粒子群算法及多阈值指数熵的遥感影像变化检测方法研究

Research on change detection approach using PSO algorithm and multiple thresholds exponential entropy in remote sensing images

在遥感影像检测中,一般采用多变化阈值来提高检测精度,但会导致运算量的增加。为解决该问题,提出利用粒子群算法及多阈值指数熵的遥感影像变化检测新方法。首先采用影像差值法构造差异影像;然后提出利用粒子群和多阈值指数熵的遥感影像分割方法,并将其用于对差异影像进行分割获取变化区域;最后对选取的实验数据进行变化检测,并与基于模糊C均值、双阈值指数熵、三阈值指数熵的非监督变化检测方法进行比较。实验结果显示,提出的变化检测方法其精度为94.77%,本案方法是一种有效地、可靠的遥感影像变化检测方法。

Although the multiple change thresholds can improve the accuracy of change detection,it will also lead to an increase in the amount of computation.In order to solve this problem,a new change detection approach using particle swarm optimization（PSO）algorithm and multiple thresholds exponential entropy in remote sensing images is proposed in this paper.Firstly,the difference image is generated by image differencing method.Then,the proposed remote sensing image segmentation method based on PSO and multiple thresholds exponential entropy is used to segment the difference image,and the change regions can be obtained.Finally,the proposed method is used to perform the change detection experiment on selected experimental data,compared with the unsupervised change detection methods,such as fuzzy C mean,double threshold exponential entropy and three threshold exponential entropy.The result shows that the accuracy of the proposed method is 94.77%,which is higher than the three methods mentioned above.The result of experiments confirms that the proposed method is an effective and reliable method for remote sensing images change detection.