Please wait a minute...
img

官方微信

遥感技术与应用  2020, Vol. 35 Issue (5): 1226-1236    DOI: 10.11873/j.issn.1004-0323.2020.5.1226
遥感应用     
基于多时相遥感观测的板栗林分布提取研究
陈继龙1,2(),魏雪馨1,2,刘洋1(),闵庆文1,刘荣高1,张文林3,郭春梅3
1.中国科学院地理科学与资源研究所,北京 100101
2.中国科学院大学,北京 100049
3.宽城满族自治县农业农村局,河北 宽城 067600
Extraction of Chestnut Forest Distribution based on Multi-temporal Remote Sensing Observations
Jilong Chen1,2(),Xuexin Wei1,2,Yang Liu1(),Qingwen Min1,Ronggao Liu1,Wenlin Zhang3,Chunmei Guo3
1.Institute of Geographic Sciences and Natural Resources,Chinese Academy of Sciences,Beijing 100101,China
2.University of Chinese Academy of Sciences,Beijing 100049,China
3.Kuancheng County Bureau of Agriculture and Rural Affairs,Kuancheng 067600,China
 全文: PDF(5262 KB)   HTML
摘要:

板栗林在欧亚、北美等地广泛分布,具有良好的生态价值和经济效益。我国板栗产量居世界首位,是重要的经济树种。使用遥感影像建立板栗林空间分布提取方法能够为其科学管理和高效经营提供定量数据,但树种分类是遥感分类的难点,并且针对板栗林的遥感提取研究较少。以河北省宽城满族自治县为研究区,结合MODIS高时间分辨率特征和Landsat数据较高空间分辨率的特征,研究板栗林提取的最佳时相以及分类特征,并采用多时相观测基于支持向量机算法实现板栗林的提取。结果表明:①4月至6月各地类光谱差异最大,是板栗林提取的关键物候期;②蓝、绿、红、近红外和短波红外波段地表反射率是分类的有效波段,NDI、NDVI、NDWI、RSI和RVI等植被指数增强了植被信息,是板栗林提取的有效分类特征;③单一时相板栗林分类中,生长季前期6月精度最高,生长季后期9月次之,非生长季1月分类结果较差;④结合生长季6月、9月和非生长季1月遥感影像的分类精度最佳,板栗林制图和用户精度分别为89.90%和87.25%。与林业局板栗林面积统计数据相比,精度可达93.45%。

关键词: 板栗林物候支持向量机季节曲线遥感    
Abstract:

Chestnut forest is widely distributed in Europe, Asia and North America, and provides notable ecological and economic benefits. Chestnut is an important economic tree species in China, with its production ranks first in the world. The method of extracting the spatial distribution of chestnut forest based on remote sensing image can provide quantitative data for its scientific management. However, the classification of tree species is difficult in remote sensing classification and there are few reports on extraction of chestnut forest based on remote sensing data. Taking Kuancheng county of Hebei province as the research area, this paper integrates MODIS high temporal resolution observations and Landsat high spatial resolution images to select the optimal time phase and classification characteristics, and then chestnut forest was mapped based on multi-temporal Landsat OLI images using Support Vector Machine. The results showed that: (1)the spectral differences were the largest among different vegetation types from April to June, followed by September, which are the key phenological periods for chestnut forest extraction, and January helps to distinguish chestnut forest and evergreen forest; (2)Reflectances in blue, green, red, near-infrared and short-wave infrared bands are the effective bands of classification. NDI, NDVI, NDWI, RSI and RVI vegetation indexes enhance the information of vegetation growth state and coverage, which are effective classification features; (3)In the classification with single temporal image, the accuracy was highest in early growing season in July, followed by late growing season in September, and poor in non-growing season in January; (4)Integrating the images of June, September and January perform best, and the mapping accuracy and user accuracy of chestnut are up to 89.90% and 87.25%. The accuracy can reach 93.45% when compared with the statistics data of chestnut forest area of local forestry bureau in 2018.

Key words: Chestnut    Phenology    Support vector machine    Seasonal curve    Remote sensing
收稿日期: 2019-07-08 出版日期: 2020-11-26
ZTFLH:  TP79  
基金资助: 国家重点研发计划项目(2019YFA0606601);中国科学院战略性先导科技专项子课题(XDA19080303);中国科学院青年创新促进会项目(2019056)
通讯作者: 刘洋     E-mail: chenjl.196@igsnrr.ac.cn;liuyang@igsnrr.ac.cn
作者简介: 陈继龙(1997-),男,河南信阳人,硕士研究生,主要从事定量遥感研究。E?mail: chenjl.196@igsnrr.ac.cn
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  
陈继龙
魏雪馨
刘洋
闵庆文
刘荣高
张文林
郭春梅

引用本文:

陈继龙,魏雪馨,刘洋,闵庆文,刘荣高,张文林,郭春梅. 基于多时相遥感观测的板栗林分布提取研究[J]. 遥感技术与应用, 2020, 35(5): 1226-1236.

Jilong Chen,Xuexin Wei,Yang Liu,Qingwen Min,Ronggao Liu,Wenlin Zhang,Chunmei Guo. Extraction of Chestnut Forest Distribution based on Multi-temporal Remote Sensing Observations. Remote Sensing Technology and Application, 2020, 35(5): 1226-1236.

链接本文:

http://www.rsta.ac.cn/CN/10.11873/j.issn.1004-0323.2020.5.1226        http://www.rsta.ac.cn/CN/Y2020/V35/I5/1226

图1  宽城县地理位置和Landsat8 OLI标准假彩色合成影像(2018年6月4日)
波段MOD09A1波长/nmLandsat8 OLI波长/nm
Band 1620~670435~451
Band 2841~876452~512
Band 3459~479533~590
Band 4545~565636~673
Band 51 230~1 250851~879
Band 61 628~1 6521 566~1 651
Band 72 105~2 1552 107~2 294
表1  MOD09A1和Landsat8 OLI波段信息
地物类型Google Earth高分辨率影像示例训练样本多边形数/像元数/个验证样本多边形数/像元数/个多边形/像元数合计/个
水域23/1 35611/439

34/1 795

建设用地28/1 25217/537

45/1 789

耕地91/1 46433/515

124/1 979

板栗林117/2 33325/495

142/2 828

落叶林地86/2 35919/593

105/2 952

常绿林99/2 21838/676

137/2 894

合计(个)444/10 982143/3 255587/14 237
表2  Google Earth获得的训练样本和验证样本
图2  典型地物训练样本和验证样本分布
图3  方法流程
图4  不同地物类型MODIS时间序列光谱反射率及植被指数曲线
图5  各地类样本点分类特征直方图
实验设计分类时相分类特征
实验11月绿、红、近红外、短波红外,NDI、NDVI、NDWI、RSI
实验26月蓝、绿、红、近红外、短波红外,NDI、NDVI、NDWI、RSI、RVI
实验39月绿、红、近红外、短波红外,NDI、NDVI、NDWI、RVI
实验41月、6月

1月:绿、红、近红外、短波红外,NDI、NDVI、NDWI、RSI

6月:蓝、绿、红、近红外、短波红外,NDI、NDVI、NDWI、RSI、RVI

实验51月、9月

1月:绿、红、近红外、短波红外,NDI、NDVI、NDWI、RSI

9月:绿、红、近红外、短波红发外,NDI、NDVI、NDWI、RVI

实验66月、9月

6月:蓝、绿、红、近红外、短波红外,NDI、NDVI、NDWI、RSI、RVI

9月:绿、红、近红外、短波红外,NDI、NDVI、NDWI、RVI

实验71月、6月和9月

1月:NDI、NDVI、NDWI、RSI

6月:蓝、绿、红、近红外、短波红外,NDI、NDVI、NDWI、RSI、RVI

9月:绿、红、近红外、短波红外,NDI、NDVI、NDWI、RVI

表3  实验设计
板栗林耕地落叶林地常绿林建设用地水体总精度/%Kappa系数
实验1制图精度/%66.6783.8886.3410090.5093.8587.500.849 3
用户精度/%68.7580.6084.0799.1294.5594.93
实验2制图精度/%92.1290.1091.4097.4992.7497.0493.550.922 3
用户精度/%83.2191.3492.4997.4998.4298.84
实验3制图精度/%79.3983.8886.6897.6391.0698.1889.680.875 7
用户精度/%74.5787.1086.2496.0798.3995.35
实验4制图精度/%89.4989.7192.9299.2693.8597.0493.920.926 7
用户精度/%82.8091.6793.0798.6897.4999.77
实验5制图精度/%79.3988.5490.8999.5694.6095.9091.860.901 9
用户精度/%79.2389.2487.6498.8396.3999.06
实验6制图精度/%90.1089.7194.9497.4996.8397.4994.560.934 5
用户精度/%87.2895.4590.6697.2098.4898.85
实验7制图精度/%89.9087.2595.7897.7896.8397.4994.750.936 7
用户精度/%89.7194.8791.3297.9398.4898.85
表4  分类结果混淆矩阵
分类器板栗林总体精度 /%Kappa系数
制图精度/%用户精度/%
支持向量机89.9087.2594.750.936 7
随机森林89.9085.9194.750.936 7
人工神经网络93.7480.8493.950.927 1
最小距离56.1664.0683.590.802 1
马氏距离92.7381.8094.040.928 2
表5  不同分类器分类结果比较
图6  宽城土地利用和板栗林分布图
板栗林落叶林地常绿林耕地
统计数据/km2480.021 094.61132.80132.92
分类结果/km2448.591 044.98111.40138.48
分类精度93.45%95.47%83.89%95.82%
表6  森林和耕地的分类结果与当地统计数据的对比
1 Wikipedia. Chestnut[DB/OL]. https://en.wikipedia.org/wiki/Chestnut, 2017-12-29, 2019-07-17.
2 Zhang Yuhe, Liang Weijian. Orchards of China[M]. Beijing: China Forestry Press, 2005.
2 张宇和,梁维坚.中国果树志[M].北京:中国林业出版社,2005.
3 Zhu Chunbo, Li Shifeng. Model of Economic Forest Interplanting to Prevent Soil Erosion[J]. Application Technology of Soil and Water Conservation, 2011(1):18-20.
3 朱春波,李世锋.经济林套种防治水土流失模式[J].水土保持应用技术,2011(1):18-20.
4 Yang Yongqiong, Wu Bozhi. Study on Intercropping Effects of Intercropping Methods[J]. Chinese Agricultural Bulletin, 2007, 23(11):192-196.
4 杨友琼,吴伯志.作物间套作种植方式间作效应研究[J].中国农学通报,2007,23(11):192-196.
5 Dong J, Xiao X, Chen B, et al. Mapping Deciduous Rubber Plantations through Integration of PALSAR and Multi-temporal Landsat Imagery[J]. Remote Sensing of Environment, 2013,134(Complete):392-402. doi: 10.1016/j.rse.2013. 03.014.
doi: 10.1016/j.rse.2013. 03.014
6 Dong J, Xiao X, Sheldon S, et al. Mapping Tropical Forests and Rubber Plantations in Complex Landscapes by Integrating PALSAR and MODIS Imagery[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2012, 74(11):20-33. doi: 10.1016/j.isprsjprs.2012.07.004.
doi: 10.1016/j.isprsjprs.2012.07.004
7 Fan H, Fu X, Zhang Z, et al. Phenology-based Vegetation Index Differencing for Mapping of Rubber Plantations Using Landsat OLI Data[J]. Remote Sensing, 2015, 7(5):6041-6058. doi: 10.3390/rs70506041.
doi: 10.3390/rs70506041
8 Wang L, Sousa W P, Gong P, et al. Comparison of IKONOS and QuickBird Images for Mapping Mangrove Species on the Caribbean Coast of Panama[J]. Remote Sensing of Environment,2004,91(3-4):432-440. doi: 10.1016/j.rse.2004. 04.005.
doi: 10.1016/j.rse.2004. 04.005
9 Giri C, Ochieng E, Tieszen L L, et al. Status and Distribution of Mangrove Forests of the World Using Earth Observation Satellite Data[J]. Global Ecology and Biogeography, 2011,20(1):154-159.doi: 10.1111/j.1466-8238.2010. 00584.x.
doi: 10.1111/j.1466-8238.2010. 00584.x
10 Li M, Li C, Jiang H, et al. Tracking Bamboo Dynamics in Zhejiang, China, Using Time-series of Landsat Data from 1990 to 2014[J]. International Journal of Remote Sensing, 2016, 37(7):1714-1729. doi: 10.1080/01431161.2016.1165885.
doi: 10.1080/01431161.2016.1165885
11 Liu C, Xiong T, Gong P, et al. Improving Large-scale Moso Bamboo Mapping based on Dense Landsat Time Series and Auxiliary Data: A Case Study in Fujian Province, China[J]. Remote Sensing Letters, 2018, 9(1):1-10. doi: 10.1080/2150704X.2017.1378454.
doi: 10.1080/2150704X.2017.1378454
12 Arbelo M, Marchetti F, Hernandez-Leal P A, et al. Multitemporal WorldView Satellites Imagery for Mapping Chestnut Trees[C]∥ Society of Photo-optical Instrumentation Engineers. Society of Photo-Optical Instrumentation Engineers (SPIE) Conference Series, 2017. doi: 10.1117/12.2278655.
doi: 10.1117/12.2278655
13 Verlic, A, Đuric, N, Kokalj, Z, et al. Tree Species Classification Using WorldView-2 Satellite Images and Laser Scanning Data in A Natural Urban Forest[J]. Sumarski list, 2014, 138(9-10):477-488.
14 Hmimina G, Dufrêne Eric , Pontailler J Y, et al. Evaluation of the Potential of MODIS Satellite Data to Predict Vegetation Phenology in Different Biomes: An Investigation Using Ground-based NDVI Measurements[J]. Remote Sensing of Environment,2013,132(6):145-158.doi:10.1016/j.rse.2013. 01.010.
doi: 10.1016/j.rse.2013. 01.010
15 Clerici, Nicola, Weissteiner, et al. Exploring the Use of MODIS NDVI-based Phenology Indicators for Classifying Forest General Habitat Categories[J]. Remote Sensing, 2012, 4(6):1781-1803. doi: 10.3390/rs4061781.
doi: 10.3390/rs4061781
16 Liu R, Liu Y. Generation of New Cloud Masks from MODIS Land Surface Reflectance Products[J]. Remote Sensing of Environment, 2013, 133(Complete):21-37. doi: 10.1016/j.rse.2013.01.019.
doi: 10.1016/j.rse.2013.01.019
17 Foga S, Scaramuzza P L, Guo S, Zhu Z, et al. Cloud Detection Algorithm Comparison and Validation for Operational Landsat Data Products[J]. Remote Sensing of Environment, 2017, 194(none):379-390. doi: 10.1016/j.rse.2017.03.026.
doi: 10.1016/j.rse.2017.03.026
18 Loveland T R, Dwyer J L. Landsat: Building a Strong Future[J]. Remote Sensing of Environment, 2012, 122(1):22-29. doi: 10.1016/j.rse.2011.09.022.
doi: 10.1016/j.rse.2011.09.022
19 Zheng Wei, Zeng Zhiyuan. A Review on Methods of Atmospheric Correction for Remote Sensing Images[J]. Remote Sensing Information, 2004(4):66-70.
19 郑伟,曾志远.遥感图像大气校正方法综述[J].遥感信息,2004(4):66-70.
20 HueteA, Didan K, Miura T, et al. Overview of the Radiometric and Biophysical Performance of the MODIS Vegetation Indices[J]. Remote Sensing of Environment, 2002, 83(1-2):195-213. doi: 10.1016/S0034-4257(02)00096-2.
doi: 10.1016/S0034-4257(02)00096-2
21 Mcdonald A J, Gemmell F M, Lewis P E. Investigation of the Utility of Spectral Vegetation Indices for Determining Information on Coniferous Forests[J]. Remote Sensing of Environment, 1998, 66(3):250-272. doi: 10.1016/S0034-4257(98)00057-1.
doi: 10.1016/S0034-4257(98)00057-1
22 Li C, Wang J, Wang L, et al. Comparison of Classification Algorithms and Training Sample Sizes in Urban Land Classification with Landsat Thematic Mapper Imagery[J]. Remote Sensing, 2014, 6(2):964-983. doi: 10.3390/rs6020964.
doi: 10.3390/rs6020964
23 Dalponte M, Bruzzone L, Gianelle D. Tree Species Classification in the Southern Alps based on the Fusion of very High Geometrical Resolution Multispectral/Hyperspectral Images and LiDAR Data[J]. Remote Sensing of Environment, 2012, 123(none):258-270. doi: 10.1016/j.rse.2012.03.013.
doi: 10.1016/j.rse.2012.03.013
24 Brabant C, Alvarez-Vanhard E, Laribi A, et al. Comparison of Hyperspectral Techniques for Urban Tree Diversity Classification[J]. Remote Sensing, 2019, 11(11). doi: 10.3390/rs11111269.
doi: 10.3390/rs11111269
25 Ballanti L, Blesius L, Hines E, et al. Tree Species Classification Using Hyperspectral Imagery: A Comparison of Two Classifiers[J]. Remote Sensing, 2016, 8(6). doi: 10.3390/rs8060445.
doi: 10.3390/rs8060445
26 Camps-Valls G, Bruzzone L. Kernel Methods for Remote Sensing Data Analysis Multi-Temporal Image Classification with Kernels[J]. 2009, 10.1002/9780470748992:125-145. doi: 10.1002/9780470748992.ch6.
doi: 10.1002/9780470748992.ch6
27 Chen Y L, Zhao S, Xie Z L, et al. Mapping Multiple Tree Species Classes Using a Hierarchical Procedure with Optimi-zed Node Variables and Thresholds based on High Spatial Resolution Satellite Data[J]. Giscience and Remote Sensing, 2020,57(4):526-542. doi:10.1080/15481603.2020.1742459.
doi: 10.1080/15481603.2020.1742459
28 Fang F, McNeil B E, Warner T A, et al. Discriminating Tree Species at Different Taxonomic Levels Using Multi-temporal WorldView-3 Imagery in Washington DC, USA[J]. Remote Sensing of Environment,2020:246.doi:10.1016/j.rse. 2020.111811.
doi: 10.1016/j.rse. 2020.111811
29 Xu K J, Tian Q J, Zhang Z Y, et al. Tree Species (Genera) Identification with GF-1 Time-Series in a Forested Landscape, Northeast China[J]. Remote Sensing, 2020, 12(10). doi: 10.3390/rs12101554.
doi: 10.3390/rs12101554
30 Feng B K, Zheng C, Zhang W Q, et al. Analyzing the Role of Spatial Features when Cooperating Hyperspectral and LiDAR Data for the Tree Species Classification in a Subtropical Plantation Forest Area[J]. Journal of Applied Remote Sensing, 2020, 14(2). doi: 10.1117/1.JRS.14.022213.
doi: 10.1117/1.JRS.14.022213
31 Miyoshi G T, Arruda M D, Osco L P, et al. A Novel Deep Learning Method to Identify Single Tree Species in UAV-based Hyperspectral Images[J]. Remote Sensing, 2020, 12(8). doi: 10.3390/rs12081294.
doi: 10.3390/rs12081294
32 Sothe C, De Almeida C M, Schimalski M B, et al. Comparative Performance of Convolutional Neural Network, Weighted and Conventional Support Vector Machine and Random Forest for Classifying Tree Species Using Hyperspectral and Photogrammetric Data[J]. Giscience and Remote Sensing, 2020, 57(3):369-394. doi: 10.1080/15481603.2020.1712102.
doi: 10.1080/15481603.2020.1712102
33 Du Huaqiang, Zhou Guomo, Ge Hongli, et al. Extraction of Bamboo Forest Remote Sensing Information based on TM Data[J]. Journal of Northeast Forestry University, 2008, 36(3):35-38.
33 杜华强,周国模,葛宏立,等.基于TM数据提取竹林遥感信息的方法[J].东北林业大学学报,2008,36(3):35-38.
34 Guo Baohua, Fan Shaohui, Guan Fengying, et al. Research on Bamboo Forest Information Extraction based on Support Vector Machine[J]. Journal of Northwest Forestry University, 2014, 29(2):80-84.
34 郭宝华,范少辉,官凤英,等.基于支持向量机的竹林信息提取研究.西北林学院学报,2014,29(2):80-84.
[1] 桑宇星,刘刚,江聪,任舒艳,朱再春. 近30 a中国叶面积指数变化趋势的不确定性评估[J]. 遥感技术与应用, 2020, 35(5): 1028-1036.
[2] 刘刚,桑宇星,赵茜,江聪,朱再春. 生态系统模型模拟中国叶面积指数变化趋势及驱动因子的不确定性[J]. 遥感技术与应用, 2020, 35(5): 1037-1046.
[3] 郭利彪,刘桂香,运向军,张勇,孙世贤. 基于数据机理的植被叶面积指数遥感反演研究[J]. 遥感技术与应用, 2020, 35(5): 1047-1056.
[4] 沈谦,朱长明,张新,黄巧华,杨程子. 1992~2013我国干旱区城市不透水遥感制图与扩张过程分析[J]. 遥感技术与应用, 2020, 35(5): 1178-1186.
[5] 刘美,杜国明,于凤荣,匡文慧. 哈尔滨城乡梯度建设用地结构变化及不透水面遥感监测分析[J]. 遥感技术与应用, 2020, 35(5): 1206-1217.
[6] 郭梦辉,季亚南,柯樱海,陈少辉. 土地利用变化下北京市热通量的时空演变[J]. 遥感技术与应用, 2020, 35(5): 1218-1225.
[7] 严欣荣,官凤英. 竹资源遥感监测研究进展[J]. 遥感技术与应用, 2020, 35(4): 731-740.
[8] 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.
[9] 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.
[10] 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.
[11] 韩天,潘小多,王旭峰,黄广辉,韦海宁. 遥感资料在WRF-Chem沙尘模拟中的应用[J]. 遥感技术与应用, 2020, 35(4): 808-819.
[12] 魏石梅, 潘竟虎, 妥文亮. 2015年中国PM2.5浓度遥感估算与时空分布特征[J]. 遥感技术与应用, 2020, 35(4): 845-854.
[13] 潘灼坤,胡月明,王广兴,刘吼海,刘江,李波,樊舒迪. 对遥感在城市更新监测应用中的认知和思考[J]. 遥感技术与应用, 2020, 35(4): 911-923.
[14] 李智礼,匡文慧,张澍. 近70 a天津主城区城市土地利用/覆盖变化遥感监测与时空分析[J]. 遥感技术与应用, 2020, 35(3): 527-536.
[15] 唐希颖,崔耀平,李楠,付一鸣,刘小燕,闰亚迪. 2000~2015年北京市土地利用强度及其辐射反馈评估[J]. 遥感技术与应用, 2020, 35(3): 587-595.