Enhanced classification algorithms for the satellite image processing

Abstract

Background and objectives: Region based approach to determine the used land details of Vellore district to improvise the vegetation in future. Methods and Statistical Analysis: To classify the used lands Modified KNN and Modified SVM algorithm are applied to get the classification. Accuracy is measured by producer's accuracy, user's accuracy, commission error and omission error. Results: MOKNN and MOSVM produce the improved accuracy in classification to determine the exact land use details. Conclusion: Proposed MOKNN and MOSVM give the improved accuracy to predict used land details. This work may be extended with combined data set to get more improved accuracy.