测绘学报 ›› 2016, Vol. 45 ›› Issue (1): 65-72.doi: 10.11947/j.AGCS.2016.20150240

• 摄影测量学与遥感 • 上一篇    下一篇

一种四叉树和测地线活动轮廓模型相结合的海陆影像分割方法

郭海涛1,2, 孙磊1, 申家双2, 陈小卫1, 张宏伟1   

  1. 1. 信息工程大学地理空间信息学院, 河南 郑州 450052;
    2. 海军海洋测绘研究所, 天津 300061
  • 收稿日期:2015-05-14 修回日期:2015-09-17 出版日期:2016-01-20 发布日期:2016-01-28
  • 作者简介:郭海涛(1976—),男,博士,副教授,研究方向为数字摄影测量、海洋遥感。
  • 基金资助:
    国家自然科学基金(41101396;41001262)

An Island and Coastal Image Segmentation Method Based on Quadtree and GAC Model

GUO Haitao1,2, SUN Lei1, SHEN Jiashuang2, CHEN Xiaowei1, ZHANG Hongwei1   

  1. 1. Institute of Suveying and Mapping, Information Engineering University, Zhengzhou 450052, China;
    2. Navey Institute of Hydrographic Surveying and Charting, Tianjin 300061, China
  • Received:2015-05-14 Revised:2015-09-17 Online:2016-01-20 Published:2016-01-28
  • Supported by:
    The National Natural Science Foundation of China(Nos.41101396;41001262)

摘要: 海陆影像分割对于后续的海岸线提取、潮间带地形反演、海岸演化状况分析等都具有十分重要的意义。本文在分析了四叉树、测地线活动轮廓(GAC)模型和Canny边缘检测算子等在海陆影像分割中优缺点的基础上,提出了一种四叉树、Canny算子和GAC模型相结合的海陆影像分割方法。该方法综合利用上述各方法的优点,将Canny算子边缘检测结果融入到基于四叉树初分割的GAC模型中,重构边界停止函数,演化水平集方程,实现海陆影像分割。试验结果表明,该方法具有海陆影像分割速度快、精度高、可靠性强和自动化程度高等优点,对于弱边缘以及严重凹陷边缘,都能实现自动和准确分割。

关键词: 四叉树, 地测线活动轮廓模型, Canny边缘检测, 海陆影像分割, 水边线提取

Abstract: Island and coastal image segmentation is of great importance for the subsequent coastline extraction, terrain inversion for intertidal zone, analysis of the situation for shore evolution, and so on. Firstly, the advantages and disadvantages of quadtree, geodesic active contour (GAC) model and Canny edge detector used in the island and coastal image segmentation are analyzed. Secondly, an island and coastal image segmentation method is proposed by integrating quadtree, GAC model and Canny edge detector. The advantages of these three kinds of method are taken in the method proposed in this paper. The method introduces the results of Canny edge detector into edge indicator function of geodesic active contour model based on quadtree segmentation, evolutes the level set equation, and realizes island and coastal image segmentation.The experimental results show that the method proposed in this paper is of high speed, precision, reliability and automation for island and coastal image segmentation, even in the weak edges and serious concave edges.

Key words: quadtree, GAC model, Canny edge detector, island and coastal image segmentation, waterline extraction

中图分类号: