基于标记的改进分水岭分割算法

针对传统分水岭分割算法中存在的过分割问题,提出了一种基于标记的改进分水岭分割算法。该方法首先对原图像进行形态学开闭预重建,去除图像中的暗纹理和噪声,在计算形态梯度之后采用开闭后重建,然后,对重建后的梯度图像利用最大熵阈值方法进行标记处理,依据标记对原始梯度图像进行修正,最后使用分水岭算法在修正后的梯度图像上进行分割。从实验结果来看,该方法能较好地抑制传统分水岭算法的过分割现象。

Improved Watershed Segmentation Algorithm Based on Marker

Aimed at the problems of over-segmentation existing in traditional watershed algorithm,a marker-based improved watershed segmentation method is proposed.Firstly,morphological opening and closing pre-reconstruction of the original image is utilized to remove the dark texture and noise.Morphological double opening and closing reconstruction is performed after the morphological gradient's calculation.Then,by using the maximum entropy threshold method reconstruction gradient image is dealed with the marks.According to the marks the original gradient image is amended.Finally,the watershed transformation of the modified gradient image is performed to achieve the segmentation results.Experimental results show that the proposed algorithm can be effective to restrain the over-segmentation phenomenon of traditional watershed method.