设为首页
加入收藏
Extraction of Peanut Planting Area Based on Dual-temporal Remote Sensing Features of Crops
-
摘要: 基于花生生长中后期2020年8月1日和15日两个时相高分多光谱数据，构建40个作物分类遥感特征，采用ReliefF-Pearson方法优选出15个特征，构造作物可分的4种特征空间。采用最大似然分类法、支持向量机和随机森林分类器，分别耦合4种特征空间，开展作物分类对比试验，进行分类精度和景观评价提出作物双时相遥感分类模型(dual-temporal remote sensing classification model for crop, C-DRSC)。结果表明：该模型具有较高的作物分类和花生识别能力，作物分类总体精度和Kappa系数分别为93.25%和0.89，平均形状指数和平均斑块分维指数分别为1.33和1.13;花生识别的用户精度和制图精度分别为96.20%和96.32%，平均形状指数和平均斑块分维指数分别为1.27和1.11。利用该模型在黄淮海地区的4个花生主产县开展夏花生种植面积遥感测算，与统计面积相比，面积测算相对误差为±16.25%，决定系数为0.9778(达到0.01显著性水平)，模型具有较好的适用性。Abstract: Various high-resolution satellite remote sensing platforms and sensors have been established recently, but there are still many challenges for crop type fine classification. New methods are needed for fine classification of crop types and extraction of peanut planting area using high-resolution optical remote sensing data. Using panchromatic and multi-spectral (PMS) data of GF-1 and GF-6 in the middle and late stages of crop growth, 40 classification features are constructed from the perspective of spectrum, texture, and temporal variation of remote sensing image by means of spatio-temporal fusion and spectral mathematical transformation.15 features are selected by ReliefF-Pearson method. To compare the effects of different feature types on classification, the selected features are combined into four feature spaces, and the separability between crops in these feature spaces is measured by Jeffries-Matusita (J-M) distance. Using maximum likelihood classification, support vector machine and random forest as classifiers, crop classification tests of four feature space schemes are carried out, and different classification results are evaluated from the perspective of remote sensing classification accuracy and landscape analysis. Compared with the original image, in four feature spaces, the J-M distance between peanut and other crops is close to the saturation value of 2.0, and their separability is significantly enhanced. For all test results, the overall accuracy of crop classification is more than 78.0%, the Kappa coefficient of crop classification is more than 0.7, and the user's accuracy and producer's accuracy of peanut are both more than 79.0%. Compared with the three classifiers, Random Forest has the highest classification accuracy, and the average values of the overall accuracy and Kappa coefficient, peanut user's accuracy and producer's accuracy of its four schemes are 91.49%, 0.89, 94.45% and 93.30%, respectively. In addition, combining dual-temporal remote sensing data, texture and vegetation index features, as well as their temporal changes, can effectively improve the classification accuracy. It shows that adding the texture feature can improve the accuracy better than the vegetation index features. The landscape shape analysis of the classification results also shows that the overall crop classification and peanut, their patch shapes are close to the reality, and random forest is the best classifier. Its mean shape indices of crop landscape-level and peanut class-level are only 1.333 and 1.270, and the corresponding mean patch fractal dimension indices are only 1.127 and 1.110. An optimal dual temporal remote sensing crop classification model is proposed. Using this model, summer peanut planting area extraction and area measurement are carried out in four main peanut producing counties in Huang-Huai-Hai Region. Compared with the statistical data, the determination coefficient is 0.98 and the relative error of area calculation is ±16.25%, showing a good application prospect.
-
图 1 典型试验区和样本分布
(方形代表训练样本, 圆形代表测试样本)
Fig. 1 Typical study area and experimental measurements
(the square denotes training sample, the circle denotes test sample)
图 6 花生种植区识别结果和统计面积的区域占比
Fig. 6 Recognition results of peanut planting area by C-DRSC and spatial distribution of regional proportion of statistical area
图 7 花生面积遥感提取的试验效果
Fig. 7 Experimental effect analysis of peanut planting area measured by remote sensing
表 1 作物样本数量情况
Table 1 Experimental measurements for each crop type
作物 训练样本 测试样本 地块 像素 地块 像素 玉米 49 59539 45 55786 花生 79 212440 86 241625 水稻 45 100386 43 94936 大豆 30 27994 29 26862 
下载: 导出CSV
表 2 特征信息表
Table 2 Information of selected features
编码 特征参量 类型 S1~S4 2020-08-01时相的红、绿、蓝和近红外波段 光谱特征 S5~S8 2020-08-15时相的红、绿、蓝和近红外波段 V1 2020-08-01时相的归一化植被指数 植被指数特征 V2 2020-08-01时相的绿度总和指数 V3 2020-08-01时相的比值植被指数 V4 2020-08-01和2020-08-15两时相归一化植被指数差 T1 2020-08-01时相纹理平均值 纹理特征 T2 2020-08-15时相纹理平均值 T3 2020-08-15和2020-08-01两时相纹理平均值差
下载: 导出CSV
表 3 不同特征空间花生与表中作物的J-M距离
Table 3 The J-M distances between peanut and the following crop in different feature spaces
作物 原始影像 构造的特征空间 2020-08-01 2020-08-15 A B C D 大豆 1.9199 1.4455 1.9527 1.9564 1.9967 1.9970 水稻 1.9669 1.4921 1.9998 1.9999 2.0000 2.0000 玉米 1.7635 1.9935 1.8510 1.8616 1.9916 1.9922
下载: 导出CSV
-
[1] 唐华俊, 周清波, 刘佳, 等. 中国农作物空间分布高分遥感制图——小麦篇. 北京: 科学出版社, 2016.Tang H J, Zhou Q B, Liu J, et al. Wheat Mapping Using High Resolution Remote Sensing Data. Beijing: Science Press, 2016. [2] 罗敬宁, 刘立葳. 遥感大数据分布式技术研究与实现. 应用气象学报, 2017, 28(5): 621-631. doi: 10.11898/1001-7313.20170510Luo J N, Liu L W. Research and implementation of remote sensing big data distributed technology. J Appl Meteor Sci, 2017, 28(5): 621-631. doi: 10.11898/1001-7313.20170510 [3] 业巧林, 许等平, 张冬. 基于深度学习特征和支持向量机的遥感图像分类. 林业工程学报, 2019, 4(2): 125-131. https://www.cnki.com.cn/Article/CJFDTOTAL-LKKF201902020.htmYe Q L, Xu D P, Zhang D. Remote sensing image classification based on deep learning features and support vector machine. Journal of Forestry Engineering, 2019, 4(2): 125-131. https://www.cnki.com.cn/Article/CJFDTOTAL-LKKF201902020.htm [4] 匡秋明, 于廷照. AI技术与卫星资料应用研究现状分析. 气象科技进展, 2020, 10(3): 21-29. doi: 10.3969/j.issn.2095-1973.2020.03.004Kuang Q M, Yu T Z. Review of leveraging AI for exploitation of satellite observations. Advances in Meteorological Science and Technology, 2020, 10(3): 21-29. doi: 10.3969/j.issn.2095-1973.2020.03.004 [5] 孙健, 曹卓, 李恒, 等. 人工智能技术在数值天气预报中的应用. 应用气象学报, 2021, 32(1): 1-11. doi: 10.11898/1001-7313.20210101Sun J, Cao Z, Li H, et al. Application of artificial intelligence technology to numerical weather prediction. J Appl Meteor Sci, 2021, 32(1): 1-11. doi: 10.11898/1001-7313.20210101 [6] 张鹏, 胡守庚. 地块尺度的复杂种植区作物遥感精细分类. 农业工程学报, 2019, 35(20): 125-134. doi: 10.11975/j.issn.1002-6819.2019.20.016Zhang P, Hu S G. Fine crop classification by remote sensing in complex planting area based on field parcel. Transactions of the CSAE, 2019, 35(20): 125-134. doi: 10.11975/j.issn.1002-6819.2019.20.016 [7] 金子琪, 王新敏, 鲍艳松, 等. 基于卷积神经网络的飑线识别算法. 应用气象学报, 2021, 32(5): 580-591. doi: 10.11898/1001-7313.20210506Jin Z Q, Wang X M, Bao Y S, et al. Squall line identification method based on convolution neural network. J Appl Meteor Sci, 2021, 32(5): 580-591. doi: 10.11898/1001-7313.20210506 [8] 李颖, 陈怀亮. 机器学习技术在现代农业气象中的应用. 应用气象学报, 2020, 31(3): 257-266. doi: 10.11898/1001-7313.20200301Li Y, Chen H L. Review of machine learning approaches for modern agrometeorology. J Appl Meteor Sci, 2020, 31(3): 257-266. doi: 10.11898/1001-7313.20200301 [9] 邓继忠, 刘其得, 王长委, 等. 基于多时相Sentinel-2卫星数据的农作物分类研究. 广东农业科学, 2020, 47(4): 129-138. https://www.cnki.com.cn/Article/CJFDTOTAL-GDNY202004019.htmDeng J Z, Liu Q D, Wang C W, et al. Crop classification based on multitemporal Sentinel-2 satellite imagery. Guangdong Agricultural Sciences, 2020, 47(4): 129-138. https://www.cnki.com.cn/Article/CJFDTOTAL-GDNY202004019.htm [10] 刘佳, 王利民, 滕飞, 等. RapidEye卫星红边波段对农作物面积提取精度的影响. 农业工程学报, 2016.32(13): 140-148. doi: 10.11975/j.issn.1002-6819.2016.13.020Liu J, Wang L M, Teng F, et al. Impact of red-edge waveband of RapidEye satellite on estimation accuracy of crop planting area. Transactions of the CSAE, 2016, 32(13): 140-148. doi: 10.11975/j.issn.1002-6819.2016.13.020 [11] 魏鹏飞, 徐新刚, 杨贵军, 等. 基于多时相影像植被指数变化特征的作物遥感分类. 中国农业科技导报, 2019, 21(2): 54-61. https://www.cnki.com.cn/Article/CJFDTOTAL-NKDB201902008.htmWei P F, Xu X G, Yang G J, et al. Remote sensing classification of crops based on the change characteristics of multi-phase vegetation index. Journal of Agricultural Science and Technology, 2019, 21(2): 54-61. https://www.cnki.com.cn/Article/CJFDTOTAL-NKDB201902008.htm [12] Zhang Y, Yang W, Sun Y, et al. Fusion of multispectral aerial imagery and vegetation indices for machine learning-based ground classification. Remote Sensing, 2021, 13(8): 1411. doi: 10.3390/rs13081411 [13] 刘二华, 周广胜, 周莉, 等. 夏玉米不同生育期叶片和冠层含水量的遥感反演. 应用气象学报, 2020, 31(1): 52-62. doi: 10.11898/1001-7313.20200105Liu E H, Zhou G S, Zhou L, et al. Remote sensing inversion of leaf and canopy water content in different growth stages of summer maize. J Appl Meteor Sci, 2020, 31(1): 52-62. doi: 10.11898/1001-7313.20200105 [14] 杨磊, 韩丽娟, 宋金玲, 等. 基于遥感数据的夏玉米高温热害监测评估. 应用气象学报, 2020, 31(6): 749-758. doi: 10.11898/1001-7313.20200610Yang L, Han L J, Song J L, et al. Monitoring and evaluation of high temperature and heat damage of summer maize based on remote sensing data. J Appl Meteor Sci, 2020, 31(6): 749-758. doi: 10.11898/1001-7313.20200610 [15] 张因国, 陶于祥, 罗小波, 等. 基于特征重要性的高光谱图像分类. 红外技术, 2020, 42(12): 1185-1191. https://www.cnki.com.cn/Article/CJFDTOTAL-HWJS202012010.htmZhang Y G, Tao Y X, Luo X B, et al. Hyperspectral image classification based on feature importance. Infrared Technology, 2020, 42(12): 1185-1191. https://www.cnki.com.cn/Article/CJFDTOTAL-HWJS202012010.htm [16] Yang S T, Gu L J, Li X F, et al. Crop classification method based on optimal feature selection and hybrid CNN-RF networks for multi-temporal remote sensing imagery. Remote Sensing, 2020, 12(19): 3119. doi: 10.3390/rs12193119 [17] Orynbaikyzy A, Gessner U, Mack B, et al. Crop type classification using fusion of Sentinel-1 and Sentinel-2 data: Assessing the impact of feature selection, optical data availability, and parcel sizes on the accuracies. Remote Sensing, 2020, 12(17): 2779. doi: 10.3390/rs12172779 [18] 邓刘洋, 沈占锋, 柯映明, 等. 基于地块尺度多时相遥感影像的冬小麦种植面积提取. 农业工程学报, 2018, 34(21): 157-164. doi: 10.11975/j.issn.1002-6819.2018.21.019Deng L Y, Shen Z F, Ke Y M, et al. Winter wheat planting area extraction using multi-temporal remote sensing images based on field parcel. Transactions of the CSAE, 2018, 34(21): 157-164. doi: 10.11975/j.issn.1002-6819.2018.21.019 [19] 马青荣, 左璇, 胡程达, 等. 涝渍对夏花生光合特性及产量影响. 应用气象学报, 2021, 32(4): 479-490. doi: 10.11898/1001-7313.20210409Ma Q R, Zuo X, Hu C D, et al. Effects of waterlogging on photosynthetic characteristics and yield of summer peanut. J Appl Meteor Sci, 2021, 32(4): 479-490. doi: 10.11898/1001-7313.20210409 [20] 周曙东, 景令怡, 孟桓宽, 等. 中国花生主产区生产布局演变规律及动因挖掘. 农业技术经济, 2018(3): 100-109. https://www.cnki.com.cn/Article/CJFDTOTAL-NYJS201803009.htmZhou S D, Jing L Y, Meng H K, et al. Analysis on spatial distribution evolution of main peanut production areas in China and its influencing factors. Journal of Agrotechnical Economics, 2018(3): 100-109. https://www.cnki.com.cn/Article/CJFDTOTAL-NYJS201803009.htm [21] 魏思成, 李凯伟, 张继权, 等. 黄淮海地区春花生旱涝灾害危险性评价. 应用气象学报, 2021, 32(5): 629-640. doi: 10.11898/1001-7313.20210510Wei S C, Li K W, Zhang J Q, et al. Hazard assessment of peanut drought and flood disasters in Huang-Huai-Hai Region. J Appl Meteor Sci, 2021, 32(5): 629-640. doi: 10.11898/1001-7313.20210510 [22] 程增书, 李玉荣, 徐桂真, 等. 河北省花生生产、科研现状与产业化发展对策. 花生学报, 2003, 32(增刊I): 60-63. https://www.cnki.com.cn/Article/CJFDTOTAL-PEAN2003S1011.htmCheng Z S, Li Y R, Xu G Z, et al. Present situation of peanut production, scientific research and industrialization development countermeasures in Hebei Province. Journal of Peanut Science, 2003, 32(Suppl I): 60-63. https://www.cnki.com.cn/Article/CJFDTOTAL-PEAN2003S1011.htm [23] 周帅, 韩彬, 李帅, 等. 河南省花生生产现状与发展对策. 天津农业科学, 2021, 27(8): 56-59. doi: 10.3969/j.issn.1006-6500.2021.08.011Zhou S, Han B, Li S, et al. Current situation and development countermeasure of peanut in Henan Province. Tianjin Agricultural Sciences, 2021, 27(8): 56-59. doi: 10.3969/j.issn.1006-6500.2021.08.011 [24] 朱保成, 李志刚, 王玉田, 等. 河南花生产业发展的思考——驻马店市正阳县花生产业发展调查. 中国粮食经济, 2019(10): 52-55. doi: 10.3969/j.issn.1007-4821.2019.10.015Zhu B C, Li Z G, Wang Y T, et al. Consideration on development of peanut industry in Henan-Investigation on the development of peanut industry in Zhengyang County, Zhumadian City. China Grain Economy, 2019(10): 52-55. doi: 10.3969/j.issn.1007-4821.2019.10.015 [25] 刘娜, 熊安元, 张强, 等. 强对流天气人工智能应用训练基础数据集构建. 应用气象学报, 2021, 32(5): 530-541. doi: 10.11898/1001-7313.20210502Liu N, Xiong A Y, Zhang Q, et al. Development of basic dataset of severe convective weather for artificial intelligence training. J Appl Meteor Sci, 2021, 32(5): 530-541. doi: 10.11898/1001-7313.20210502 [26] 王娜, 李强子, 杜鑫, 等. 单变量特征选择的苏北地区主要农作物遥感识别. 遥感学报, 2017, 21(4): 519-530. https://www.cnki.com.cn/Article/CJFDTOTAL-YGXB201704004.htmWang N, Li Q Z, Du X, et al. Identification of main crops based on the univariate feature selection in Subei. Journal of Remote Sensing, 2017, 21(4): 519-530. https://www.cnki.com.cn/Article/CJFDTOTAL-YGXB201704004.htm [27] 陈昱文, 黄小猛, 李熠, 等. 基于ECMWF产品的站点气温预报集成学习误差订正. 应用气象学报, 2020, 31(4): 494-503. doi: 10.11898/1001-7313.20200411Chen Y W, Huang X M, Li Y, et al. Ensemble learning for bias correction of station temperature forecast based on ECMWF products. J Appl Meteor Sci, 2020, 31(4): 494-503. doi: 10.11898/1001-7313.20200411 [28] 韩丰, 杨璐, 周楚炫, 等. 基于探空数据集成学习的短时强降水预报试验. 应用气象学报, 2021, 32(2): 188-199. doi: 10.11898/1001-7313.20210205Han F, Yang L, Zhou C X, et al. An experimental study of the short-time heavy rainfall event forecast based on ensemble learning and sounding data. J Appl Meteor Sci, 2021, 32(2): 188-199. doi: 10.11898/1001-7313.20210205 [29] Kononenko I. Estimating Attributes: Analysis and Extensions of RELIEF//European Conference on Machine Learning, 1994: 171-182. [30] 刘莹, 孟庆岩, 王永吉, 等. 基于特征优选与支持向量机的不透水面覆盖度估算方法. 地理与地理信息科学, 2018, 34(1): 24-31. doi: 10.3969/j.issn.1672-0504.2018.01.005Liu Y, Meng Q Y, Wang Y J, et al. A method for estimating impervious surface percentage based on feature optimization and SVM. Geography and Geo-Information Science, 2018, 34(1): 24-31. doi: 10.3969/j.issn.1672-0504.2018.01.005 [31] 童庆禧, 张兵, 郑兰芬. 高光谱遥感: 原理、技术与应用. 北京: 高等教育出版社, 2006.Tong Q X, Zhang B, Zheng L F. Hyperspectral Remote Sensing: Principles, Techniques, and Applications. Beijing: Higher Education Press, 2006. [32] 耿修瑞. 高光谱遥感图像目标探测与分类技术研究. 北京: 中国科学院研究生院(遥感应用研究所), 2005.Geng X R. Target Detection and Classification for Hyperspectral Imagery. Beijing: University of Chinese Academy of Sciences(Institute of Remote Sensing Applications), 2005. [33] 邓乃扬, 田英杰. 数据挖掘中的新方法: 支持向量机. 北京: 科学出版社, 2004.Deng N Y, Tian Y J. A New Method in Data Mining: Support Vector Machine. Beijing: Science Press, 2004. [34] Breiman L. Random forests. Machine Learning, 2001, 45(1): 5-32. doi: 10.1023/A:1010933404324 [35] Congalton R G. A review of assessing the accuracy of classifications of remotely sensed data. Remote Sensing of Environment, 1991, 37(1): 35-46. doi: 10.1016/0034-4257(91)90048-B [36] 张永生, 欧阳芳, 袁哲明. 华北农田生态系统景观格局的演变特征. 生态科学, 2018, 37(4): 114-122. https://www.cnki.com.cn/Article/CJFDTOTAL-STKX201804014.htmZhang Y S, Ouyang F, Yuan Z M. Change characteristics of landscape pattern in farmland ecosystems in North China. Ecological Science, 2018, 37(4): 114-122. https://www.cnki.com.cn/Article/CJFDTOTAL-STKX201804014.htm [37] 罗志军, 赵越, 赵杰, 等. 基于景观格局与空间自相关的永久基本农田划定研究. 农业机械学报, 2018, 49(10): 195-204. doi: 10.6041/j.issn.1000-1298.2018.10.022Luo Z J, Zhao Y, Zhao J, et al. Defining of permanent basic farmland based on landscape pattern and spatial autocorrelation. Transactions of the Chinese Society for Agricultural Machinery, 2018, 49(10): 195-204. doi: 10.6041/j.issn.1000-1298.2018.10.022 -
点击查看大图
图(7) / 表(3)
计量
- 摘要浏览量: 1241
- HTML全文浏览量: 304
- PDF下载量: 86
- 被引次数: 0
