面向对象的多尺度分割算法是以模糊数学为基础，通过对影像中对象的多个像元

Object-oriented multiscale segmentation algorithm is based on fuzzy mathematics, statistical pixel gray scale, size, and texture information on a plurality of objects in the image [34], the size of the objects belonging to a category obtained probability, in the same between the image element to ensure maximum retention characteristic similarity premise, the video segmentation process at different scales.

面向对象的多尺度分割算法是以模糊数学为基础，通过对影像中对象的多个像元的灰度、大小和纹理等信息进行统计[34]，得出对象属于某一类别概率的大小，在保证同类像元之间保持最大的特征相似性前提下，对影像进行不同尺度下的分割处理。

The object-oriented multi-scale segmentation algorithm is based on the fuzzy mathematics, through the statistics of the gray level, size and texture information of multiple pixels in the image [34], we can get the probability of the object belonging to a certain category. Under the premise of ensuring the maximum characteristic similarity between similar pixels, the image is segmented at different scales.<br>