摘要
提出了一种基于数学形态学算子的多结构元素多尺度边缘检测方法.首先构造6种具有代表性的结构元素,用改进的形态学边缘检测算子对灰度图像进行边缘检测,然后改变结构元素的尺寸大小得到多尺度下的边缘图像,根据不同尺度抗噪性能不同来确定权值,将不同尺度下的边缘图像合成,得到在噪声存在条件下较为理想的图像边缘.实验结果表明,与传统边缘检测算法相比,该算法在保持图像边缘清晰的同时,有很强的去除噪声能力,具有一定的实用性和可行性.
An image edge detection method of multi-structure elements and multi-scale based on mathematical morphology is proposed. Firstly, image edge is detected by using six representative structural elements and improved morphological edge detection operator. Then, the edges of various sizes are combined according to anti-noise performance of different seales,which are detected by changing the size of structure elements. Finally, the more ideal image edge under the environment of existing noise is obtained. Compared with traditional edge detection methods, the experimental result shows that the algorithm has strong ability of eliminating noise as well as keeps clear image edge. Therefore, it has very high usability and feasibility.
出处
《微电子学与计算机》
CSCD
北大核心
2009年第8期76-79,共4页
Microelectronics & Computer
关键词
边缘检测
多尺度
形态学
多结构元素
edge detection
multi-scale
morphology
multi-structure elements
