基于二维模糊熵的图像非监督变化检测

Unsupervised image change detection based on 2D fuzzy entropy

针对图像变化检测,提出了一种基于二维模糊熵的方法。首先采用LMS方法寻找二维直方图的最佳分割方向;再根据最佳分割方向构建一种新的二维隶属度函数,并用其搜索差异图像二维模糊最大熵来确定最佳二维隶属度函数;最后按最佳二维隶属度函数检测差异图像中的变化区域。理论分析和实验结果表明,本文方法相对一般二维检测方法具有更好的检测性能。

Aiming at image change detection,a new change detection method based on 2D fuzzy entropies was proposed to detect change areas of the difference image.First,the best segmentation direction of a 2D histogram was found by using Least Square Method（LSM）.Then,a kind of new 2D membership function was defined based on the best segmentation direction,which was used to obtain the optimal membership function by searching 2D maximal fuzzy entropy.Finally,the image change area was detected by using the optimal membership function.Theoretical analysis and experiment results show that the proposed method has predominant change detection performance.