With Weiku oasis in Xinjiang as the study area, the authors used two polarization methods, i.e., Freeman-Durden and H/α, to decompose and treat 4-polarization data of the Radarsat-2, got the corresponding characteristic parameters, extracted the salinization information of the study area combined with the SVM-Wishart semi-supervised classification method, and finally checked and analyzed the result of the classification with the visual interpretation and the field investigation. Some conclusions have been reached: ① When the impact categories are identified and the parameter feature space is built to get the characteristic parameters, different polarization decompositions yield different resolutions of parameter information, and the distributions of parameters characteristic space are different; after decomposing with H/α, the characteristic space constituted by characteristic parameters are different; ② The effect of using semi-supervised classification method to classify the endings of the Freeman Durden and H/α,Freeman Durden classification is superior to that of H/a; ③SVM-Wishart semi-supervised classification is superior to traditional SVM classification and hence it can be well used to extract the salinization information. SVM-Wishart semi-supervised classification can fully excavate the characteristic parameters after the coherent decomposition of polarization and can improve the classification accuracy, and it has certain advantages in the extraction of salinization information.
Ding J L, Yao Y, Wang F , et al. Detecting soil salinization in arid regions using spectralfeature space derived from remote sensing data[J]. Acta Ecologica Sinica, 2014,34(16):4620-4631.
Cao L, Ding J L, Yu H Y . Relationship between multi-scale landscape pattern and salinity in Weigan and Kuqa Rivers Delta oasis[J]. Transactions of the Chinese Society of Agricultural Engineering, 2016,32(3):101-110.
Nurmemet I, Ghulam A, Tiyip T , et al. Monitoring soil salinization in Keriya River Basin,Northwestern China using passive reflective and active microwave remote sensing data[J]. Remote Sensing, 2015,7(7):8803-8829.
doi: 10.3390/rs70708803
url: http://www.mdpi.com/2072-4292/7/7/8803
Lhissou R, Chokmani K, El Harti A, et al. Soil salinity estimation using RADARSAT 2 polarimetric data in arid and sub-arid regions:Morocco and Tunisia cases[C]// EGU General Assembly Conference.Vienna:EGU, 2013.
[14]
Li Y Y, Zhao K, Ding Y L, et al. An empirical method for soil salinity and moisture inversion in west of Jilin[C]// Proceedings of the International Conference on Remote Sensing Environment and Transportation Engineering(RSETE).Nanjing:AISR, 2013: 19-21.
Ding J L, Zhang F, Jiang H N , et al. Spatial variability of soil conductivity and salt content in the north Tarim Basin:A case study in the delta oasis of Weigen-Kuqa Rivers[J]. Arid Land Geography, 2008,31(4):624-632.
Yao Y, Ding J L, Lei L , et al. Monitoring spatial variability of soil salinity in dry and wet seasons in the North Tarim Basin using remote sensing and electromagnetic induction instruments[J]. Acta Ecologica Sinica, 2013,33(17):5308-5319.
Ding J L, Yao Y, Wang F . Quantitative remote sensing of soil salinization in arid regions based on three dimensional spectrum eigen spaces[J]. Acta Pedologica Sinica, 2013,50(5):853-861.
Cloude S R, Pottier E . An entropy based classification scheme for land applications of polarimetric SAR[J]. IEEE Transactions on Geoscience and Remote Sensing, 1997,35(1):68-78.
doi: 10.1109/36.551935
url: http://ieeexplore.ieee.org/document/551935/
[22]
Lee J S, Grunes M R, Pottier E , et al. Unsupervised terrain classification preserving polarimetric scattering characteristics[J]. IEEE Transactions on Geoscience and Remote Sensing, 2004,42(4):722-731.
doi: 10.1109/TGRS.2003.819883
url: http://ieeexplore.ieee.org/document/1288367/
2019,
Vol. 31
), Wenyu WEI1,2
