一种基于谱图理论的阵列目标识别方法

A New Algorithm for Array Target Recognition Based on spectral graph theory

阵列目标是遥感图像中一类重要目标,具有规则的空间分布关系。该文以空间分布关系特征作为阵列目标的识别特征,研究了阵列目标空间分布基元特征的提取及场景结构图邻接矩阵的建立技术,并在此基础上提出了一种基于迭代谱图划分的阵列目标识别方法。实验结果表明,该方法能有效识别多种阵列目标,具有良好的实用性。

This paper presents a new recognition method of array target by applying the characteristic of spatial relationship of target. The method is based on spectral graph theory. In the method, the spatial relationship primitive（SRP） of array target is extracted, and the adjacency matrix of the Scene Structure Graph（SSG） is established. Then, a fast recognition algorithm, namely Iterative Spectral Graph Partitioning Algorithm（ ISGPA）, is designed. Results of experiments indicate that the recognition effect on array targets is satisfactory.