A comparative analysis of remote sensing image classification techniques

In this paper, we have compared the accuracy of four supervised classification as Mahalanobis, Maximum Likelihood Classification (MLC), Minimum distance and Parallelepiped classification with remote sensing Landsat images of different time period and sensors. We have used Landsat Multispectral Scanner (MSS), Thematic Mapper (TM) and Enhanced Thematic Mapper+ (ETM+) images of 1972, 1998 and 2013 respectively of Jaipur district, Rajasthan, India. Accuracy has been calculated using Producer accuracy, User accuracy, Overall accuracy and Kappa statistics. We have found that remote sensing images with a different time period and sensors, when classified with supervised algorithms produced different results. Minimum distance classification produced better accuracy with Landsat MSS image than other three classifications while Maximum Likelihood Classification produced better accuracy with Landsat TM and ETM+ images than other three classifications.