华南理工大学学报(自然科学版) ›› 2016, Vol. 44 ›› Issue (5): 48-57.doi: 10.3969/j.issn.1000-565X.2016.05.008

• 电子、通信与自动控制 • 上一篇    下一篇

基于二维 Arimoto 灰度熵的图像阈值分割快速迭代算法

吴一全1,2,3,4 朱丽1 吴诗婳1   

  1. 1. 南京航空航天大学 电子信息工程学院,江苏 南京 211106; 2. 江苏省制浆造纸科学与技术重点实验室,江苏 南京 210037; 3. 中国水产科学研究院淡水渔业研究中心 农业部淡水渔业与种质资源利用重点实验室,江苏 无锡 214081; 4. 农业部东海海水健康养殖重点实验室,福建 厦门 361021
  • 收稿日期:2015-06-11 修回日期:2016-01-07 出版日期:2016-05-25 发布日期:2016-04-12
  • 通信作者: 吴一全(1963-),男,博士,教授,博士生导师,主要从事图像处理与分析、目标检测与识别、智能信息处理研究. E-mail:nuaaimage@163.com
  • 作者简介:吴一全(1963-),男,博士,教授,博士生导师,主要从事图像处理与分析、目标检测与识别、智能信息处理研究.
  • 基金资助:
    国家自然科学基金资助项目(61573183);江苏省制浆造纸科学与技术重点实验室开放基金资助项目(201313);农业部淡水渔业与种质资源利用重点实验室开放基金资助项目(KF201313);农业部东海海水健康养殖重点实验室基金资助项目(2013ESHML06);江苏高校优势学科建设工程项目(2012)

Fast Iterative Algorithm for Image Threshold Segmentation Based on Two-Dimensional Arimoto Gray Entropy

WU Yi-quan1,2,3,4 ZHU Li1 WU Shi-hua1   

  1. 1.College of Electronic and Information Engineering,Nanjing University of Aeronautics and Astronautics,Nanjing 211106,Jiangsu,China; 2.Jiangsu Provincial Key Laboratory of Pulp and Paper Science and Technology,Nanjing 210037,Jiangsu,China; 3.Key Laboratory of Freshwater Fisheries and Germplasm Resources Utilization of the Ministry of Agriculture,Freshwater Fisheries Research Center,Chinese Academy of Fishery Sciences,Wuxi 214081,Jiangsu,China; 4.Key Laboratory of Healthy Mariculture for the East China Sea,Ministry of Agriculture,Xiamen 361021,Fujian,China
  • Received:2015-06-11 Revised:2016-01-07 Online:2016-05-25 Published:2016-04-12
  • Contact: 吴一全(1963-),男,博士,教授,博士生导师,主要从事图像处理与分析、目标检测与识别、智能信息处理研究. E-mail:nuaaimage@163.com
  • About author:吴一全(1963-),男,博士,教授,博士生导师,主要从事图像处理与分析、目标检测与识别、智能信息处理研究.
  • Supported by:
    Supported by the National Natural Science Foundation of China(61573183),the Jiangsu Provincial Key Labora- tory of Pulp and Paper Science and Technology(201313),the Key Laboratory of Freshwater Fisheries and Germplasm Resources Uti- lization(KF201313),the Key Laboratory of Healthy Magriculture for the East China Sea,Ministry of Agriculture(2013ESHML06) and the Priority Academic Program Development of Jiangsu Higher Education Institutions(2012)

摘要: 现有的 Arimoto 熵阈值法仅依赖于灰度直方图分布,且计算最佳阈值时需搜索整个解空间,效率不高. 为此,文中提出了一种二维 Arimoto 灰度熵阈值分割的快速迭代算法. 首先,提出了一维 Arimoto 灰度熵阈值选取的快速迭代算法;然后,考虑图像目标和背景的类内灰度均匀性,导出了基于灰度 - 平均灰度级直方图的 Arimoto 灰度熵阈值法,并给出了中间变量的快速递推公式;最后,提出了二维 Arimoto 灰度熵阈值选取的快速迭代算法,推导了相应的公式,大大减少了运算量. 实验结果表明,文中所提算法运行速度快,分割性能优于现有的 5 种同类阈值分割算法,分割后图像中的目标完整,边缘纹理清晰,细节更为丰富.

关键词: 图像分割, 阈值选取, 二维 Arimoto 灰度熵, 快速迭代算法

Abstract: As the existing Arimoto entropy-based thresholding methods only depend on the probability information from gray histogram and need to search the whole solution space to obtain the optimal threshold with low efficiency,a fast iterative algorithm for threshold segmentation on the basis of two-dimensional Arimoto gray entropy is pro- posed.Firstly,a fast iterative algorithm for threshold selection using one-dimensional Arimoto gray entropy is pro- posed.Secondly,by taking into consideration the gray level uniformity within the object cluster and the back- ground cluster,a two-dimensional Arimoto gray entropy thresholding method on the basis of gray level-average gray level histogram is derived.Then,fast recursive formulae for intermediate variables are given.Finally,a fast ite- rative algorithm is proposed for threshold selection on the basis of two-dimensional Arimoto gray entropy,and the corresponding algorithmic formulae are derived,which helps reduce computation burden greatly.Experimental results show that the proposed algorithm is superior to five existing threshold segmentation algorithms because it runs more rapidly and is more effective in obtaining segmented images with complete objects,clear edges and rich details.

Key words: image segmentation, threshold selection, two-dimensional Arimoto gray entropy, fast iterative algorithm

中图分类号: