中文摘要：

鉴于SAR图像获取的困难，无法保证机器学习算法时需要的大数据量训练样本，因此影响了识别结果。首次提出了应用虚拟样本来扩大SAR图像目标识别训练集，提高SAR图像目标识别率的方法。通过使用重采样算法，奇异值重构与轮廓波重构等方法构建虚拟样本，与原有样本组成训练集，并通过SVM支持向量机进行训练识别在MSTAR公共数据集上的识别实验结果表明，对于不同数量的实际训练样本，通过添加本文方法构建的虚拟样本扩大训练集后，SAR图像目标识别率均会得到提高，尤其在小样本的情况下，识别率提高非常显著。该文证实了虚拟样本应用于SAR图像目标识别的有效性。在样本数目有限时，添加虚拟样本对SAR图像目标识别性能具有明显的改善作用。

英文摘要：

Machine learning algorithm is one of the main methods of recognizing ground target SARimage; however, because of the difficulties of SAR image acquisition, it is impossible to guarantee a suffi-cient number of training samples during the machine learning algorithm. Thus, the recognition results willbe affected. For the first time, this paper proposed the way to improve the recognition rate of SAR imagesby using virtual samples to expand the training set of SAR image target recognition. Through such meth-ods as re--sampling algorithm, singular value reconstruction and contourlet reconstruction to generate vir-tual samples, this paper combined the training set with the original samples. An experiment was conductedon recognizing MSTAR data sets by means of training vector machine with SVM support. The resultsshow that for different numbers of training samples, the recognition rate of SAR images can be improvedby adding the virtual samples, especially in the case of small samples. This paper proves the effectivenessof using virtual sample in SAR image target recognition. When the number of samples is limited, addingvirtual samples can significantly improve the performance of SAR image target recognition.