Please wait a minute...
img

官方微信

遥感技术与应用  2022, Vol. 37 Issue (3): 580-588    DOI: 10.11873/j.issn.1004-0323.2022.3.0580
农业遥感专栏     
基于波段反射率日均增量的花生干旱灾情遥感评估
齐文栋1,李志刚2,顾晓鹤3()
1.北京尚德智汇科技有限公司,北京 100088
2.中国太平保险集团有限责任公司,广东 深圳 518046
3.北京农业信息技术研究中心,北京 100097
Remote Sensing Evaluation of Peanut Drought Disaster based on Daily Average Increment of Multi-band Reflectivity
Wendong Qi1,Zhigang Li2,Xiaohe Gu3()
1.Beijing Sun-Golden Technology Co. ,Ltd. Beijing 100088,China
2.Ping An Property Insurance Co. ,Ltd. Shenzhen 518046,China
3.Beijing Research Center for Information Technology in Agriculture,Beijing 100097,China
 全文: PDF(3269 KB)   HTML
摘要:

干旱是影响花生产量的主要气象因素之一。开展花生干旱灾情遥感评估对于产量估算、防灾减灾和保险理赔具有重要意义。当前花生旱灾遥感评估主要依赖于光谱指数变化信息,容易受不同地区生育进程干扰,限制了光谱指数方法的普适性。研究在多时相Sentinel-2遥感影像和野外实测样本的支持下,分析时序波段反射率日均增量信息与花生干旱受灾程度之间的内在联系,利用决策树、随机森林、逻辑回归等方法对花生干旱等级进行分类,并以总体精度和Kappa系数评价各种方法的精度。结果表明:单一波段的近红外反射率日均增量对花生受灾情况的指示性较强。多光谱波段组合方式对花生旱灾程度的指示性均优于单个波段,其中红波段、蓝波段、近红外光谱波段反射率日均增量组合的指示性最强,整体精度达到89.93%,Kappa系数0.847 1。与逻辑回归和决策树算法相比,随机森林算法对花生旱灾评估精度最高。在旱情等级最优时相组合分析中,利用花生生长旺盛期(7月~8月)的多波段反射率日均增量信息,灾情等级遥感识别的总体精度可达88.62%,Kappa系数为0.827 4。说明基于生长旺盛期时序多波段反射率日均增量的干旱灾情遥感评估方法能有效提取花生受灾范围与灾情严重程度。

关键词: 花生旱灾多光谱遥感    
Abstract:

Drought is one of the main meteorological factors affecting peanut yield. Remote sensing assessment of peanut drought disaster is of great significance for yield estimation, disaster prevention and mitigation, and insurance claims. At present, the remote sensing evaluation of peanut drought mainly depends on the change information of spectral index, which is easily disturbed by the growth process in different regions, which limits the universality of spectral index method. Supported by multi temporal Sentinel-2 remote sensing images and field samples, this study analyzed the internal relationship between the daily average reflectance increment information of time-series bands and the drought disaster degree of peanut. Decision tree, random forest, logistic regression and other methods were used to classify the Drought Grades of peanut, and the overall accuracy and kappa coefficient were used to evaluate the accuracy of various methods. The results showed that the daily average increment of NIR reflectance in a single band was a strong indicator of peanut disaster. The results showed that the combination of multi spectral bands was better than single band in indicating drought degree of peanut, and the combination of red band, blue band and near infrared spectral band had the strongest indication, with the overall accuracy of 89.93% and Kappa coefficient of 0.847 1. Compared with Logistic regression and decision tree algorithm, random forest algorithm has the highest accuracy in drought assessment of peanut. In the analysis of the optimal time combination of drought grade, using the multi band daily average reflectance increment information of peanut growth peak period (July and August), the overall accuracy of disaster grade remote sensing recognition can reach 88.62%, and the Kappa coefficient is 0.827 4. The results show that the drought disaster assessment method based on multi band reflectance daily increment in the growing period can effectively extract the disaster area and severity of peanut.

Key words: Peanut    Drought    Multispectral remote sensing    Yield estimation
收稿日期: 2021-05-25 出版日期: 2022-08-25
ZTFLH:  TP79  
通讯作者: 顾晓鹤     E-mail: guxh@nercita.org.cn
作者简介: 齐文栋(1987-),男,山西大同人,硕士,中级工程师,主要从事农业遥感应用研究。E?mail:qiwendong@sun?golden.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  
齐文栋
李志刚
顾晓鹤

引用本文:

齐文栋,李志刚,顾晓鹤. 基于波段反射率日均增量的花生干旱灾情遥感评估[J]. 遥感技术与应用, 2022, 37(3): 580-588.

Wendong Qi,Zhigang Li,Xiaohe Gu. Remote Sensing Evaluation of Peanut Drought Disaster based on Daily Average Increment of Multi-band Reflectivity. Remote Sensing Technology and Application, 2022, 37(3): 580-588.

链接本文:

http://www.rsta.ac.cn/CN/10.11873/j.issn.1004-0323.2022.3.0580        http://www.rsta.ac.cn/CN/Y2022/V37/I3/580

图1  研究区位置分布图审图号:GS(2019)1822号
图2  研究区月累计降雨量
图3  研究区耕地地块
物候播种期出苗期拔节期开花期幼果期饱果期成熟期
时间6月上旬6月中旬7月上旬7月下旬8月中旬8月底~9月初国庆前后
多波段(变化)上升期上升期上升期最大值下降期下降期
获取影像日期

7月2日(t1

7月7日(t2

8月11日(t3

8月16日(t4

8月31日(t5

9月5日(t6

表1  花生种植物候信息
图4  外业调查数据等级分布图
受灾程度未受灾轻度中度重度绝产
对应减产程度<10%10%~30%30%~50%50%~80%80%
调查点个数734553317
表 2  野外调查点数量
图5  研究区2019年花生种植分布图

受灾程度

样本

未受灾

轻度

受灾

中度

受灾

重度

受灾

绝产
训练样本数量54551892304115
测试样本数量54551892304115
表3  训练样本与测试样本像元数
图6  花生旱灾灾情等级多波段日均增量图
分类器绿近红外绿+近红外蓝+近红外绿+蓝+近红外
随机森林总体精度82.57%82.72%79.70%84.60%89.20%89.82%89.93%
Kappa系数0.734 10.733 50.686 90.764 90.835 30.8460.847 1
决策树C4.5总体精度76.83%77.30%71.76%76.88%81.83%84.45%81.16%
Kappa系数0.646 20.651 30.572 40.648 90.724 40.764 80.712 1
逻辑回归总体精度56.16%56.11%57.41%64.61%65.24%66.65%67.69%
Kappa系数0.287 80.281 40.301 70.435 10.457 20.480 80.502 2
表 4  花生旱情等级识别精度评估
图7  花生旱灾灾情最早识别时间评估图
图8  研究区2019年花生旱灾灾情识别结果
图9  研究区2019年花生受灾面积统计
1 Song Yang, Fang Shibo, Wei Yaxing. Comparison of typical remote sensing drought indexes and their adaptability in agriculture[J]. Science & Technology Review,2016,34(5):45-52.
1 宋扬,房世波,卫亚星.农业干旱遥感监测指数及其适用性研究进展[J].科技导报,2016,34(5):45-52.
2 Fang Shibo, Qi Yue, Han Guojun, et al.Trends and possible impacts of meteorological drought in winter and spring in main wheat regions of China from 1961 to 2010[J].Scientia Acricultura Sinica,2014,47(9):1754-1763.
2 房世波,齐月,韩国军,等.1961—2010年中国主要麦区冬春气象干旱趋势及其可能影响[J].中国农业科学,2014,47(9):1754-1763.
3 Wilhite D A. Drought as a natural hazard: Concepts and definitions[J]. Drought, A Global Assessment, 2000, 1: 3-18.
4 Yan Feng, Qin Zhihao, Li Maosong, et al. Progress in soil moisture estimation from remote sensing data for agricultural drought monitoring[J].Journal of Natural Disasters,2006,15(6):114-121.
4 闫峰,覃志豪,李茂松,等.农业旱灾监测中土壤水分遥感反演研究进展[J].自然灾害学报,2006,15(6):114-121.
5 Meng Jihua, Fu Wei, Xu Jin,et al. Application of remote sensing in loss assessment of planting insurance[J].Remote Sensing Technology and Application,2017,32(2):238-246.
5 蒙继华,付伟,徐晋,等.遥感在种植业保险估损中的应用[J].遥感技术与应用,2017,32(2):238-246.
6 Wu Chunlei, Qin Qiming, Li Mei,et al.Soil moisture monitoring of vegetative area in farmland by remote sensing based on spectral feature space[J].Transactions of the Chinese Society of Agricultural Engineering,2014,30(16):106-112.
6 吴春雷,秦其明,李梅,等.基于光谱特征空间的农田植被区土壤湿度遥感监测[J].农业工程学报,2014,30(16):106-112.
7 Wang Siyuan, Li Qiangzi, Wang Hongyan,et al.Winter wheat drought monitoring based on TROPOMI chlorophyll fluorescence remote sensing[J].Remote Sensing Technology and Application,2021,36(5):1057-1071.
7 王思远,李强子,王红岩,等.基于TROPOMI叶绿素荧光遥感的冬小麦旱情监测[J].遥感技术与应用,2021,36(5):1057-1071.
8 Guo Hu, Wang Ying, Wang Fang. Overview of remote sensing approaches to drought monitoring[J]. Remote Sensing Technology and Application,2008,23(1):111-116.
8 郭虎,王瑛,王芳.旱灾灾情监测中的遥感应用综述[J].遥感技术与应用,2008,23(1):111-116.
9 Mo Weihua, Wang Zhenhui, Sun Han, al et, Remote sensing monitoring of farmland drought based on vegetation supply water index[J]. Journal of Nanjing Institute of Meteorology, 2006,29(3):396-401.
9 莫伟华,王振会,孙涵,等.基于植被供水指数的农田干旱遥感监测研究[J].南京气象学院学报,2006,29(3):396-401.
10 Liu Liangming. Study on remote sensing drought early warning model based on EOS/MODIS data[D]. Wuhan: Wuhan University,2004.
10 刘良明.基于EOS/MODIS数据的遥感干旱预警模型研究[D]. 武汉:武汉大学,2004.
11 Wu Bingfang, Zhang Feng, Liu Chenglin,et al.An integrated method for crop condition monitoring[J]. Journal of Remote Sensing, 2005,8(6):498-514.
11 吴炳方,张峰,刘成林,等.农作物长势综合遥感监测方法[J].遥感学报,2005,8(6):498-514.
12 Li Xinghua, Li Yunpeng, Yang Liping.Study on comprehensive application of drought monitoring and evaluation method in Inner Mongolia[J]. Journal of Arid Land Resources and Environment, 2014,28(3):162-166.
12 李兴华,李云鹏,杨丽萍.内蒙古干旱监测评估方法综合应用研究[J].干旱区资源与环境,2014,28(3):162-166.
13 Huang Youxin, Liu Xiuguo, Shen Yonglin,et al.Advances in remote sensing derived agricultural drought monitoring indices and adaptability evaluation methods[J].Transactions of the Chinese Society of Agricultural Engineering, 2015,31(16):186-195.
13 黄友昕,刘修国,沈永林,等.农业干旱遥感监测指标及其适应性评价方法研究进展[J].农业工程学报,2015,31(16):186-195.
14 Hatfield J L, Prueger J H. Value of using different vegetative indices to quantify agricultural crop characteristics at different growth stages under varying management practices[J]. Remote Sensing, 2010, 2(2): 562-578.
15 Sun H, Zhao X, Chen Y, et al. A new agricultural drought monitoring index combining MODIS NDWI and day-night land surface temperatures: A case study in China[J]. International Journal of Remote Sensing, 2013, 34(24): 8986-9001.
16 Li Huapeng, Zhang Shuqing, Gao Ziqiang,et al.Evaluating the utility of MODIS vegetation index for monitoring agricultural drought[J].Spectroscopy and Spectral Analysis,2013,33(3):756-761.
16 李华朋,张树清,高自强,等.MODIS植被指数监测农业干旱的适宜性评价[J].光谱学与光谱分析,2013,33(3):756-761.
17 Sun Hao, Chen Yunhao, Sun Hongquan.Comparisons and classification system of typical remote sensing indexes for agricultural drought[J].Transactions of the Chinese Society of Agricultural Engineering,2012,28(14):147-154.
17 孙灏,陈云浩,孙洪泉.典型农业干旱遥感监测指数的比较及分类体系[J]. 农业工程学报,2012,28(14):147-154.
18 Jia Dewei, Zhou Lei, Huang Canhui, et al.Research progress on remote sensing monitoring methods and applications of agricultural drought[J].Anhui Agricultural Sciences, 2016, 44 (36): 233-235.
18 贾德伟,周磊,黄灿辉,等.农业干旱遥感监测方法及其应用的研究进展[J].安徽农业科学, 2016,44(36):233-235.
19 Rouse J W, Haas R H, Schell J A, et al. Monitoring vegetation systems in the great plains with ERTS[C]∥ Stanley C F. Proceedings of Third Earth Resources Technology Satellite Symposium, Washington: NASA Special Publication, 1974: 310-317.
20 Wang Fei, Yang Shengtian, Ding Jianli,et al.Environmental sensitive variable optimization and machine learning algorithm using in soil salt prediction at oasis[J].Transactions of the Chinese Society of Agricultural Engineering,2018,34(22):102-110.
20 王飞,杨胜天,丁建丽,等.环境敏感变量优选及机器学习算法预测绿洲土壤盐分[J].农业工程学报,2018,34(22):102-110.
21 Chen Liping, Wu Wenbo.Remote sensing image classification based on object-oriented technique and C 4.5 algorithm[J].Remote Sensing Information, 2013,28(2):116-120.
21 陈丽萍,武文波.基于决策树C4.5算法的面向对象分类方法研究[J].遥感信息,2013,28(2):116-120.
22 Yu Qiao, Jiang Shujuan, Zhang Yanmei,et al. The impact study of class imbalance on the performance of software defect prediction models[J]. Chinese Journal of Computers,2018,41(4):809-824.
22 于巧,姜淑娟,张艳梅,等.分类不平衡对软件缺陷预测模型性能的影响研究[J]. 计算机学报,2018,41(4):809-824.
23 Zhou Peicheng, Cheng Gong, Yao Xiwen, et al. Machine learning paradigms in high-resolution remote sensing image interpretation[J]. Journal of Remote Sensing,2021,25(1):182-197.
23 周培诚,程塨,姚西文,等.高分辨率遥感影像解译中的机器学习范式[J].遥感学报,2021,25(1):182-197.
24 Breiman L .Random forests[J].Machine Learning,2001,45(1):5-32.
25 Wu Wei, Zhang Yuan, Li Qiangzi,et al. A hierarchical classification and iterative model based method for remote sensing classification of land cover[J]. Remote Sensing Technology and Application, 2019, 34(1): 68-78.
25 吴薇,张源,李强子,等.基于迭代CART算法分层分类的土地覆盖遥感分类[J].遥感技术与应用, 2019, 34(1): 68-78.
26 Nie Bin, Li Huan, Luo Jigen,et al.The study on classification of C4.5 algorithms with GINI index[J]. Journal of Jiangxi Normal University: Natural Science Edition,2019,43(5):469-472.
26 聂斌,李欢,罗计根,等.融合GINI指数的C4.5算法的分类研究[J]. 江西师范大学学报(自然科学版),2019,43(5):469-472.
27 Fang Miao, Zhang Jinlong, Xu Zhen. Landslide susceptibility zoning study in Lanzhou city based on GIS and Logistic Regression Model[J].Remote Sensing Technology and Application, 2011, 26(6): 845-854.
27 方苗,张金龙,徐瑱.基于GIS和Logistic回归模型的兰州市滑坡灾害敏感性区划研究[J].遥感技术与应用,2011,26(6):845-854.
28 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.
28 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.
[1] 尹芳,封凯,吴朦朦,拜得珍,王蕊,周园园,尹春涛,尹翠景,刘磊. 一种基于分段偏最小二乘模型的土壤重金属遥感反演方法[J]. 遥感技术与应用, 2021, 36(6): 1321-1328.
[2] 蒋嘉锐,朱文泉,乔琨,江源. 基于Sentinel-2数据的天山山地针叶林识别方法研究[J]. 遥感技术与应用, 2021, 36(4): 847-856.
[3] 刘鹤,顾玲嘉,任瑞治. 基于无人机遥感技术的森林参数获取研究进展[J]. 遥感技术与应用, 2021, 36(3): 489-501.
[4] 郭云开,张源,张文博. 植被健康指数及高等级公路路域环境评价研究[J]. 遥感技术与应用, 2013, 28(6): 1120-1124.
[5] 温 奇,范一大,陈世荣,孙 灏,胡卓玮. 环境减灾卫星数据在旱灾预警监测中的应用[J]. 遥感技术与应用, 2012, 27(4): 591-599.
[6] 张晓艳, 刘锋, 王丽丽, 封文杰, 刘淑云, 朱建华. 花生叶面积指数与特征导数光谱的相关性[J]. 遥感技术与应用, 2010, 25(5): 668-674.
[7] 郭 虎,王 瑛,王 芳. 旱灾灾情监测中的遥感应用综述[J]. 遥感技术与应用, 2008, 23(1): 111-116.
[8] 冯燕,何明一,魏江. 基于神经网络的多光谱遥感图像无损压缩[J]. 遥感技术与应用, 2004, 19(1): 42-46.
[9] 王介民,高 峰,刘绍民. 流域尺度ET的遥感反演[J]. 遥感技术与应用, 2003, 18(5): 332-338.