Classification of remote sensing image is an important aspect of image processing, which is the basis of the follow-up extraction of thematic information, detection of dynamic change, production of thematic maps, databases of remote sensing image, etc. With the development of modern remote sensing technology, remote sensing information has become richer. Therefore, the accuracy and real-time requirement of classification have become more important. However, the recent research which focuses on the new algorithm can not improve the classification efficiency significantly to meet the needs of practical applications.On the basis of existing classification algorithms, the paper analyzed the process of classification, and then found out the way to implement fast remote sensing images classification. The main idea of the method is: To improve the classification automation and accuracy, an AOI (Area of Interest) sample database is employed; parallel computing was used to advance the computing efficiency. Features of this paper are as follows:Firstly, as two main steps slowing down the speed of classification, samples selection and classification computing are found out. The accuracy of samples selection is limited by operators and tools. The efficiency of computing is decided by algorithms and hardware. Thus the main problems could be figured out.Secondly, two problems mentioned above are solved. Parallel computing was used in the step of classification computing to improve the computing speed. Sample database based on AOI was used in the step of selecting sample to improve the classification automation and accuracy.Finally, a fast, automatic and stable classification system based on the architecture of server/client was proposed by means of engineering, modular, process-oriented, hierarchical design model. This system can implement the fast classification of remote sensing images. And the experiments prove the efficiency of the parallel classification. |