基于自适应形态学的医学图像边缘检测

摘要

针对人体医学图像组织结构复杂、模糊、噪声大的特点，提出一种新的多尺度结构元的自适应边缘检测算法，给出新算法中计算各结构元权值的方法，并将其与传统的算法进行比较。实验结果表明，新方法能克服传统边缘检测算法抗干扰能力小的缺点，较好地实现了噪声图像的弱边缘检测。具有检测灵活性强，获得边缘信息平滑、丰富的特点，而且，算法编程实现容易。%As to the characteristics of complex structure,vague and big noise in medical images of human body,in this paper we propose a new adaptive edge detection algorithm with multi-scale structural elements,present the method of calculating each structural element weight in the new algorithm,and compare it with the traditional algorithms.Experimental results show that the new algorithm can overcome the defect of weak anti-interference ability in traditional edge detection algorithm,and better realises the weak edge detection of noisy image.It has the characteristics of high detection flexibility,smooth and rich edge information acquired,and easy to realise the algorithm programming.