:: Volume 4, Issue 1 (6-2016) ::
jgit 2016, 4(1): 1-15 Back to browse issues page
Comparison of backscattering soil surface models to validate and inversion physical parameters
S.Mohammad MirMazloumi , Mahmod Reza Sahebi *
K.N.Toosi University of Technology
Abstract:   (4596 Views)

Using SAR data is the best and the most popular method for retrieving soil surface parameters. This research is trying to choose the best model to estimate soil surface parameters from SMEX03 data over southern Oklahoma. First, backscattering coefficients obtained from the Oh, Dubois, and IEM backscattering models in L, C, and P bands were compared with backscattering coefficients of the images in HH and VV polarization and the best model was chosen. Then soil roughness, dielectric constant and correlation length estimated from inversion backscattering models based on the mathematical algorithms and compared with corresponding result measured by ground truth.

 In backscatter coefficients comparison section, the Oh model in band C presented the most accurate results. In this section, Dubois and IEM models in band L had appropriate accuracies. However, their results were weaker than the best result. In retrieving soil surface parameters section, the main purpose of the research, variables solution with adjustment on IEM provided better accuracy for estimating all parameters; however, the Oh model in band C for estimating surface roughness and in band L for estimation of dielectric constant presented the appropriate results. Inaccurate results of band P was another result calculated in this research. The results of band P presented the least satisfactory results in this work. In conclusion, the results of empirical models are weaker than of the results obtained from the theoretical model (IEM). In addition, using more accurate ground truth data might result in obtaining more accurate validation part.

Keywords: Backscattering models, Remote Sensing, SAR, Soil surface parameters
Full-Text [PDF 1607 kb]   (1583 Downloads)    
Type of Study: Research | Subject: RS
Received: 2015/02/1 | Accepted: 2015/08/22 | Published: 2016/11/6

XML   Persian Abstract   Print

Rights and permissions
Creative Commons License This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
Volume 4, Issue 1 (6-2016) Back to browse issues page