Früher|Nächster

Spectral unmixing method considering endmember variability of vegetation

  • role： First author第一作者
  • Affiliation：

    School of Geography, South China Normal University, Guangzhou 510631, China

  • Email：weiiqnhua@m.scnu.edu.cn
  • Introduction：韦钦桦，研究方向为高光谱遥感图像处理。E-mail：weiiqnhua@m.scnu.edu.cn
WEI Qinhua,  
  • role： Corresponding author通信作者
  • Affiliation：

    School of Geography, South China Normal University, Guangzhou 510631, China

  • Email：luowenfei@m.scnu.edu.cn
  • Introduction：罗文斐，研究方向为高光谱遥感图像处理、混合像元分解及其应用。E-mail：luowenfei@m.scnu.edu.cn
LUO Wenfei*,  
  • Affiliation：

    School of Geography, South China Normal University, Guangzhou 510631, China

TANG Kaifeng

Resümee

As ground features are affected by various factors, the problem of endmember variability will occur. Endmember variability greatly affects the accuracy of spectral unmixing results. This study is based on the vegetation-soil binary scene, and spectral unmixing is performed under the framework of NMF (Nonnegative Matrix Factorization). The PROSAIL model is used to describe the variability of vegetation endmembers from the mechanism so that the results of spectral unmixing have clear physical meaning. To improve efficiency, we set up two neural networks for model calculation and model inversion. In this way, the spectral unmixing algorithm can obtain the endmembers of the vegetation pixel by pixel, which can more accurately describe the variability of the vegetation endmember.In addition, there is diversity in the spatial resolution of remote sensing data products. The problem of the scale effect is widespread and is a key issue in the field of remote sensing. Analysis of the reason is largely due to the mixed pixels that universally exist. The method studied in this paper describes the variability in the vegetation endmember. A spectral unmixing algorithm that can describe the variability of vegetation endmembers pixel by pixel is obtained. This result can be used to invert vegetation parameters. Therefore, this study attempts to correct the scale effect of remote sensing products by considering the spectral unmixing method of vegetation endmember variability.This paper takes the LAI scale effect as an example. The effectiveness of the method is verified by an Unmanned Aerial Vehicle (UAV) image experiment. Three subimages were selected, and then they were resampled to two different levels of spatial resolution for experimentation. Among them, exponential function fitting is performed through the simulated spectrum of the PROSAIL model, and the relationship model is constructed to invert the LAI. The experimental results show that (1) the spectral unmixing method that uses the PROSAIL model to describe the variability of vegetation endmembers can obtain higher unmixing accuracy; (2) after using this spectral unmixing method, the Root Mean Square Error (RMSE) of the LAI scale effect is significantly reduced, and it has a certain effect on the correction of the LAI scale effect. This can improve the remote sensing scale effect problem to a certain extent.In summary, the spectral unmixing method has a certain effect on the correction of LAI scale differences and can improve the problem of the remote sensing scale effect to a certain extent. However, the research has the following issues that need further consideration: (1) This article only considers the vegetation-soil binary scene, but this method has not verified the multiple endmember scene. (2) The variability of soil and other background endmembers can be further considered. (3) The model can be further optimized to reduce the difference between the PROSAIL model spectrum and the real image spectrum, thereby improving the accuracy of unmixing. (4) In this paper, the evaluation is carried out by the method of upscaling. In the future, more complicated practical factors can be considered for evaluation. Simultaneously, real images collected at different flight altitudes can also be used for evaluation.

Schlüsselwort

remote sensing;spectral unmixing;endmember variability;prosail model;leaf area index;neural network;scale effect

References

  1. 1.
    Asner G P, Bustamante M M C and Townsend A R. 2003. Scale dependence of biophysical structure in deforested areas bordering the Tapajós National Forest, Central Amazon. Remote Sensing of Environment, 87(4): 507-520
  2. 2.
    Bateson C A, Asner G P and Wessman C A. 2000. Endmember bundles: a new approach to incorporating endmember variability into spectral mixture analysis. IEEE Transactions on Geoscience and Remote Sensing, 38(2): 1083-1094
  3. 3.
    Berry M W, Browne M, Langville A N, Pauca V P and Plemmons R J. 2007. Algorithms and applications for approximate nonnegative matrix factorization. Computational Statistics and Data Analysis, 52(1): 155-173
  4. 4.
    Chen Y H, Zhang W C and Yong B. 2007. Retrieving leaf area index using a neural network based on classification knowledge. Acta Ecologica Sinica, 27(7): 2785-2793
  5. 5.
    Drumetz L, Chanussot J and Jutten C. 2016a. Variability of the endmembers in spectral unmixing: recent advances//8th Workshop on Hyperspectral Image and Signal Processing: Evolution in Remote Sensing (WHISPERS). Los Angeles: IEEE: 1-5
  6. 6.
    Drumetz L, Veganzones M A, Henrot S, Phlypo R, Chanussot J and Jutten C. 2016b. Blind hyperspectral unmixing using an extended linear mixing model to address spectral variability. IEEE Transactions on Image Processing, 25(8): 3890-3905
  7. 7.
    Du X X, Zare A, Gader P and Dranishnikov D. 2014. Spatial and spectral unmixing using the beta compositional model. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 7(6): 1994-2003
  8. 8.
    Eches O, Dobigeon N, Mailhes C and Tourneret J Y. 2010. Bayesian estimation of linear mixtures using the normal compositional model. Application to hyperspectral imagery. IEEE Transactions on Image Processing, 19(6): 1403-1413
  9. 9.
    Eismann M T and Hardie R C. 2004. Stochastic spectral unmixing with enhanced endmember class separation. Applied Optics, 43(36): 6596-6608
  10. 10.
    Feret J B, François C, Asner G P, Gitelson A A, Martin R E, Bidel L P R, Ustin S L, Le Maire G and Jacquemoud S. 2008. PROSPECT-4 and 5: advances in the leaf optical properties model separating photosynthetic pigments. Remote Sensing of Environment, 112(6): 3030-3043
  11. 11.
    Gillis N and Glineur F. 2010. Using underapproximations for sparse nonnegative matrix factorization. Pattern Recognition, 43(4): 1676-1687
  12. 12.
    Guo Y K, Liu Y L, Zhang X J and Xu M. 2020. LAI inversion using radiation transfer model and random forest regression. Engineering of Surveying and Mapping, 29(3): 33-38
  13. 13.
    Hasan U, Sawut M, Chen S S and Li D. 2020. Inversion of leaf area index of winter wheat based on GF-1/2 image. Acta Agronomica Sinica, 46(5): 787-797
  14. 14.
    Heinz D, Chang C I and Althouse M L G. 1999. Fully constrained least-squares based linear unmixing [hyperspectral image classification]//IEEE 1999 International Geoscience and Remote Sensing Symposium. Hamburg: IEEE: 1401-1403
  15. 15.
    Hong D F, Yokoya N, Chanussot J and Zhu X X. 2019. An augmented linear mixing model to address spectral variability for hyperspectral unmixing. IEEE Transactions on Image Processing, 28(4): 1923-1938
  16. 16.
    Jacquemoud S and Baret F. 1990. PROSPECT: a model of leaf optical properties spectra. Remote Sensing of Environment, 34(2): 75-91
  17. 17.
    Jacquemoud S, Verhoef W, Baret F, Bacour C, Zarco-Tejada P J, Asner G P, François C and Ustin S L. 2009. PROSPECT + SAIL models: a review of use for vegetation characterization. Remote Sensing of Environment, 113 Suppl 1: S56-S66
  18. 18.
    Jiang J L, Ji X S, Yao X, Tian Y C, Zhu Y, Cao W X and Cheng T. 2018. Evaluation of three techniques for correcting the spatial scaling bias of leaf area index. Remote Sensing, 10(2): 221
  19. 19.
    Lee D D and Seung H S. 1999. Learning the parts of objects by non-negative matrix factorization. Nature, 401(6755): 788-791
  20. 20.
    Li J J, Zhu X H, Ma L L, Zhao Y G, Qian Y G and Tang L L. 2017. Leaf area index retrieval and scale effect analysis of multiple crops from UAV-based hyperspectral data. Remote Sensing Technology and Application, 32(3): 427-434
  21. 21.
    Li X C, Xu X G, Bao Y S, Huang W J, Luo J H, Dong Y Y, Song X Y and Wang J H. 2012. Retrieving LAI of winter wheat based on sensitive vegetation index by the segmentation method. Scientia Agricultura Sinica, 45(17): 3486-3496
  22. 22.
    Li X W. 2005. Retrospect, prospect and innovation in quantitative remote sensing. Journal of Henan University (Natural Science), 35(4): 49-56
  23. 23.
    Li X W. 2006. Review of the project of quantitative remote sensing of major factors for spatial-temporal heterogeneity on the land surface. Advances in Earth Science, 21(8): 771-780
  24. 24.
    Li X W, Wang J D and Strahler A H. 1999. Scale effect of Planck’s law over nonisothermal blackbody surface. Science in China Series E: Technological Sciences, 42(6): 652-656
  25. 25.
    Li X W, Wang J D and Strahler A H. 2000. Scale effects and scaling-up by geometric-optical model. Science in China Series E: Technological Sciences, 43(1): 17-22
  26. 26.
    Li X W and Wang Y T. 2013. Prospects on future developments of quantitative remote sensing. Acta Geographica Sinica, 68(9): 1163-1169
  27. 27.
    Li X W, Zhao H R, Zhang H and Wang J D. 2002. Global change study and quantitative remote sensing for land surface parameters. Earth Science Frontiers, 9(2): 365-370
  28. 28.
    Li Y N, Lu L and Liu Y. 2017. Tasseled cap triangle (TCT)-leaf area index (LAI) model of rice fields based on PROSAIL model and its application. Chinese Journal of Applied Ecology, 28(12): 3976-3984
  29. 29.
    Liang L, Di L P, Zhang L P, Deng M X, Qin Z H, Zhao S H and Lin H. 2015. Estimation of crop LAI using hyperspectral vegetation indices and a hybrid inversion method. Remote Sensing of Environment, 165: 123-134
  30. 30.
    Lin H, Liang L, Zhang L P and Du P J. 2013. Wheat leaf area index inversion with hyperspectral remote sensing based on support vector regression algorithm. Transactions of the Chinese Society of Agricultural Engineering, 29(11): 139-146
  31. 31.
    Liu L Y. 2014a. Vegetation Quantitative Remote Sensing Principles and Applications. Beijing: Science Press
  32. 32.
    Liu L Y. 2014b. Simulation and correction of spatialscaling effects for leaf area index. Journal of Remote Sensing, 18(6): 1158-1168
  33. 33.
    Ma J J and Yuan J G. 2014. Relationship between NDVI, view zenith angle and LAI based on model simulation. Remote Sensing Technology and Application, 29(4): 539-546
  34. 34.
    Ma Y J. 2019. Calibration of Regional MODIS LAI Products Considering the Difference of Model, Scale and Image Parameter. Beijing: China University of Geosciences (Beijing): 50-56
  35. 35.
    Pan H Z and Chen Z X. 2018. Application of UVA hyperspectral remote sensing in winter wheat leaf area index inversion. Chinese Journal of Agricultural Resources and Regional Planning, 39(3): 32-37
  36. 36.
    Pinty B, Lavergne T, Widlowski J L, Gobron N and Verstraete M M. 2009. On the need to observe vegetation canopies in the near-infrared to estimate visible light absorption. Remote Sensing of Environment, 113(1): 10-23
  37. 37.
    Ren H Y, Zhao Y L, Li X B and Ge Y J. 2020. Cultivated land fragmentation in mountainous areas based on different resolution images and its scale effects. Geographical Research, 39(6): 1283-1294
  38. 38.
    Roberts D A, Gardner M, Church R, Ustin S, Scheer G and Green R O. 1998. Mapping chaparral in the Santa Monica Mountains using multiple endmember spectral mixture models. Remote Sensing of Environment, 65(3): 267-279
  39. 39.
    Schmidt F, Schmidt A, Treguier E, Guiheneuf M, Moussaoui S and Dobigeon N. 2010. Implementation strategies for hyperspectral unmixing using bayesian source separation. IEEE Transactions on Geoscience and Remote Sensing, 48(11): 4003-4013
  40. 40.
    Somers B, Asner G P, Tits L and Coppin P. 2011. Endmember variability in spectral mixture analysis: a review. Remote Sensing of Environment, 115(7): 1603-1616
  41. 41.
    Tao H L, Feng H K, Yang G J, Yang X D, Liu M X and Liu S B. 2020. Leaf area index estimation of winter wheat based on UAV imaging hyperspectral imagery. Transactions of the Chinese Society for Agricultural Machinery, 51(1): 176-187
  42. 42.
    Thouvenin P A, Dobigeon N and Tourneret J Y. 2016. Hyperspectral unmixing with spectral variability using a perturbed linear mixing model. IEEE Transactions on Signal Processing, 64(2): 525-538
  43. 43.
    Tian Y H, Woodcock C E, Wang Y J, Privette J L, Shabanov N V, Zhou L M, Zhang Y, Buermann W, Dong J R, Veikkanen B, Häme T, Andersson K, Ozdogan M, Knyazikhin Y and Myneni R B. 2002. Multiscale analysis and validation of the MODIS LAI product: I. Uncertainty assessment. Remote Sensing of Environment, 83(3): 414-430
  44. 44.
    Verhoef W. 1984. Light scattering by leaf layers with application to canopy reflectance modeling: the SAIL model. Remote Sensing of Environment, 16(2): 125-141
  45. 45.
    Verhoef W, Jia L, Xiao Q and Su Z. 2007. Unified optical-thermal four-stream radiative transfer theory for homogeneous vegetation canopies. IEEE Transactions on Geoscience and Remote Sensing, 45(6): 1808-1822
  46. 46.
    Wang L J and Niu Z. 2014. Sensitivity analysis of vegetation parameters based on PROSAIL model. Remote Sensing Technology and Application, 29(2): 219-223
  47. 47.
    Wu H and Li Z L. 2009. Scale issues in remote sensing: a review on analysis, processing and modeling. Sensors, 9(3): 1768-1793
  48. 48.
    Xiao Y F, Zhou D M, Gong H L and Zhao W J. 2015. Sensitivity of canopy reflectance to biochemical and biophysical variables. Journal of Remote Sensing, 19(3): 368-374
  49. 49.
    Yang B,Luo W F.2015.Constrained NMF-based high-dimension adaptive particle swarm optimization algorithm for endmember extraction from a hyper spectral remote sensing image.Journal of Remote Sensing, 19(2):240-253.
  50. 50.
    Yang Y W. 2018. LAI Scale Effect study in Heihe Oasis Based on CASI Data. Chengdu: Chengdu University of Technology: 5-54
  51. 51.
    Yu Q H, Yang G J and Wang C C. 2019. Chlorophyll inversion of winter wheat based on ground hyperspectral data and PROSAIL model. Science of Surveying and Mapping, 44(11): 96-102, 136
  52. 52.
    Zare A and Ho K C. 2014. Endmember variability in hyperspectral analysis: addressing spectral variability during spectral unmixing. IEEE Signal Processing Magazine, 31(1): 95-104
  53. 53.
    Zhan H M and Xu S Z. 2011. Generalized linear mixed model for segregation distortion analysis. BMC Genetics, 12: 97
  54. 54.
    Zhang R H, Sun X M, Su H B, Tang X Z and Zhu Z L. 1999. Remote sensing and scale transfering of levity parameters on earth surface. Remote Sensing for Land and Resources, 41(3): 51-58
  55. 55.
    Zhao C H, Cui S L and Liu W. 2014. Multi-endmember hierarchical mixture analysis algorithm for spectra. Journal of Optoelectronics·Laser, 25(9): 1830-1836

Lesen Sie die ganze Passage

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