Land Cover Feature Extraction and Analysis using Biogeography based Optimization (BBO) Algorithm

Abstract

The entropy vector is used to identify the homogeneity between the regions but in this proposed approach we are using a wavelet inspired segmented approach for the region match so that more clearly differentiate the homogenous and heterogeneous regions over the image. In this existing work, the K-means clustering is used as the initial classification and then the BBO is implemented. An initial segmented analysis is implemented to identify the number of classes more accurately and then the classification is performed. The main objective of the work is to perform the region classification for the land cover images development and implement wavelet inspired BBO approach to perform the classification process. The initial segmentation based similarity measure is performed to identify the number of classes using Clustering method and then the clustering process is done using Cmeans. The major objective of the work is to define more efficient classification under different regions.