摘要
边缘检测是图像处理中的重要内容,是图像的最基本特征,在图像分割、图像检索、模式识别、机器视觉等领域中都有重要的应用。本文提出了一种基于多结构元素的形态学抗噪边缘检测算法,该算法利用形态学的基本运算膨胀、腐蚀、开、闭及它们的组合,并通过构造4个不同方向的结构元素,得到图像4个方向的边缘检测结果,并将这些结果加权平均,得到最终的图像边缘。结果表明,该算法的抗噪性能优于经典的Log算子和Canny算子,并且检测出的边缘平滑性好,特征清晰,因而有一定的实用性。
Edge detection,one of the most important features of image, plays a crucial role in image processing. It is defined as a region where there is a sudden change of pixels＇ gray level. Edge detection has many important applications in fields like image segmentation, image retrieval,pattern recognition,machine vision. This paper puts forward a kind of noise-immune morphology algorithm of edge-detection based on multi-structural elements. The algorithm adapts the basic morphological operations such as dilation, erosion, opening, closing and their combination, builds four structure elements to get four edge-detections,and averages these edges to gain the final detecting result. The experiment results show that the noise-immune capacity of this algorithm is evidently better than classical Laplacian operator and Canny operator, further more, it deblurs the detected edge efficiently and eliminated the influence of noise appropriately.
出处
《电子测量技术》
2008年第4期36-37,81,共3页
Electronic Measurement Technology
关键词
边缘检测
图像
数学形态学
结构元素
edge-detection
image
mathematical morphology
structure element
