The Comparative Study Of Remote Sensing Image Classification

Remote sensing technology has become the important means land use / land cover information sources. Classification, in its study, which occupies an important position, the merits directly related to the classification accuracy. This paper reviews the classification of remote sensing images on the background , explain the basic theory and remote sensing image preprocessing of the ways, a brief overview of the remote sensing image classification concepts and principles. detailed study of the traditional classification of remote sensing methods -- supervised and unsupervised classification, and the emergence of some relatively new classification in recent years—artificial neural networks and fuzzy classification. Compared with the various methods of principle, algorithms and their respective advantages and disadvantages. Finally, the use of different remote sensing classification of the actual image classification, the classification analysis, come to different methods used in the actual process characteristics. The results show that the artificial neural network classification methods in the classification effect is better than the traditional method of categorization.In short, remote sensing image classification in pattern recognition is a more complicated issue. remote sensing image classification of supervision and non-supervised classification methods, is the most basic and general method in image classification. Traditional supervised and unsupervised classification despite their different strengths, but there is some deficiency. And the new classification methods such as neural networks with adaptive, self-learning, associative memory storage and distribution of good character, by the people and to be widely used in image classification, break the traditional method of statistical classification of limitations, improve the speed and accuracy of classification. Although the various characteristics of each classification, but in practical work also needs to integrate multiple classifications, to improve the classification accuracy and precision.