摘要
针对分水岭图像分割算法对噪声比较敏感和过度分割现象,提出一种自适应阈值与数学形态学相结合的改进分水岭分割算法。首先进行中值滤波、计算梯度,然后进行数学形态学处理、自适应阈值分割,最后对前述处理的图像运用分水岭算法实现精确分割。实验结果表明,该方法具有很强的噪声抑制能力,较好地削弱了过分割现象,实现精确分割。
This paper proposes an improved watershed segmentation algorithm based on the combination of adaptive threshold and mathematical morphology for watershed image segmentation algorithm which is sensitive to noise and over segmentation.First,the median filtering and gradient computing is made,and then mathematical morphology processing and adaptive threshold segmentation are performed.Finally,a watershed algorithm is used to accurately segment the previously processed image.Experimental results show that this method has a strong ability to suppress noise,which can effectively weaken the over segmentation and achieve accurate segmentation.
作者
罗山
张冬梅
Luo Shan;Zhang Dongmei(School of Traffic and Automotive Engineering,Panzhihua University,Panzhihua Sichuan 617000,China)
出处
《山西电子技术》
2018年第6期3-5,共3页
Shanxi Electronic Technology
基金
攀枝花学院大学生创新创业训练计划项目(2017cxcy081)
关键词
中值滤波
分水岭
图像分割
数学形态学
自适应阈值
median filtering
watershed
image segmentation
mathematical morphology
adaptive threshold
