Land use and land cover classification of individual and stacked satellite images using supervised maximum likelihood method

Digital image classification is the process of sorting pixels into a finite number of individual classes, or categories of data, based on their data file values. If a pixel satisfies a certain set of criteria, the pixel is assigned to the class that corresponds to those criteria. In the present study, supervised maximum likelihood (ML) classification method has been used to classify images of Ranchi area of the years 2008, 2009 and stacked images. LISS III (Linear Imaging Self Scanning Sensor) images have been used to classify images of individual years as well as stacked images. Individual year images have four bands and stacked images have eight bands because stacked images are created after merging of both the individual years’ images i.e., 2008 and 2009. Land use and land cover classification of images have been done for pervious and impervious categories for training signatures. After classification of images, producer's accuracy, user's accuracy, overall accuracy and kappa coefficients have been calculated with the help of contingency/error matrix for training signatures for pervious and impervious categories. It is observed that the accuracies of stacked images are higher than the individual year's images