基于形态学算法的灰度图像边缘检测.doc

基于形态学算法的灰度图像边缘检测 院 系 学院 专 业 班 级 学 号 姓 名 指导教师 负责教师 沈阳航空工业学院2009年6月摘 要 数学形态学建立在集合论的基础上，是一种应用于图像处理和模式识别 领域的新的方法，它的基本思想是用一定形态的结构元素去度量和提取图像 中的对应形状以达到对图像分析和识别的目的。边缘检测在图像处理与计算 机视觉中占有特殊的位置，它是底层视觉处理中最重要的环节之一。基于数 学形态学的边缘信息提取处理优于基于微分运算的边缘提取算法，它不像微 分算法那样对噪声敏感，同时，提取的边缘也比较光滑，在边缘检测上既能 够体现图像集合特征，很好地检测图像边缘，又能满足实时性要求，并且可 以在边缘检测的基础上，通过改变形态尺度更好地克服噪声影响。 本文首先深入地讨论了数学形态学的基本理论，从二值形态学推广到灰 度形态学，并分析和介绍了数学形态学在图像处理中的应用。然后详细地论 述了基本的形态学滤波器、广义形态滤波器以及多结构元素滤波器，并对各 种滤波器的特性加以分析和比较。最后，本文基于传统的形态学边缘检测算子， 采用了抗噪型多结构元形态学边缘检测算法，完成人体运动图像的边缘 检测。 关键词：Abstract Mathematical morphology founded on set theory is a new method applied in the field of image processing and pattern recognition. Its prime principle is to use a certain structuring element to measure and extract the corresponding form in an image so that we can analyze and recognize the image. Edge detection plays a special role in an image processing and computer vision. It is one of the most important links in the basic vision processing. The edge detection processing based on mathematical morphology excels that based on differential operation. It is not so sensitive to noise as differential operation and the extracted edge is smooth. It can not only embody the set characteristics of an image and efficiently detect the edge , but also meet the real-time expectations. At the same time，it can reduce the noise by changing morphological scale. Above all in this paper the basic theory of mathematical morphology is discussed，its application in image processing is analysed and introduced as well. Then the basic morphological filter, the generalized morphological filter and the morphological filter with omnidirectional structuring element are expounded in detail. Considering the advantage of mathematical morphology exploited in image edge detection, a multi-scale morphological edge detection algorithm is adopted in this paper. At the end a new antinoise multi-structuring elements edge detector is used by improving the edge detectot ment