A Review of Soft Classification Approaches on Satellite Image and Accuracy Assessment

Classification is a widely used technique for image processing and is used to extract thematic data for preparing maps in remote sensing applications. A number of factors affect the classification process. But classification is only half part of image processing and incomplete without accuracy assessment. Accuracy assessment of classification tells how accurately the classification process has been carried out. This research paper presents a review study of image classification through soft classifiers and also presents accuracy assessment of soft classifiers using entropy. Soft classifiers help in the development of more robust methods for remote sensing applications as compared to the hard classifiers. In this paper, two supervised soft classifiers, FCM, and PCM have been used to demonstrate the improvement in the classification accuracy by membership vector, RMSE, and also it has tried to generate fraction output from FCM, PCM, and noise with entropy.