遥感影像分割是指把一幅影像划分为互不重叠的一组区域的过程,它要求得到的每个区域的内部具有某种一致性或相似性,而任意两个相邻的区域则不具有此种相似性。遥感影像分割是面向对象的遥感影像数据挖掘与应用中的一项关键技术,对于影像目标信息自动化提取与智能识别尤为重要,在面向对象的遥感影像处理工程中具有重要意义。本文对常见的高空间分辨率遥感影像分割方法与应用策略进行了分析,比较了各种分割方法的应用范围、优缺点及目前存在的改进措施。建立了面向对象的遥感影像分割方法的分类体系,最后指出了面向对象的遥感影像分割方法目前所存在的问题及应用前景。
Remote sensing imagery segmentation is a process of dividing an image into different regions such that each region is,but the union of any two adjacent regions is not,homogeneous.It is one of the key techniques in the object-oriented remote sensing imagery data mining and its application,also quite essential in remote sensing image processing engineering.In this paper,we have a thorough survey on different methods of high spatial resolution remote sensing imagery segmentation,categorizing them into four groups according to the gray or color information they are exploiting.The disadvantage of current methods and the proper progress which can be obtained in the near future are pointed out at the end of this essay.