支持向量机（SVM）是一种基于超平面分类的新的学习方法，具有很强的泛化能力。研究了支持向量机的学习机理，以及实现支持向量机的序贯最小优化算法（SMO），并用来对舰船图像进行识别。首先将待识别目标进行二维小波分解，获取不同尺度下的小波系数，然后对其进行主元分析，得到的主元分量作为支持向量机的特征量输入。实验结果表明，该方法具有良好的分类性能。-Support Vector Machine (SVM) is a hyperplane-based classification of new learning method, has strong generalization ability. Of support vector machine learning mechanism, and the realization of support vector machines Sequential Minimal Optimization (SMO), and used to identify the image on the ship. Targets to be identified first two-dimensional wavelet decomposition, for different scales of wavelet coefficients, and then carry out the main element of their analysis, the main element component as the feature of SVM input. Experimental results show that the method has good classification performance.