一种面向地理对象的遥感影像变化检测方法

目的 根据高空间分辨率影像上变化区域呈聚集状分布的特点,提出了一种面向地理对象的遥感影像变化检测算法。在利用 Mean-Shift分割算法的基础上,获得不同时相地理对 象 的 灰 度特征 信息,结合 变 化 矢量 分析,采用最大数学期望算法自动提取变化区域。以 QuickBird、SPOT、TM 三组不同空间分辨率的影像进行算法验证并比较了该方法与单像素变化检测算法的差异。结果表明,三组影像中面向对象的变化检测算法的检测精度分 别 为 91.1%,87.3% 和 84.3%,单像素 的 变 化 检 测算法 检测精度分别为 86.41%,82.48% 和81.02%。试验结果显示面向对象的算法检测精度高于基于单像素的变化检测算法,且对高空间分辨率的影像检测效果要优于对中低空间分辨率的影像的检测效果。该算法减少了变化阈值确定中的人工干预,克服了以像素为单位的变化检测算法中由于缺少空间邻域信息而产生孤立、离散、不连通变化结果的问题,能够满足在不同土地覆盖类型下的变化检测要求,在国土资源监测中具有一定的使用价值。

The Geographic Object-based Method for Change Detection withRemote Sensing Imagery

Objective This paper proposes a geographical object-based method for change detection with high reso-lution images based on the changing areas distributed as a clustered type.This algorithm utilizes theMean-Shift segmentation algorithm to extract a geographic object,and then uses the gray informationof the geographic object with the EM algorithm to automatically extract changed and unchanged areas.This method considers spatial neighborhood information which can avoid the isolation and discrete dis-connected areas in change results when using apixel-based method.This method also reduces inter-vention when determining the change threshold value.Groups of three different spatial resolution ima-ges(QuickBird,SPOT,TM images)are used to verify this proposed geographic object-based changedetection algorithm and compared the accuracy and precision with a pixel-base method.Our resultsshow that the accuracy with object-based change detection method on QuickBird,SPOT and TM ima-ges was 91.1%,87.3% and 84.3%,while for the pixel-based method are 86.41%,82.48% and81.02%respectively.These results illustrate that the object-based change detection method has high-er change detection accuracy than the pixel based approach.Moreover,the object-based method hasbetter accuracy for high spatial resolution than in middle or low resolution images.