图像线条关系研究

线条的提取和分析是图像处理和计算机视觉研究的基础性任务,也一直是计算机视觉研究的一个难点和瓶颈。本文将图像中的线条关系归纳为3种：伴随、接连和包围关系。现有文献对接连关系有很多讨论和应用,但是对伴随和包围关系却缺少研究。本文对图像线条的伴随和包围关系进行讨论,给出刻画和计算方法,并通过人工数据实验和实际图像数据实验,验证算法的有效性。

Research on Relationship between Image Lines

Line extraction and analysis is a basic task, as well as a bottleneck, of image processing and computer vision research. In this paper, we summarize the relationship between image lines into three categories: concomitance, succession, and surrounding. The succession relation has been extensively researched and applied by the existing theories, but the concomitance and surrounding relations were seldom discussed. This paper introduces methods and algorithms to describe and compute the concomitance and surrounding relations. The effectiveness of the algorithms was verified by experiments on synthetic data and real image data.