图像分割是一种重要的图像分析技术，对图像分割的研究一直是图像技术研究中的热点和焦点。分水岭算法是基于数学形态学理论的图像分割算法，但是对噪声敏感且存在过分割的现象。为提高图像分割效果，本文通过对影像进行滤波处理，用改进的快速区域合并算法优化分水岭算法进行影像分割。-In this work a technique for cloud detection and classification from MSG–SEVIRI (Meteosat Second Generation–Spinning Enhanced

Visible and Infra-red Imager) imagery is presented. It is based on the segmentation of the multispectral images using order-invariant

watershed algorithms, which are applied to the corresponding gradient images, computed by a multi-dimensional morphological operator.

To reduce the over-segmentation produced by the watershed method, a RAG (Region Adjacency Graph) based region merging

technique is applied, using region dissimilarity functions.