Support Vector Machine with Linear and Radial Bias Function to Classify the Soil Erosion and Land Degradation

The analysis of Land Use and Land Cover (LULC) changes needed update for remotely sensed information for assessing its impacts in erosion. This is especially significant evaluation to regions with semi-arid landscapes the soils cause scarce and appropriate management requirements corresponding LULC for resident conditions to obtain the suitable land use. To addresses this issue, in this article developed the Support Vector Machine (SVM) with linear and Radial Bias Function (RBF) which classifies the land degradation and erosion effectively. The developed SVM with linear and RBF method which is able to handle the mask generation, ranking and overfitting problems. Moreover, it minimizes the computational overhead through dividing the samples from various classes. The images are uniformly rescaled by using min-max normalization and the meaningful features are extracted by utilizing Normalized Difference Vegetation Index (NDVI) and Bare Soil Index (BSI) techniques. Then, it is effectively classified by using the developed SVM with linear and RBF technique.