Wetland Information Extraction from RS Image Based on Wavelet Packet and the Active Learning Support Vector Machine
Wetlands which are the planets most important ecosystem have high scientific research-value and will bring us both social and economic benefits. However, duing to various natural and man made factors, more and more wet-lands have converted to agricultural land and urban land. Now, the changes in wetlands area and quantity have caused publics widespread concern. And wet-lands management and protection will benefit from the improvement of the wetland information abstractions precision. Improving the classification preci-sion of the RS image is a difficult problem because of the small scale of remote sensing images. This paper which is about the wetland remote sensing images extraction is based on the LANDSAT ETM remote sensing data, and the result of the Wavelet Packet reconstruction will be used as the sample set of the Ac-tive Support Vector Machine .At the end of this paper, a comparative analysis of the experimental results will show between the single classification (SVM, BPNN) method and the solution which is proposed in this article. This method can be proved to obtain very good classification results through many experi-ments on remote sensing image classification Ive done. Experimental results show that this algorithms classification accuracy is better than the single classi-fications. Moreover, in the active learning process, the bad influence of the im-ages isolated and intersection points on the classification is avoided, and the number of training samples are reduced greatly.
Wavelet Packet ASVM wetland classification remote sensing information extraction
Pu Wang Wenxing Bao
School of Computer Science and Engineering, Beifang University for Ethnics,Yinchuan Ningxia, P.R. China 750021
国际会议
南昌
英文
491-499
2010-10-22（万方平台首次上网日期，不代表论文的发表时间）