基于形态学重建的分水岭图像分割实验教学研究  

Research on experimental teaching of watershed image segmentation based on morphological reconstruction

作　　者：马学条[1] 周彦均 王永慧 郑雪峰 MA Xuetiao;ZHOU Yanjun;WANG Yonghui;ZHENG Xuefeng(School of Electronics and Information,Hangzhou Dianzi University,Hangzhou 310018,China)

机构地区：[1]杭州电子科技大学电子信息学院,浙江杭州310018

出　　处：《实验技术与管理》2021年第3期93-97,共5页Experimental Technology and Management

基　　金：浙江省高等教育十三五第一批教学改革研究项目(jg20180135);浙江省高等教育十三五教学改革研究项目(jg20190161)。

摘　　要：现有分水岭算法对噪声敏感,易出现过度分割现象,导致图像分割边缘不明显。在传统分水岭算法基础上,通过形态学开闭重建来清除图像中的噪声点,并采用最小覆盖运算修改梯度图像,使得局部最小区域仅出现在标记位置,从而消除过分割现象。实验结果表明,与传统标记分水岭算法相比,用改进的算法对硬币图像和火焰图像进行处理,硬币轮廓识别率提高了56.67%,火焰目标分割效果提高了16.15%,取得了较好的图像处理效果。The existing watershed algorithm is sensitive to noise and prone to over segmentation, resulting in that the image segmentation edge is not obvious. Based on the traditional watershed algorithm, the noise points in the image are removed by morphological open close reconstruction, and the gradient image is modified by the minimum cover operation, so that the local minimum area only appears in the marked position, thereby eliminating the over segmentation phenomenon. The experimental results show that, compared with the traditional marking watershed algorithm, the recognition rate of coin contour is improved by 56.67%, and the segmentation effect of flame target is improved by 16.15% by using the algorithm designed in this paper to process the coin image and flame image.

关 键 词：形态学重建 分水岭 图像分割 

分 类 号：G642.0[文化科学—高等教育学] TP751.1[文化科学—教育学]