Object-oriented Classification Of Remote Sensing Image Based On SPM Feature Extraction

The development of the remote sensing technology makes us obtain very abundant information of nature, especially with the appearance of high resolution remote sensing image it extends the visual field of the nature. But the challenge that faces us is how to make use of the data effectively and obtain more useful information through some processing. High resolution remote sensing data have a lot of characteristics such as spectral, shape, texture and context and so on compared to the other remote sensing data. Though the technology of the remote sensing image classification has made considerable progress, it will result in not only reducing the classification accuracy but also making the spatial data redundant and wasting the resource when the single traditional classification method is applied to the high resolution remote sensing image.Therefore, this paper closely around the high-resolution remote sensing image classification to improve the accuracy. Take a piece of land in southern California of the United States for example, we focus on the factors that affect the classification accuracy and improving the accuracy of the method of these two aspects. Innovating the previous object-oriented remote sensing image classification method, before segment the remote sensing image we pretreatment it first, change the image gray level; improve the image contrast, elimination of edges and noise, smooth image; prominent edges or linear features, sharpening an image; synthetic color image; compressed image data, highlighting important information. This laid the foundation for remote sensing image segmentation. Then use mean shift image segmentation algorithm for remote sensing image segmentation, get the target image, the image features and some of the characteristics of the parameters to convert the original image into a more abstract, more compact form. making more high-level image analysis and understanding possible.This paper introduces the basic principles of support vector machine(svm). Specifically explain the linear separable, nonlinear separable and some method to hand the nonlinear separable, finally, detailed analysis of the SVM algorithm. Focuses on the basic principles of spatial pyramid matching kernel, and introduces the spatial pyramid matching kernel method (SPM) for feature extraction. In the object-oriented remote sensing image classification, this paper successful introduction the spatial pyramid matching kernel method (SPM) for feature extraction, the algorithm can extract the features accurately. Then using support vector machines for remote sensing image classification. Comparing the results, we can see that the object-oriented classification of remote sensing image which based on SPM feature extraction can greatly improve the accuracy.