一种遥感影像边界无效像元的检测方法及应用  

Implementation of a Method for Detecting the Invalid Boundary Pixels of Remote Sensing Image

作　　者：高娟[1] 王荣彬[2,3] 马晶[1] 曾超[2] 沈焕锋[2] 

机构地区：[1]武汉大学资源与环境科学学院,湖北武汉430079 [2]武汉大学 测绘遥感信息工程国家重点实验室,湖北武汉430079 [3]中国土地勘测规划院,北京100035

出　　处：《地理空间信息》2012年第2期12-15,179,共4页Geospatial Information

基　　金：国家973计划资助项目(2011CB707103);国家自然科学基金资助项目(41071269);中国土地勘测规划院资助项目(20101709369)

摘　　要：遥感数据在进行几何校正或投影变换后常常会在边界形成一些无效像元,这些无效像元会对影像的辐射信息统计、质量评价产生影响。提出了一种基于区域生长的影像边界无效像元检测方法,以4个顶点为种子点,考察种子点周围的所有边界来进行区域增长,通过这些点的区域增长将与无效像元具有相同灰度值属性的相邻像素合并到此区域。以信息熵为例,阐述了该方法在影像统计与质量评价中的应用意义。During geometric correction or after the projection transformation,remote sensing data often form invalid pixels in the boundary.These invalid pixels would affect the radiation information statistics and quality evaluation of the image.A method based on region growing was proposed to detect remote sensing image which had invalid boundary pixels.With four vertices as seed points,examining all the boundaries around the seed point and merging the adjacent pixels which had the same gray value with the invalid pixels into this area through regional growth of these points.Using information entropy as an example,the significance of the application of this method in image statistics and quality evaluation is described.

关 键 词：边界无效像元 区域生长法 阈值法 信息熵 

分 类 号：P237[天文地球—摄影测量与遥感]