Please wait a minute...
 
Remote Sensing for Land & Resources    2019, Vol. 31 Issue (3) : 43-50     DOI: 10.6046/gtzyyg.2019.03.06
|
The development of plastic greenhouse index based on Logistic regression analysis
Jun CHEN1, Runping SHEN1(), Bolun LI1, Chaopu TI2, Xiaoyuan YAN2, Minyue ZHOU1, Shaowu WANG1
1. School of Geographical Sciences, Nanjing University of Information Science and Technology, Nanjing 210044,China
2. State Key Laboratory of Sustainable Soil and Agriculture, Nanjing Institute of Soil Science, Chinese Academy of Sciences, Nanjing 210008, China
Download: PDF(5641 KB)   HTML
Export: BibTeX | EndNote | Reference Manager | ProCite | RefWorks    
Abstract  

In order to accurately extract a large range of plastic greenhouse distribution information, the authors took Changzhou City, which is located in the Taihu Lake basin, as the study area, used Landsat8 imagery, employed plastic greenhouses spectral analysis and spectral separability analysis, selected seven multi-spectral data and one thermal infrared datum of Landsat8 image and three remote sensing indexes(NDVI, NDBaI and MNDWI)and, based on Logistic regression analysis, constructed a new plastic greenhouse index (NewPGI). Accuracy verification results show that, in the sample area, the high-resolution image of the plastic greenhouse reference map shows that NewPGI’s overall classification accuracy is 94.9%, and Kappa coefficient is 0.74. Throughout Changzhou, the verification sample points were selected based on the Google Earth image. The overall accuracy of NewPGI is 91.28%, and the Kappa coefficient is 0.78. Compared with the existing plastic greenhouse index, NewPGI can better extract plastic greenhouses under complex surface coverage.

Keywords plastic greenhouse      remote sensing index      Logistic regression analysis     
:  S127TP79  
Corresponding Authors: Runping SHEN     E-mail: rpShen@nuist.edu.cn
Issue Date: 30 August 2019
Service
E-mail this article
E-mail Alert
RSS
Articles by authors
Jun CHEN
Runping SHEN
Bolun LI
Chaopu TI
Xiaoyuan YAN
Minyue ZHOU
Shaowu WANG
Cite this article:   
Jun CHEN,Runping SHEN,Bolun LI, et al. The development of plastic greenhouse index based on Logistic regression analysis[J]. Remote Sensing for Land & Resources, 2019, 31(3): 43-50.
URL:  
https://www.gtzyyg.com/EN/10.6046/gtzyyg.2019.03.06     OR     https://www.gtzyyg.com/EN/Y2019/V31/I3/43
Fig.1  Location and image of study area
Fig.2  Satellite data in the sample area
Fig.3  Reference of classification in the sample area
Fig.4  Validation data of study area
Landsat8波段信息 研究数据 数据或算法介绍 参考资料
http: //landsat.usgs.gov/
B1: Coastal(深蓝) 海岸带环境监测
B2: Blue (蓝光) 可见光波段,合成模拟真彩色影像用于地物识别等
B3: Green(绿光)
B4: Red(红光)
B5: NIR(近红外) 植被信息提取
B6: SWIR1(短波红外) 植被旱情监测、强火监测和部分矿物信息提取
B7: SWIR2(短波红外)
B10: TIRS1(热红外) 地表温度反演、火灾监测、土壤湿度评价和夜间成像等
遥感指数 NDVI (B5-B4)/(B5+B4) 许剑辉等[10]
NDBI (B6-B5)/(B6+B5) As-syakur等[11]
NDBaI (B6-B10)/(B6+B10)
MNDWI (B3-B7)/(B3+B7) Xu等[12]
Tab.1  Spectral data based on Landsat8 image and various remote sensing indexes
Fig.5  Spectral curves of different land cover types (mean values)
波段及遥
感指数
大棚 / 人
造地表
大棚 / 裸地
和休耕地
大棚 /
植被
大棚 /
水体
B1 0.924 7 1.136 5 2.214 8 1.351 5
B2 0.931 0 1.034 8 2.211 9 1.275 6
B3 1.023 0 0.691 6 1.906 9 1.019 9
B4 0.767 8 0.352 4 2.035 1 1.034 2
B5 1.444 5 0.823 5 0.392 5 1.680 3
B6 0.876 6 0.232 1 0.956 7 2.025 0
B7 0.403 5 0.056 7 1.359 9 1.886 4
B10 0.608 4 0.148 4 0.5071 1.363 5
NDVI 0.482 9 0.191 5 2.042 4 1.045 8
NDBI 0.740 1 0.636 1 0.902 2 0.487 8
NDBaI 0.949 3 0.230 5 0.913 0 1.951 2
MNDWI 0.348 3 0.726 6 1.809 1 1.396 8
Tab.2  Separation degree of plastic greenhouses from typical land cover types
Fig.6  Image of Sigmoid function
Xk ak Sig. Xk ak Sig.
B1 76.943 0.019 B7 -43.667 0.024
B2 -91.195 0.012 B10 155.886 0
B3 -146.302 0 NDVI 32.461 0.001
B4 60.4 0.04 NDBaI 138.95 0
B5 -34.773 0.011 MNDWI 83.31 0
B6 -63.933 0.018 常量 24.98 0.089
Tab.3  Parameter variables in NewPGI
X2 波段数 Sig. Cox&Snell Nagelkerke
247.03 11 0.000 0.414 0.697
Tab.4  Omnibus test of Logistic regression model coefficients and test of R-squared
Fig.7  Plastic greenhouse information extraction in the sample area
分类 非塑料大棚 塑料大棚 总计 用户精度/%
非塑料大棚 38 795 1 348 40 143 96.64
塑料大棚 947 3 910 4 857 80.50
总计 39 742 5 258 45 000
制图精度/% 97.62 74.36
总体精度 94.9% Kappa系数 0.74
Tab.5  PGs/No PGs confusion matrix of the sample area
Fig.8  Plastic greenhouse information extraction in Changzhou City
遥感指数 非塑料大棚 塑料大棚 总计 用户精度/%
NewPGI 非塑料大棚 1 657 123 1 780 93.09
塑料大棚 92 594 686 86.59
总计 1 749 717 2 466
制图精度/% 94.74 82.85
总体精度 91.28% Kappa系数 0.78
PGI 非塑料大棚 1 603 318 1 921 73.21
塑料大棚 146 399 545 83.45
总计 1 749 717 2 466
制图精度/% 91.65 55.65
总体精度 81.18% Kappa系数 0.51
Tab.6  PGs/No PGs confusion matrix of the study area
[1] 国家统计局国务院第三次全国农业普查领导小组. 第三次全国农业普查主要数据公报[R]. 北京:国家统计局, 2017.
[1] The Third National Agricultural Census Leading Group of the State Council, the National Bureau of Statistics. The Main Data Bulletin of the Third National Agricultural Census[R]. Beijing:National Bureau of Statistics, 2017.
[2] Navulur K. Multispectral Image Analysis Using the Object-Oriented Paradigm[M]. Boca Raton: CRC Press, 2006.
[3] Agüera F, Aguilar M A, Aguilar F J , et al. Detecting greenhouse changes from QuickBird imagery on the Mediterranean coast[J]. International Journal of Remote Sensing, 2006,27(21):4751-4767.
[4] Aguilar M A, Bianconi F, Aguilar F J , et al. Object-based greenhouse classification from GeoEye-1 and WorldView-2 stereo imagery[J]. International Journal of Remote Sensing, 2014,6(5):3554-3582.
[5] Aguilar M A, Nemmaoui A, Novelli A , et al. Object-based greenhouse mapping using very high resolution satellite data and Landsat-8 time series[J]. International Journal of Remote Sensing, 2016,8(6):513.
[6] Zhao G X, Li J, Li T , et al. Utilizing Landsat TM imagery to map greenhouses in Qingzhou,Shandong Province,China[J]. Pedosphere, 2004,14(3):363-369.
[7] Novelli A, Aguilar M A, Nemmaoui A , et al. Performance evaluation of object based greenhouse detection from Sentinel-2 MSI and Landsat 8 OLI data:A case study from Almería (Spain)[J]. International Journal of Applied Earth Observation and Geoinformation, 2016,52:403-411.
[8] Yang D D, Chen J, Zhou Y , et al. Mapping plastic greenhouse with medium spatial resolution satellite data:Development of a new spectral index[J]. Photogrammetry and Remote Sensing, 2017,128:47-60.
[9] 约翰逊, 威克恩 . 实用多元统计分析[M]. 陆漩译.2版.北京: 清华大学出版社, 2001.
[9] Johnson R A, Wichern D W. Applied Multivariate Statistical Analysis[M].Translated by Lu X. 2nd ed.Beijing: Tsinghua University Press, 2001.
[10] 许剑辉, 赵怡, 肖明虹 , 等. 基于空间自回归模型的广州市NDVI和NDBI与气温关系研究[J]. 国土资源遥感, 2018,30(2):186-194.doi: 10.6046/gtzyyg.2018.02.25.
[10] Xu J H, Zhao Y, Xiao M H , et al. Study on the relationship between NDVI and NDBI and temperature in Guangzhou based on spatial autoregressive model[J]. Remote Sensing for Land and Resources, 2018,30(2):186-194.doi: 10.6046/gtzyyg.2018.02.25.
[11] As-syakur A R, Adnyana I W S, Arthana I W , et al. Enhanced built-up and bareness index (EBBI) for mapping built-up and bare land in an urban area[J]. Remote Sensing, 2012,4(10):2957-2970.
[12] Xu H Q . Modification of normalised difference water index (NDWI) to enhance open water features in remotely sensed imagery[J]. International Journal of Remote Sensing, 2006,27(14):3025-3033.
[13] Kaufman Y J, Remer L . Detection of forests using mid-IR reflectance:An application for aerosol studies[J]. IEEE Transactions on Geoscience and Remote Sensing, 1994,32(3), 672-683.
[14] 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.
url: http://d.wanfangdata.com.cn/Periodical/ygjsyyy201106020
[14] Fang M, Zhang J L, Xu Z . 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.
[15] 王济川, 郭志刚 . Logistic回归模型:方法与应用[M]. 北京: 高等教育出版社, 2001.
[15] Wang J C, Guo Z G. Logistic Regression Model:Method and Application[M]. Beijing: Higher Education Press, 2001.
[16] Cox D R . The analysis of binary data[J]. Economica, 1971,150:210-211.
[1] Yueru WANG, Pengpeng HAN, Shujing GUAN, Yu HAN, Lin YI, Tinggang ZHOU, Jinsong CHEN. Information extraction of Dracaena sanderiana planting area based on Landsat8 OLI data[J]. Remote Sensing for Land & Resources, 2019, 31(1): 133-140.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
京ICP备05055290号-2
Copyright © 2017 Remote Sensing for Natural Resources
Support by Beijing Magtech