Précédent|Prochain

Hybrid feature selection for cropland identification using GF-5 satellite image

  • role： First author第一作者
  • Affiliation：

    State Key Laboratory of Remote Sensing, Faculty of Geographical Science, Beijing Normal University, Beijing 100875, China

    Beijing Engineering Research Center for Global Land Remote Sensing Products, Beijing Normal University, Beijing 100875, China

  • Email：chenzhulin@mail.bnu.edu.cn
  • Introduction：陈珠琳，研究方向为资源环境遥感。E-mail： chenzhulin@mail.bnu.edu.cn
CHEN Zhulin12,  
  • role： Corresponding author通信作者
  • Affiliation：

    State Key Laboratory of Remote Sensing, Faculty of Geographical Science, Beijing Normal University, Beijing 100875, China

    Beijing Engineering Research Center for Global Land Remote Sensing Products, Beijing Normal University, Beijing 100875, China

  • Email：jiakun@bnu.edu.cn
  • Introduction：贾坤，研究方向为植被定量遥感，土地覆盖遥感分类和生态遥感。E-mail： jiakun@bnu.edu.cn
JIA Kun12*,  
  • Affiliation：

    Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100101, China

LI Qiangzi3,  
  • Affiliation：

    Land Satellite Remote Sensing Application Center, Ministry of Natural Resource of the People's Republic of China, Beijing 100048, China

XIAO Chenchao4,  
  • Affiliation：

    Land Satellite Remote Sensing Application Center, Ministry of Natural Resource of the People's Republic of China, Beijing 100048, China

WEI Dandan4,  
  • Affiliation：

    State Key Laboratory of Remote Sensing, Faculty of Geographical Science, Beijing Normal University, Beijing 100875, China

    Beijing Engineering Research Center for Global Land Remote Sensing Products, Beijing Normal University, Beijing 100875, China

ZHAO Xiang12,  
  • Affiliation：

    Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100101, China

WEI Xiangqin3,  
  • Affiliation：

    State Key Laboratory of Remote Sensing, Faculty of Geographical Science, Beijing Normal University, Beijing 100875, China

    Beijing Engineering Research Center for Global Land Remote Sensing Products, Beijing Normal University, Beijing 100875, China

YAO Yunjun12,  
  • Affiliation：

    Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100101, China

LI Juan3

résumé

Accurate farmland area identification is the basis of crop yield estimation and an important indicator in food security assessment. As an important data source for farmland identification, remote sensing data can provide dynamic and fast observation results for classification. GF-5, which is the only hyperspectral satellite in the China High-resolution Earth Observation System, has great research and application potential in farmland identification. However, the dimensionality curse caused by the redundant bands in hyperspectral data seriously affects the calculation speed and classification accuracy of models. To solve this problem, this research proposes a hybrid feature selection algorithm for farmland identification. First, on the basis of the feature importance provided by the feature selection algorithm, the feature dimension is gradually reduced from 295 to 5 with a step length of 10. The overall accuracy of the classification results corresponding to each feature dimension is recorded. Second, the turning point (a dimension number whose corresponding overall accuracy hardly decreases when the input variable number is smaller than it) is determined based on the overall accuracy, and the corresponding variables are adopted as the feature subset. Lastly, the Sequential Backward Selection (SBS) method is used to search for the best subset.Three feature selection algorithms (i.e., Random Forest (RF), Multi-Information (MI), and L1 regularization (L1)) and three classification algorithms (RF, Support Vector Machine (SVM), and K-Nearest Neighbor (KNN)) are examined. Results indicate that the autocorrelations of the three subsets differ significantly. Most of the bands selected by the MI method are continuous and concentrated in the blue and shortwave infrared range. Therefore, the extremely high autocorrelation that exists in this subset has a negative effect on classification accuracy. By contrast, the correlation between bands in the RF and L1 feature subsets is relatively weak. However, the two feature sets still result in different classification accuracy. According to the variable distribution, many red-edge and near-infrared bands are contained in the L1 feature subset. These bands demonstrate better ability to distinguish farmland, forest, and soil than the blue and red bands selected by the RF algorithm. The classification algorithms also have different capacities. In the high-dimensional space, the SVM algorithm exhibits high robustness to noise, resulting in high accuracy. However, when the dimension decreases to a critical value, the accuracy of SVM decreases sharply. By contrast, although RF is not as robust as SVM in the high-dimensional space, it has excellent generalization ability in the low-dimensional space. Compared with the subsets obtained after the first dimensionality reduction process, the optimal feature subsets obtained by SBS searching improve the classification accuracy of each model.The L1-SVM-SBS model with a 23-dimensional input achieves the highest overall classification accuracy (94.64%) and cropland recall rate (95.83%). This study provides a new method of farmland identification using hyperspectral data. By selecting numerous representative and informative bands, this method not only improves farmland classification accuracy, but can also be used as a reference for other classification problems involving hyperspectral remote sensing.

mots-clés

cropland identification;GF-5;feature selection;hyperspectral remote sensing;L1 regularization;sequential backward selection

References

  1. 1.
    Alvarez-Meza A M, Lee J A, Verleysen M and Castellanos-Dominguez G. 2017. Kernel-based dimensionality reduction using Renyi’s α-entropy measures of similarity. Neurocomputing, 222: 36-46
  2. 2.
    Aneece I and Thenkabail P. 2018. Accuracies achieved in classifying five leading world crop types and their growth stages using optimal earth observing-1 Hyperion hyperspectral narrowbands on Google earth engine. Remote Sensing, 10: 2027
  3. 3.
    Bolón-Canedo V and Alonso-Betanzos A. 2019. Ensembles for feature selection: a review and future trends. Information Fusion, 52: 1-12
  4. 4.
    Ding X H, Zhang S Q, Li H P, Wu P, Dale P, Liu L J and Cheng S. 2020. A restrictive polymorphic ant colony algorithm for the optimal band selection of hyperspectral remote sensing images. International Journal of Remote Sensing, 41(3): 1093-1117
  5. 5.
    Dong C and Zhao G X. 2020. Influence of time series data quality on land cover classification accuracy. Remote Sensing Technology and Application, 35(3): 558-566.
  6. 6.
    Dong X F, Gan F P, Li N, Yan B K, Zhang L, Zhao J Q, Yu J C, Liu R Y and Ma Y N. 2020. Fine mineral identification of GF-5 hyperspectral image. Journal of Remote Sensing, 24(4), 454-464
  7. 7.
    Fan D D, Li Q Z, Wang H Y, Zhang Y, Du X and Shen Y. 2019. Improvement in recognition accuracy of minority crops by resampling of imbalanced training datasets of remote sensing. Journal of Remote Sensing, 23(4): 730-742
  8. 8.
    Ghorbanian A, Kakooei M, Amani M, Mahdavi S, Mohammadzadeh A and Hasanlou M. 2020. Improved land cover map of Iran using sentinel imagery within Google earth engine and a novel automatic workflow for land cover classification using migrated training samples. ISPRS Journal of Photogrammetry and Remote Sensing, 167: 276-288
  9. 9.
    González J, Ortega J, Damas M, Martín-Smith P and Gan J Q. 2019. A new multi-objective wrapper method for feature selection-Accuracy and stability analysis for BCI. Neurocomputing, 333: 407-418
  10. 10.
    Han Z and Song W. 2019. Spatiotemporal variations in cropland abandonment in the Guizhou-Guangxi karst mountain area, China. Journal of Cleaner Production, 238: 117888
  11. 11.
    Hao P Y, Chen Z X, Tang H J, Li D D and Li H. 2019. New workflow of plastic-mulched farmland mapping using multi-temporal sentinel-2 data. Remote Sensing, 11(11): 1353
  12. 12.
    Htitiou A, Boudhar A, Lebrini Y, Hadria R, Lionboui H, Elmansouri L, Tychon B and Benabdelouahab T. 2019. The performance of random forest classification based on Phenological metrics derived from sentinel-2 and landsat 8 to map crop cover in an irrigated semi-arid region. Remote Sensing in Earth Systems Sciences, 2(4): 208-224
  13. 13.
    Jia K and Li Q Z. 2013. Review of features selection in crop classification using remote sensing data. Resources Science, 35(12): 2507-2516
  14. 14.
    Kussul N, Lavreniuk M, Skakun S and Shelestov A. 2017. Deep learning classification of land cover and crop types using remote sensing data. IEEE Geoscience and Remote Sensing Letters, 14(5): 778-782
  15. 15.
    Li M, Chen H, Shi X, Liu S, Zhang M and Lu S F. 2019. A multi-information fusion “triple variables with iteration” inertia weight PSO algorithm and its application. Applied Soft Computing, 84: 105677
  16. 16.
    Li Q Z. 2018. Prospect of grain production and supply service mode of China in internet plus era. China Agricultural Informatics, 30(1): 93-99
  17. 17.
    Li Y, Li T and Liu H. 2017. Recent advances in feature selection and its applications. Knowledge and Information Systems, 53(3): 551-577
  18. 18.
    Liu D and Sun K. 2019. Random forest solar power forecast based on classification optimization. Energy, 187: 115940
  19. 19.
    Liu J, Liu J K, An J J and Zhang C. 2020. Precise crop classification based on multi-features from time-series Landsat8 OLI images and random forest algorithm. Agricultural Research in the Arid Areas, 38(3): 281-288, 298
  20. 20.
    Liu X P, Li X, Tan Z Z and Chen Y M. 2011. Zoning farmland protection under spatial constraints by integrating remote sensing, GIS and artificial immune systems. International Journal of Geographical Information Science, 25(11): 1829-1848
  21. 21.
    Liu X S, Gong Z W and Wu J. 2018. Land use information extraction using multiple features derived from hyperspectral images. Journal of Nanjing Forestry University (Natural Science Edition), 42(4): 141-147
  22. 22.
    Liu Y N, Xun X D, Hu X N, Liu S F, Cao K Q, Chai M Y, Liao Q J, Zuo Z Q, Hao Z Y, Duan W B, Zhou W Y N, Zhang J and Zhang Y. 2020. Development of visible and short-wave infrared hyperspectral imager onboard GF-5 satellite. Journal of Remote Sensing, 24(4): 333-344
  23. 23.
    Maldonado S and López J. 2018. Dealing with high-dimensional class-imbalanced datasets: embedded feature selection for SVM classification. Applied Soft Computing, 67: 94-105
  24. 24.
    Park M Y and Hastie T. 2007. L1‐regularization path algorithm for generalized linear models. Journal of the Royal Statistical Society: Series B (Statistical Methodology), 69(4): 659-677
  25. 25.
    Radley S, Sybi C J and Premkumar K. 2020. Multi information amount movement aware- routing in FANET: flying ad-hoc networks. Mobile Networks and Applications, 25(2): 596-608
  26. 26.
    Sánchez-Maroño N, Alonso-Betanzos A and Tombilla-Sanromán M. 2007. Filter methods for feature selection – a comparative study//Proceedings of the 8th International Conference on Intelligent Data Engineering and Automated Learning. Birmingham, UK, December: Springer: 178-187
  27. 27.
    Sukawattanavijit C, Chen J and Zhang H S. 2017. GA-SVM algorithm for improving land-cover classification using SAR and optical remote sensing data. IEEE Geoscience and Remote Sensing Letters, 14(3): 284-288
  28. 28.
    Sylvester E V A, Bentzen P, Bradbury I R, Clément M, Pearce J, Horne J and Beiko R G. 2018. Applications of random forest feature selection for fine-scale genetic population assignment. Evolutionary Applications, 11(2): 153-165
  29. 29.
    Wang X X, Gao X W, Zhang Y Z, Fei X Y, Chen Z, Wang J, Zhang Y Y and Zhao H M. 2019. Land-cover classification of coastal wetlands using the RF algorithm for Worldview-2 and Landsat 8 images. Remote Sensing, 11(16): 1927
  30. 30.
    Wei Y H, Wang Y, He X M, Guo K and Chang R C. 2020. Method of terrain classification based on GF-5 satellite remote sensing images. Modern Electronics Technique, 43(18): 85-88
  31. 31.
    Xu L, Ming D P, Zhou W, Bao H Q, Chen Y Y and Ling X. 2019. Farmland extraction from high spatial resolution remote sensing images based on stratified scale pre-estimation. Remote Sensing, 11(2): 108
  32. 32.
    Yan X A and Jia M P. 2018. A novel optimized SVM classification algorithm with multi-domain feature and its application to fault diagnosis of rolling bearing. Neurocomputing, 313: 47-64
  33. 33.
    Yin H, Prishchepov A V, Kuemmerle T, Bleyhl B, Buchner J and Radeloff V C. 2018. Mapping agricultural land abandonment from spatial and temporal segmentation of Landsat time series. Remote Sensing of Environment, 210: 12-24
  34. 34.
    Yoo C, Han D, Im J and Bechtel B. 2019. Comparison between convolutional neural networks and random forest for local climate zone classification in mega urban areas using Landsat images. ISPRS Journal of Photogrammetry and Remote Sensing, 157: 155-170
  35. 35.
    Yu Q Y, Xiang M T, Wu W B and Tang H J. 2019. Changes in global cropland area and cereal production: an inter-country comparison. Agriculture, Ecosystems and Environment, 269: 140-147
  36. 36.
    Yuan J W, Wu C, Du B, Zhang L P and Wang S G. 2020. Analysis of landscape pattern on urban land use based on GF-5 hyperspectral data. Journal of Remote Sensing, 24(4): 465-478
  37. 37.
    Zhang S C. 2020. Cost-sensitive KNN classification. Neurocomputing, 391: 234-242
  38. 38.
    Zhou X L, Su G Q, Wang L J, Nie S D and Ge X M. 2017. The inversion of 2D NMR relaxometry data using L1 regularization. Journal of Magnetic Resonance, 275: 46-54

Lire l'article complet

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