基于广义形态滤波和分水岭算法的图像分割

传统的分水岭算法直接对图像的梯度图进行分割，梯度算子由于受噪声或量化误差的影响而产生很多局部最小值，从而导致严重的过分割现象，如何消除其过分割现象一直以来都是研究的热点。本文介绍一种广义形态滤波器，使用两组适应不同形状要求的含有四个结构的结构元素来构成一个滤波器组，用于分水岭算法的预处理，在平滑原始图像的同时去除了易造成过分割的区域细节和噪声，有效地抑制了分水岭算法的过分割现象，然后使用分水岭算法，获得了较好的分割效果。

Image Segmentation Based on Generalized Morphological Filters and Watershed Algorithm

The traditional watershed algorithm carries on the segmentation to the gradient map of the image directly,because the gradient operator is influenced by noise or quantization error,the image produces a lot of local minimum values and results in the serious problem of over segmentation.How to resolve the problem of over segmentation is always a hotspot of research.This article introduces a kind of generalized morphological filters,which contains two groups of four-structure structuring elements in order to adapt the need of different shape,used to pretreat the watershed algorithm,smooth the original image,at the same time,wipe off the region detail and noise which are prone to cause over segmentation.It can restrain the over segmentation phenomenon of watershed algorithm effectively.Then we use the watershed algorithm,the method can achieve preferable segmentation result.