Anterior|Siguiente

Identification of main crops based on the univariate feature selection in Subei

  • role： First author第一作者
  • Affiliation：

    Renewable Energy Laboratory of Institute of Remote Sensing and Digital Earth, Chinese Academy of Sciences, Beijing 100101, China

    College of Resources and Environmental, University of Chinese Academy of Sciences, Beijing 100049, China

  • Email：wangna@radi.ac.cn
  • Introduction：王娜(1990—　)，女，硕士研究生，研究方向为农业遥感分类方法和目标识别。E-mail：wangna@radi.ac.cn
WANG Na12,  
  • role： Corresponding author通信作者
  • Affiliation：

    Renewable Energy Laboratory of Institute of Remote Sensing and Digital Earth, Chinese Academy of Sciences, Beijing 100101, China

  • Email：liqz@radi.ac.cn
  • Introduction：李强子(1970—　)，男，研究员，研究方向为农业遥感和生态遥感。E-mail：liqz@radi.ac.cn
LI Qiangzi1*,  
  • Affiliation：

    Renewable Energy Laboratory of Institute of Remote Sensing and Digital Earth, Chinese Academy of Sciences, Beijing 100101, China

DU Xin1,  
  • Affiliation：

    Renewable Energy Laboratory of Institute of Remote Sensing and Digital Earth, Chinese Academy of Sciences, Beijing 100101, China

ZHANG Yuan1,  
  • Affiliation：

    Renewable Energy Laboratory of Institute of Remote Sensing and Digital Earth, Chinese Academy of Sciences, Beijing 100101, China

ZHAO Longcai1,  
  • Affiliation：

    Renewable Energy Laboratory of Institute of Remote Sensing and Digital Earth, Chinese Academy of Sciences, Beijing 100101, China

WANG Hongyan1

resumen

Timely accurate crop type identification and Crop Acreage Estimates (CAE) are essential for food security. Remote sensing technology has been successfully applied to crop identification because of its macro, rapid monitoring capabilities at large scales and its ability to quickly obtain accurate agricultural information. However, when identifying crop types, both simple and too many identifiable features might lead to low classification accuracies. Thus, multi-source and optimally selected features are obviously crucial to crop classification using remotely-sensed images. This paper considered a series of features, including multi-temporal spectra, vegetation indexes, textures, and band differences. Multiple experiments were designed and conducted in Sihong County, Jiangsu Province, China using Gaofen-1 and Huanjing-1 images to evaluate the influence of different features on the identification accuracy and determine the combination of preferred features which can improve the classification effect. The combination of random forest classification and univariate feature selection methods was expected to have a considerably positive effect on distinguishing and extracting the main crops in remote sensing images. In this study, the crop classification was implemented using random forests and univariate feature selection. The random forest method, which constructs many CART decision trees during each classification process, is one of themost effective classification methods. Univariate feature selection is a statistical testing method, which tests each feature to measure the relationship between the feature and the corresponding variable and then removes features that obtain low scores. First, the random forest classifier was applied to classify the images using the preceding multisource features mentioned. Second, we analyzed the contributions of different types of features or feature combinations to the classification accuracy. Third, features were selected by using the univariate feature selection method. Finally, we re-combined the optimal features and random forest to classify the image and distinguish the main crop types with high accuracy. The results showed that overall classification accuracy based on the combination of optimal features reached 97.07% with the corresponding Kappa coefficient being 0.96, which indicated that the feature selection method used in this paper has a considerably positive effect on high classification accuracy because it efficiently reduced feature dimension. The classification results also showed that the crop classification using multi-source features outperformed the one which only used spectral features. In addition, the accuracy of the experiment which simultaneously used spectral and VI features was the second highest among all experiments. The optimal feature combination has 19 features, including five spectral features, six vegetation indexes, seven band difference features, and 1 texture feature, which suggested that vegetation indexes and band differences were more important to the crop identification than the other two. This study demonstrated the following: (1) the addition of different types of features could improve classification accuracy; (2) too many features would decrease classification accuracies; (3) univariate feature selection was effective for choosing the optimal subset of features. The optimally selected features can be relatively beneficial to reduce the computation load and improve the worse accuracies caused by applied features blindly. Therefore, the combination of random forest and univariate feature selection is effective in improving classification accuracy and efficiency.

palabra clave

univariate feature selection;spectrum feature;Vegetation Index (VI) feature;texture feature;band difference features

References

  1. 1.
    Atkinson J T, Ismail R and Robertson M. 2014. Mapping bugweed (solanum mauritianum) infestations in pinus patula plantations using hyperspectral imagery and support vector machines. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 7(1): 17–28
  2. 2.
    Blum A L and Langley P. 1997. Selection of relevant features and examples in machine learning. Artificial Intelligence, 97(1/2): 245–271
  3. 3.
    Breiman L. 2001. Random forest. Machine Learning, 45(1): 5–32
  4. 4.
    Chakraborty M and Panigrahy S. 2000. A processing and software system for rice crop inventory using multi-date RADARSAT ScanSAR data. Isprs Journal of Photogrammetry and Remote Sensing, 55(2): 119–128
  5. 5.
    Chen Q H. 2007. Researeh and realization of multi-source remote sensing data on objeet oriente. Wuhan: China University of Geosciences: 17–18
  6. 6.
    Chen X M, Shen Z F, Xing T Y, Xia L G and Wu T J. 2016. Efficiency and accuracy analysis of multispectral image classification based on mRMR feature selection method. Journal of Geo-information Science, 18(6): 816–823
  7. 7.
    Ding Y P. 2013. Research on dryland crops identification and area extraction method based on microwave remote sensing. Beijing: Chinese Academy of Agricultural Sciences: 4–42
  8. 8.
    Drotár P, Gazda J and Smékal Z. 2015. An experimental comparison of feature selection methods on two-class biomedical datasets. Computers in Biology and Medicine, 66: 1–10
  9. 9.
    Jarvis E. 2016. The introduction of some common feature selction methods with scikit-learn[EB/OL]. 2016-06-27.
  10. 10.
    Jia K, Li Q Z, Tian Y C, Wu B F, Zhang F F and Meng J H. 2012. Crop classification using multi-configuration SAR data in the North China Plain. International Journal of Remote Sensing, 33(1): 170–183
  11. 11.
    Jia K and Li Q Z. 2013. Review of features selection in crop classification using remote sensing data. Resources Science, 35(12): 2507–2516
  12. 12.
    Kaya G T. 2013. A comprehensive analysis of earthquake damage patterns using high dimensional model representation feature selection//Proc. SPIE 8892, Image and Signal Prcessing for Remote Sensing XIX. Dresden, Germany: SPIE [DOI: 10.1117/12. 2030100]
  13. 13.
    Lei Z. 2012. Random Forest and its application in remote sensing. Shanghai: Shanghai Jiaotong University: 9–12
  14. 14.
    Liu H and Yu L. 2005. Toward integrating feature selection algorithms for classification and clustering. IEEE Transactions on Knowledge and Data Engineering, 17(4): 491–502
  15. 15.
    Liu Y, Fu Z Y and Zheng F B. 2015. Review on high resolution remote sensing image classification and recognition. Journal of Geo-information Science, 17(9): 1080–1091
  16. 16.
    Lu D and Weng Q. 2007. A survey of image classification methods and techniques for improving classification performance. International Journal of Remote Sensing, 28(5): 823–870
  17. 17.
    Ma Y, Jiang Q G, Meng Z G, Li Y H, Wang D and Liu H X. 2016. Classification of land use in farming area based on Random Forest algorithm. Transactions of The Chinese Society of Agricultural Machinery, 47(1): 297–303
  18. 18.
    Persello C and Bruzzone L. 2009. A novel approach to the Selection of spatially invariant features for classification of hyperspectral images//2009 IEEE International Geoscience and Remote Sensing Symposium. Cape Town: IEEE: II-61-II-64 [DOI: 10.1109/ igarss.2009.5418001]
  19. 19.
    Ren J Q, Chen Z X, Tang H J, Zhou Q B and Qin J. 2011. Regional crop yield simulation based on crop growth model and remote sensing data. Transactions of the CSAE, 27(8): 257–264
  20. 20.
    Soares J V, Rennó C D, Formaggio A R, da Costa Freitas Yanasse C and Frery A C. 1997. An investigation of the selection of texture features for crop discrimination using SAR imagery. Remote Sensing of Environment, 59(2): 234–247
  21. 21.
    Tuia D, Pacifici F, Kanevski M and Emery W J. 2009. Classification of very high spatial resolution imagery using mathematical morphology and support vector machines. IEEE Transactions on Geoscience and Remote Sensing, 47(11): 3866–3879
  22. 22.
    Wang D, Lin H, Chen J S, Zhang Y Z and Zeng Q W. 2010. Application of multi-temporal ENVISAT ASAR data to agricultural area mapping in the Pearl River Delta. International Journal of Remote Sensing, 31(6): 1555–1572
  23. 23.
    Wang S Y, Zhang Z W and Yu Z H. 2014. Classification of Honghe Wetland remote sensing image based on random forests. Geomatics and Spatial Information Technology, 37(4): 83–93
  24. 24.
    Wu B F. 2004. China crop watch system with remote sensing. Joural of Remote Sensing, 8(6): 481–497
  25. 25.
    Yang C, Liu S C, Bruzzone L, Guan R C and Du P J. 2013. A feature-metric-based affinity propagation technique for feature selection in hyperspectral image classification. IEEE Geoscience and Remote Sensing Letters, 10(5): 1152–1156
  26. 26.
    Yang J W, Zhang J S, Zhu X F, Xie D F and Yuan Z M Q. 2015. Random Forest applied for dimension reduction and classification in hyperspectral data. Journal of Beijing Normal University (Natural Science), 51(S1): 82–88
  27. 27.
    Yao G J, Yang and Zhan X J. 2014. Feature selection agorithm based on random forest. Journal of Jilin University (Engineering and Technology Edition), 44(1): 137–141
  28. 28.
    Zortea M and Haertel V. 2004. Experiments on feature extraction in remotely sensed hyperspectral image data//IEEE International Geoscience and Remote Sensing Symposium. [s.l.]: IEEE: 964–967[DOI: 10.1109/igarss.2004.1368569]

Leer el texto completo

The above content is generated by Large Model Translation. The translated content is for reference only. We do not assume any commercial or legal responsibilty for any consequences arising from the use of our website