基于MSE模型的高分辨率遥感图像变化检测

目前，传统高光谱分辨率遥感图像变化检测方法大多基于单一特征信息进行处理，没有综合影像所有特征信息，难以检测出完整的变化信息。为此，提出了一种基于多特征流形嵌入模型（multiview spectral embedding, MSE）的高分辨遥感图像变化检测方法，利用该模型对图像特征变化信息向量中所有的变化信息进行融合，获得完整的检测结果。实验结果表明，本文方法的检测精度好于传统的变化检测方法，并且稳定性良好。

Change Detection from High-Resolution Remote Sensing Image Based on MSE Model

At present, most of the traditional change detection methods from high-resolution remote sensing image are based on a feature information, the information of multi-feature information cannot be extracted, so it is difficult to detect the complete information. In order to solve this problem, a change detection algorithm of high-resolution remote sensing image based on multiview spectral embedding is proposed in the present paper. Firstly, change image is obtained using traditional difference change detection method, and multi-feature information is extracted. The feature vector information is fused by a MSE model and the complete change information can be obtained. The experimental results show that the detection accuracy of the proposed method is better than the accuracy of traditional methods, and its stability is outstanding.