图像轮廓提取方法研究

Contour extraction research of image

对现有的轮廓提取现状进行分析,把主要的轮廓提取方法划分为先验知识法、数学形态法、基于梯度的方法、水平集方法、活动轮廓模型方法、以及神经动力学方法六大类,并研究这些方法的主要特点。对轮廓提取方法研究进行展望,提出神经动力学方法是轮廓提取方法的发展方向。

The current research methods for contour extracting are discussed.The main methods are classified into six types of contour extracting,those are the methods of using priori knowledge,the methods of using mathematical morphology,the methods using the gradient of a image,the methods based on level set,the methods using the active contour model,and the methods using neutral dynamics,the characteristics of them are studied.The future for contour extracting field is forecasted,and the methods using neutral dynamics is considered as a very promising field.