Change detection approach is capable of mapping the proportional land cover of each class for every pixel. Fuzzy image classification is useful, if not indispensable, for advanced change detection techniques considering processes and therefore the intensity of land cover change. Clear land cover changes from one class to the other. The Overall Classification Accuracy obtained is 77.34%, which can be further improved. As the successful change estimation in our work mainly depends on the classification of our data sets with high accuracies linearly. Also our study does not include knowledge base. If knowledge base is incorporated into the system, the accuracy of classification would improve. As a post classification procedure for change detection, in the place of Fuzzy classification, Maximum likelihood classifiers can also be used. Cross-Correlation Analysis (CCA) can also be used to identify changes that have occurred in a previously mapped area. The procedure uses a recent multi-spectral image with a thematic land cover map in a two-step process.