Abstract
Abstract. In this paper, a framework is developed based on Support Vector Machines (SVM) for crop classification using polarimetric features extracted from multi-temporal Synthetic Aperture Radar (SAR) imageries. The multi-temporal integration of data not only improves the overall retrieval accuracy but also provides more reliable estimates with respect to single-date data. Several kernel functions are employed and compared in this study for mapping the input space to higher Hilbert dimension space. These kernel functions include linear, polynomials and Radial Based Function (RBF). The method is applied to several UAVSAR L-band SAR images acquired over an agricultural area near Winnipeg, Manitoba, Canada. In this research, the temporal alpha features of H/A/α decomposition method are used in classification. The experimental tests show an SVM classifier with RBF kernel for three dates of data increases the Overall Accuracy (OA) to up to 3% in comparison to using linear kernel function, and up to 1% in comparison to a 3rd degree polynomial kernel function.
Highlights
Crop classification and mapping using synthetic aperture radar (SAR) is an important application of remote sensing and earth observation technology
The possibility of identifying the individual classes is based on the fact that the dielectric properties and the structure of the different crop types are different
We have developed a framework for the classification of the temporal alpha features of H/A/α decomposition method, using Support Vector Machines (SVM) classifier and studied the effect of various kernel function on classification accuracy and performance
Summary
Crop classification and mapping using synthetic aperture radar (SAR) is an important application of remote sensing and earth observation technology. The discrimination potential of SAR data is based on the sensitivity of the radar backscatter to the dielectric properties of the objects and their structure (i.e., the size, shape, and orientation distribution of the scatterers) [1], [2]. The radar backscatter is sensitive to, e.g., the dielectric properties of the soil, the surface roughness, the terrain slope, and the vegetation canopy structure (e.g., the row direction and spacing, and the cover fraction) [1]. These properties are not necessarily specific for the individual classes and may cause variability of the backscatter within the classes. Differences in the development stages at a specific point in time due to, for instance, differences in sowing time may cause such variability [2]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
