基于空时域约束模糊核聚类的红外多时相地下目标探测

提出了一种具有空时域约束的模糊核聚类红外多时相浅层地下目标探测算法.首先在经典模糊聚类算法基础上引入空间和时相变化信息,给出时相信息指数；然后根据时相信息指数修正时相权重因子,将空时域约束加入到多时相模糊核聚类中对红外多时相图像进行分类；最后由分类结果给出地下目标的位置及目标大致种类数,并根据地下目标的红外成像机理获得目标的大体物理性质来提高地下目标探测的准确性.结果表明,所提算法对地下浅层目标具有较好探测准确度.

Detection of Buried Targets in Multi-temporal Infrared Imagery Using Fuzzy Kernel Cluster with Spatio and Temporal Restriction

Due to the existence of buried targets influenced the surface' images of IR changing with times,a fuzzy kernel cluster algorithm with spatio and temporal restriction (STKFCM) was proposed to detect buried targets based on multitemporal images of IR.Firstly,the information of spatio and the change in temporal were put into the classical fuzzy cluster algorithm.The index of temporal information was given.Then the weight factor of temporal was modificated by the index of temporal information.The spatio and temporal restriction was added to the multitemporal fuzzy cluster,which was used to classify the multitemporal images of IR.At last the targets' position and the number of classes were estimated by the results of class.Take the mechanism of buried targets' images into account,the general character of targets were obtained.And the detection' s exactness of buried targets was improved.Experiment results demonstrate the algorithm's effectiveness in detecting buried targets shallow.Some interested research was offered to detect buried targets using the technology of IR.