Hyderabad City Land Use/Land Cover Changes Multi-spatio Temporal Comparison Using Spectral Angle Mapper with Maximum Likelihood Classification

Main Article Content

Rakesh Kumar Appala, Vidhya Lakshmi Sivakumar


Aim: In this study the objective is to predict land use and land cover changes by using Spectral Angle Mapper (SAM) and Maximum Likelihood Classifier (MLC) and also to identify the algorithm that gives more accuracy. Materials and Methods: Landsat7 ETM+ (Enhanced Thematic Mapper plus) and Landsat 8 were used for the years 2001, 2011 and 2021 of study region. These satellite images were classified using two classifiers, namely, SAM classifier and MLC forming two groups. Each group contains 3 samples with a total of N = 6 samples. The pretest power is to be determined with 80% and with alpha value of 0.05 and Confidence Interval of 95%. Results and Discussion: The land use and land cover changes have been analyzed with novel supervised classifiers of SAM and MLC and percentages of different types of region have been noted.An independent samples - t test from SPSS statistical analysis it was observed that from a single tail test p<0.05 hence there is a significance difference between two groups of classifiers, namely, SAM and MLC.The mean and standard deviation of overall classification accuracy is 91.42 ± 6.13 and 98.89 ± 1.26 respectively. The mean and standard deviation for kappa coefficient is 0.87 ± 0.86 and 0.98 ± 0.17 for SAM and MLC respectively. Conclusion: From this research it can be concluded that Maximum Likelihood Classifier performs better than Spectral Angle Mapper.

Article Details