计算机应用 ›› 2019, Vol. 39 ›› Issue (9): 2695-2700.DOI: 10.11772/j.issn.1001-9081.2019030543

• 虚拟现实与多媒体计算 • 上一篇    下一篇

耦合先验拉普拉斯坐标的半监督图像分割算法

曹昀炀1, 王涛2   

  1. 1. 华东师范大学 统计学院, 上海 200241;
    2. 南京理工大学 计算机科学与工程学院, 南京 210094
  • 收稿日期:2019-04-03 修回日期:2019-05-14 出版日期:2019-09-10 发布日期:2019-05-24
  • 通讯作者: 王涛
  • 作者简介:曹昀炀(1998-),男,江苏南京人,主要研究方向:应用统计、图像分割;王涛(1990-),男,江苏扬州人,副教授,博士,主要研究方向:计算机视觉、图像分割。
  • 基金资助:

    国家自然科学基金资助项目(61802188);江苏省自然科学基金资助项目(BK20180458);江苏省博士后科研基金计划项目。

Semi-supervised image segmentation based on prior Laplacian coordinates

CAO Yunyang<sup>1</sup>, WANG Tao<sup>2</sup>   

  1. 1. School of Statistics, East China Normal University, Shanghai 200241, China;
    2. College of Computer Science and Engineering, Nanjing University of Science and Technology, Nanjing Jiangsu 210094, China
  • Received:2019-04-03 Revised:2019-05-14 Online:2019-09-10 Published:2019-05-24
  • Supported by:

    This work is partially supported by the National Natural Science Foundation of China (61802188), the Natural Science Foundation of Jiangsu Province (BK20180458), the Jiangsu Planned Project for Postdoctoral Research Fund.

摘要:

针对传统半监督图像分割方法难以精确分割分散或细小区域的缺陷,提出了一种耦合标签先验和拉普拉斯坐标模型的半监督图像分割算法。首先,扩展拉普拉斯坐标(LC)模型,通过引入标签先验项进一步精确表征未标记像素点与已标记像素点之间的关系。然后,基于矩阵方程的求导优化,有效估计像素属于标签的后验概率,以实现图像目标分割的任务。得益于标签先验的引入,所提算法对分散或细小区域的分割更加鲁棒。最后,在多个公开的半监督分割数据集上实验结果表明,相比拉普拉斯坐标算法,所提算法的分割准确率获得了显著提升,验证了所提算法的有效性。

关键词: 图像分割, 彩色图像, 半监督图像分割, 拉普拉斯坐标, 先验概率

Abstract:

Focusing on the issue that classic semi-supervised image segmentation methods have difficulty in accurately segmenting scattered or small regions, a semi-supervised segmentation algorithm based on label prior and Laplacian Coordinates (LC) was proposed. Firstly, the Laplacian coordinates model was extended, and further the relationship between unlabeled pixels and labeled pixels accurately characterized by introducing the label prior. Secondly, based on the derivation of matrix equation, the posterior probability that the pixel belongs to the label was able to be effectively estimated, thus achieving the segmentation of the image. Thanks to the introduction of the label prior, the algorithm was more robust to the segmentation of scattered and small regions. Lastly, the experimental results on several public semi-supervised segmentation datasets show that the segmentation accuracy of the proposed algorithm is significantly improved compared with that of the Laplacian coordinates algorithm, which verifies the effectiveness of the proposed algorithm.

Key words: image segmentation, color image, semi-supervised image segmentation, Laplacian Coordinates (LC), prior probability

中图分类号: