Affinity Propagation Clustering Of Applied Research In Remote Sensing Image Classification

Remote sensing(RS)image classification is always a pivotal part of remote sensing study.How to improve the accuracy of RS interpretation is a hot topics and important problems in RS application.it plays a key role in the RS image'application and promotion .Classification of remote sensing image using computer is data disposal of remote sensing image pixels by compute.There are mainly two methods in traditional computer classification:non-supervised classification and supervised classification.the former is a clustering process,it needs lesser man'attended operation while the latter is a study and training process,and it needs preliminary knowledge.While RS image classification methods emerged in endlessly with development of new methods and technique.for example:fuzzy classification,artificial neural network,expert system and so on.they all enhance accuracy,improve effect and application to a great extent.Whereas different ways are different advantage and disadvantage,classification effect is affected by some factors.threr is not a method which is optimal method to all RS image classifications.only can say that some methods fit some classifications well.consequence, who want a best classification effect,it needs to widely study land feature on the ground and classifier'optimal composite mode.Based on world RS image classification methods analysis,this paper mainly research supervised classification methods,non-supervised classification methods ,fuzzy classification and artificial neural network classification of remote sensing images were classified. And then introduces the affinity propagation clustering in remote sensing image classification in the Applied Research. Through the use of Matlab software, Intelligent classification of remote sensing images were made by the pilot study to deal with. the results show that the effect of AP algorithm is not less than the traditional algorithm, more closely clustering structure can give accurate results, and analysis the advantages and disadvantages of affinity propagation clustering. In conclusion, the paper sums up lack of research work and provides a reference for further study in the future.