论文部分内容阅读
常规多时相遥感影像变化检测主要基于光谱信息,没有充分利用纹理、几何、形状等多种特征信息,不足以体现检测目标的完整性和准确性。本文针对不同特征在变化检测中应用的优势,在提取影像多种特征的基础上,构建了1维和多维两种基于信息融合策略的变化检测方法,即利用1维特征空间加权距离相似度运算、多维特征空间的模糊集融合和支持向量机融合策略进行变化检测。利用多时相QuickBird高分辨率遥感影像进行城市土地覆盖变化检测试验,结果表明,本文方法可以有效集成不同特征的优势与表征变化信息的能力,提高变化检测过程的稳定性和适用性,同时能够更好地保持变化地物的结构和形状,突出主要变化目标。
Conventional multi-temporal remote sensing image change detection based on spectral information does not make full use of texture, geometry, shape and other features of the information, not enough to reflect the integrity and accuracy of the detection target. In this paper, based on the advantages of different features used in the change detection, two kinds of change detection methods based on information fusion strategy are constructed based on the extraction of multiple features of the image. That is, by using 1-D feature space weighted distance similarity calculation, Multi-dimensional feature space fuzzy set fusion and support vector machine fusion strategy for change detection. The test of urban land cover change using the multi-temporal QuickBird high resolution remote sensing imagery shows that the proposed method can effectively integrate the advantages of different features and the ability of characterizing the change information to improve the stability and applicability of the change detection process, Goodly maintain the structure and shape of the changed features and highlight the major changes.