J].农业工程学报,2015,31(11):194-201.
14 Wu M Q , Huang W J , Niu Z , et al . Fine Crop Mapping by Combining High Spectral and High Spatial resolution Remote Sensing Data in Complex Heterogeneous Areas[J]. Computers and Electronics in Agriculture,2017,139:1-9.
15 Li Xiaofeng , Zhang Shuqing , Liu Qiang ,et al . Fast Segmenation Method of High-resolution Remote Sensing Image[J].Journal of Infrared and Millimeter Waves,2009, 28(2):146-150.
15 李晓峰,张树清,刘强,等 .高分辨率遥感影像的快速分割方法[J].红外与毫米波学报,2009,28(2):146-150.
16 Luo B , Zhang L . Robust Autodual Morphological Profiles for the Classification of High-resolution Satellite Images[J]. IEEE Transactions on Geoscience and Remote Sensing, 2014, 52(2): 1451-1462.
17 Xu Xingang , Li Qiangzi , Zhou Wancun ,et al .Classification Application of QuickBird Imagery to Obtain Crop Planting Area[J].Remote Sensing Technology and Application, 2008, 23(1):17-23.
17 徐新刚,李强子,周万村,等 .应用高分辨率遥感影像提取作物种植面积[J].遥感技术与应用,2008, 23(1):17-23.
18 Cui B G , Meng A X . Fast Remote Sensing Image Segmentation Algorithm based on Nearest Neighbor Direct Graph[J].Computer Science, 2013, 40(10):274-278.
18 崔宾阁,孟翱翔 .基于最近邻有向图的遥感图像快速分割算法[J].计算机科学, 2013,40(10):274-278.
19 Wu Nailong .Maximum Entropy Method[C]//Beijing Branch Academic Annual Conference, 1987.
19 吴乃龙 . 最大熵方法[C]//北京分会学术年会,1987.
20 Huang Xin , Zhang Liangpei , Li Pingxiang . Classification of High Spatial Resolution Remotely Sensed Imagery based Upon Fusion of Multiscale Features and SVM[J]. Journal of Rmote Sensing, 2007, 11(1):48-54.
20 黄昕, 张良培, 李平湘,等 . 基于多尺度特征融合和支持向量机的高分辨率遥感影像分类[J]. 遥感学报, 2007, 11(1):48-54.
21 Wu C L , Chau K W , Fan C . Prediction of Rainfall Time Series Using Modular Artificial Neural Networks Coupled with Data-preprocessing Techniques[J]. Journal of Hydrology,2010, 389(1): 146- 167.
22 Hu F , Xia G S , Hu J , et al . Transferring Deep Convolutional Neural Networks for the Scene Classification of High-resolution Remote Sensing Imagery[J]. Remote Sensing, 2015, 7(11): 14680- 14707.
23 Chen L C , Papandreou G , Kokkinos I , et al . Semantic Image Segmentation with Deep Convolutional Nets and Fully Connected CRFs[J]. Computer Science, 2015(4):357-361.
24 Ronneberger O , Fischer P , Brox T . U-Net: Convolutional Networks for Biomedical Image Segmentation[J].arXiv: 1505.04597.
25 Badrinarayanan V , Kendall A , Cipolla R . SegNet: A Deep Convolutional Encoder-Decoder Architecture for Scene Segmentation[J]. IEEE Transactions on Pattern Analysis & Machine Intelligence, 2017(99): 2481-2495. doi: 10.1109 /TPAMI. 2016. 2644615 .
doi: 10.1109 /TPAMI. 2016. 2644615
26 Chen L C , Papandreou G , Kokkinos I , et al . DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs[J]. arXiv,2017;arXiv:1606.00915V2.
27 Lin G , Milan A , Shen C , et al . RefineNet: Multi-Path Refinement Networks for High-resolution Semantic Segmentation[J]. arxiv,2016. arxiv:1611.06612v3.
28 Pan Xuran , Yang Fan , Pan Guofeng . Extraction of Residential Areas in GF-1 Remote Sensing Images based on Improved Fully Convolutional Network[J]. Telecommunication Engineering, 2018(2):119-125 .
28 潘旭冉,杨帆,潘国峰 . 采用改进全卷积网络的“高分一号”影像居民地提取[J]. 电讯技术,2018(2):119-125.
29 Liu Wending , Tian Hongbao , Xie Jiangjian ,et al . Identification Methods for Forest Pest Areas of UAV Aerial Photography based on Fully Convolutional Networks[J]. Transactions of the Chinese Society for Agricultural, 2019,50(3):179-185.
29 刘文定,田洪宝,谢将剑,赵恩庭,张军国 .基于全卷积网络的林业航拍图像虫害区域识别方法[J].农业机械学报,2019,50(3):179-185.
30 Wang Jing , Ding Xiangqian , Wang Xiaodong , et al . Study of Near Infrared Spectrum Classification for Tobacco Leaf Position based on Deep Belief Network[J]. Infrared and Laser Engineering, 2019,48(4):0404001.
30 王静,丁香乾,王晓东,等 .基于深度信念网络的烟叶部位近红外光谱分类方法研究[J].红外与激光工程,2019,48(4):0404001.
31 Li Feiteng . The Convolutional Neural Network and Its Applications[D]. Qingdao: Dalian University of Technology,2014 [
31 李飞腾 . 卷积神经网络及其应用[D]. 青岛:大连理工大学,2014.]
32 Fu G , Liu C , Zhou R , et al . Classification for High Resolution Remote Sensing Imagery Using a Fully Convolutional Network[J]. Remote Sensing.2017, 9(6):498.doi:10.3390/rs9050498 .
doi: 10.3390/rs9050498
33 Li Yandong , Hao Zongbo , Lei Hang . Research Summary of Convolutional Neural Network [J]. Computer Application, 2016, 36(9): 2508 -2515.
33 李彦冬, 郝宗波, 雷航 . 卷积神经网络研究综述[J].计算机应用, 2016, 36(9): 2508 -2515.
[1] 蒋乔灵, 徐涵秋. GF-1 PMS2与GF-2 PMS2传感器数据的交互对比[J]. 遥感技术与应用, 2018, 33(6): 1084-1094.
[2] 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.
[3] 姜涛,朱文泉,詹培,唐珂,崔雪锋,张天一. 一种抗时序数据噪声的冬小麦识别方法研究[J]. 遥感技术与应用, 2017, 32(4): 698-708.
[4] 王苏芸,孙中昶,郭华东,申维. 基于面向对象的东营市城乡建设用地信息提取[J]. 遥感技术与应用, 2017, 32(4): 780-786.
[5] 冯艾琳,何洪林,刘利民,任小丽,张黎,葛蓉,赵凤华. 基于多源数据的禹城农田生态系统冬小麦生育期识别方法比较研究[J]. 遥感技术与应用, 2016, 31(5): 958-965.
[6] 潘一凡,张显峰,于泓峰,饶俊峰. 联合快舟一号影像纹理信息的城市土地覆盖分类[J]. 遥感技术与应用, 2016, 31(1): 194-202.
[7] 王琦,柴琳娜,赵少杰,张涛. 基于多角度微波辐射亮温数据反演冬小麦光学厚度[J]. 遥感技术与应用, 2015, 30(3): 424-430.
[8] 权文婷,周辉. HJ星数据在关中冬小麦种植面积遥感监测中的应用 [J]. 遥感技术与应用, 2014, 29(6): 930-934.
[9] 郭钇宏,王博,刘勇,杨亦宁 . 综合优度法和不一致性法的最优分割参数选择方法[J]. 遥感技术与应用, 2014, 29(3): 489-497.
[10] 余其鹏,张晓祥,梅丹丹,徐盼. 结合地籍数据的高密度城区面向对象遥感分类 [J]. 遥感技术与应用, 2014, 29(2): 344-351.
[11] 别强,何磊,赵传燕. 基于影像融合和面向对象技术的植被信息提取研究[J]. 遥感技术与应用, 2014, 29(1): 164-171.
[12] 范伟,荀尚培,杨元建,何彬方,张宏群,吴必文,高锷,孙喜波,陈磊. 星载SAR在涡阳县冬小麦测产中的应用[J]. 遥感技术与应用, 2013, 28(6): 1101-1106.
[13] 王志波,高志海,王琫瑜,徐先英,白黎娜,王红岩,吴俊君,孙 斌. 基于面向对象方法的沙化土地遥感信息提取技术研究[J]. 遥感技术与应用, 2012, 27(5): 770-777.
[14] 陈燕丽,莫伟华,莫建飞,王君华,钟仕全. 基于面向对象分类的南方水稻种植面积提取方法[J]. 遥感技术与应用, 2011, 26(2): 163-168.
[15] 白黎娜, 王琫瑜, 田 昕, 卢 颖, 杨永恬. 基于多时相ENVISat ASAR数据的冬小麦识别方法—以北京通州试验区为例[J]. 遥感技术与应用, 2010, 25(4): 458-463.