关键词: 分水岭算法, 彩色图像的梯度, 标记提取, Otsu算法, h-minima变换

Abstract: A new watershed algorithm based on improved gradient and self-adaptive marker-extraction is proposed to overcome over-segmentation. Two groups of structural elements are used to calculate gradient. The markers of reconstructed gradient are extracted by Otsu algorithm. h-minima transformation is used to modify markers, and the union set of a series of markers is regarded as the final marker image. The watershed transformation of modified gradient is performed. The experimental results show that the proposed algorithm can more efficiently reduce the over-segmentation, obtain meaningful and more reasonable regions than the traditional watershed algorithm.

Key words: watershed algorithm, color image gradients, marker extraction, Otsu algorithm, h-minima transformation