应用SVM算法进行TM多光谱图像地物分类

TM multi-spectral image's land cover classification using SVM algorithm

SVM算法是近年来统计模式识别领域流行的算法。因为有统计学习理论(SLT)作为基础,SVM算法具有良好的计算有效性、健壮性和统计稳定性,被广泛地应用在图像识别、语音识别、文字识别等多种模式识别领域。TM多光谱遥感图像的地物分类一直是遥感图像处理领域比较困难的问题。现尝试应用SVM算法对北京市怀柔水库附近地区的地物进行分类,收到了较好的效果。

Support vector machine （SVM） algorithm is a kind of statistical pattern recognition algorithm which is popular at research field in recent years. For the reason of its academic foundation of statistical leaming theory （SLT）, SVM algorithm has features of better computational efficiency, robtistness and statistical stabihty, it＇s widely used in many fields such as digital image recognition, voice detection and character recognition which belongs to the application of pattern classification. The classification for land cover of TM multi - spectral image is a difficult problem for a long time. This paper tries to apply SVM algorithm to classify the land cover of area nearby Huai - rou reservoir of Beijing. The result of experiment shows it works well.