Supervised Classification of Remote Sensing Images Using Fuzzy Technique

Abstract

In the convent onal remote sensing supervised classification, training information and classification results are represented in a one-pixel one class method. Class mixture cannot he taken into consideration in training classifier and in- determining pixels membership. The expressive limitation has reduced the classification accuracy level and led to the poor extraction of information. This paper describes a fuzzy supervised close fication method in which geographical information is represented as Fuzzy sets. The algoritem consists of two major stops: The estimate of fuzzy parameters. Som fary training data, und fuzzy partitions of spectral space Purtial membershin of pixels allows component coven classes of mixed pixels to be identified and more securite statistica accuracy by be achieved. Results of classifying a landast TV images are presented and their accuracy is analyzed