Review of soft classification approaches on satellite Image and accuracy assessment

Image classification is a process that may be affected by many factors. This paper is in regard of the classification techniques used for image processing and analysis using the concept of Fuzzy and Possibilities techniques, applied in each pixel of false-color satellite image By this paper examine classification approaches and the technique used for improving classification accuracy. By the introduction of fuzzy methodology provides the basis for the development of more robust approaches to the remote sensing classification problem. Effective use of multiple features of remotely sensed data and the selection of a suitable classification method are especially significant for improving classification accuracy. Reference data can be generated from this system in two modes i.e. pure reference data as well as mixed reference data. There is also a provision in this system for saving the membership values generated using different classifiers for sub-pixel land cover mapping.Integration of remote sensing, geographical information systems (GIS), and expert system emerges as a new research frontier. Although, a number of Fuzzy set theory based classifiers may be adopted, Fuzzy based classifier their approaches used to partition pixel for remote sensing image into different class membership value. More research however is needed to identify and reduce uncertainties in the image processing chain to improve classification accuracy. From amongst a number of soft classifiers, this study has focused on two statistical classifiers maximum likelihood classifier (MLC) and linear mixture model (LMM), two fuzzy set theory based classifiers fuzzy, c-means (FCM) and possibilistic cmeans (PCM), the Other soft classification method can be used as an Artificial Neural Network algorithm both as supervised as well as unsupervised. Other emerging classifiers are nonparametric learning algorithm known as Support Vector Machine (SVM). SVM has been used for estimating the parameters of a Bayes classifier for remote sensing multispectral data.The paper reviews soft classification techniques for Remotely Sensed Data. Keywords-Fuzzy C-Means, Fuzzy Possibilistic C-Means, kmean, Maximum Likelihood Classifier (MLC) ,support vector machine(SVM),Artificial neural network algorithm(ANN), membership value and entropy.