Abstract: A novel target identification method based on corner character and self-adaptive kernel clustering algorithm is proposed. Validity measure function is introduced to the kernel clustering algorithm, and it can get the sorts’ number automatically. The method is used for detecting the airplane. Experiments show that the algorithm is efficient.

Key words: Corner character, Kernel clustering, Validity measure function, Airplane identification

摘要： 提出了基于角点特征和自适应核聚类的目标识别方法，将有效性函数引入核聚类算法中，提出了一种可动态估计聚类数目的自适应核聚类算法。该方法用于飞机识别中，通过对飞机角点特征的自适应核聚类，完成定位识别。实验结果表明，该方法是有效的。

关键词: 角点特征, 核聚类, 有效性函数, 飞机识别