Change detection using image differencing: A study over area surrounding Kumta, India

Major resources for spatial change detection are remotely sensed data. In this paper, we discuss the change detection using image differencing technique over Kumta region, Karnataka, India. Accurate change detection results are very important to know the relationship between human interactions with environment in order to have a better understanding and decision making. An attempt has been made to recognize the changes in Land-Use/Land-Cover (LU/LC) over a period of four years from 2012 to 2016. Support Vector Machine (SVM) algorithm is used to classify the study area. A considerable amount of classification accuracy has been attained to have a very decent image differencing results. No major misclassifications were observed between the classes and change detection statistics are presented in terms of percentage of change and are also analyzed with respect to annual rainfall over the considered area.