面向对象的迭代加权多变量变化检测方法

发布时间：2019-07-03 17:46

【摘要】：针对传统变化检测方法应用于高分辨率遥感影像变化检测时出现的变化信息分散、椒盐噪声影响严重等问题,将面向对象技术和迭代加权多变量变化检测方法结合起来,提出了一种面向对象的迭代加权多变量变化检测方法(iteratively reweighted multivariate alternative detection,IR-MAD)。该方法主要通过结合卡方分布的概率密度函数和面向对象技术来对传统多变量变化识别方法(multivariate alternative detection,MAD)进行改进,卡方分布的概率密度函数对变化信息进行融合以获取信息集中的IR-MAD变量。此外,在对影像进行分割时结合叠置分割技术获取边界一致、同质性较好的影像对象。实验表明,面向对象IR-MAD方法能够有效集成变化信息,准确提取变化区域,同时较好地保持变化目标的结构与形状,减少椒盐噪声的影响,检测结果具有较高的可靠性。
[Abstract]:Aiming at the problems that the traditional change detection method is applied to the problem that the change information of the high-resolution remote sensing image change detection is dispersed and the pepper salt noise is serious and the like, the object-oriented technology and the iterative weighted multi-variable change detection method are combined, An object-oriented iterative weighted multi-variable detection method (IR-MAD) is proposed. The method is mainly used to improve the traditional multi-variable detection (MAD) by combining the probability density function of the chi-square distribution and the object-oriented technique, and the function of probability density function of the chi-square distribution is used for fusing the change information to obtain the IR-MAD variable in the concentrated information. In addition, when the image is divided, the image object with uniform boundary and good homogeneity is obtained by combining the overlay segmentation technique. The experiment shows that the object-oriented IR-MAD method can effectively integrate the change information and accurately extract the change area, while keeping the structure and shape of the changing target better, and reducing the influence of the pepper salt noise, and the detection result has higher reliability.
【作者单位】：中国测绘科学研究院;兰州交通大学甘肃省地理国情监测工程实验室;中国地质大学(武汉);重庆市地理信息中心;
【基金】：国土资源部公益性行业科研专项(201311038)
【分类号】：TP751

【相似文献】