电子学报 ›› 2019, Vol. 47 ›› Issue (12): 2495-2504.DOI: 10.3969/j.issn.0372-2112.2019.12.007

• 学术论文 • 上一篇    下一篇

边缘引导和轮廓约束下的跨域香农熵最大化导向的自动阈值选取方法

邹耀斌1, 乔焰2, 孙水发1, 臧兆祥1, 夏平1, 王俊英1, 董方敏1, 龚国强1   

  1. 1. 三峡大学计算机与信息学院, 湖北宜昌 443002;
    2. 安徽农业大学信息与计算机学院, 安徽合肥 230036
  • 收稿日期:2019-01-18 修回日期:2019-06-28 出版日期:2019-12-25
    • 通讯作者:
    • 邹耀斌
    • 作者简介:
    • 乔焰 女,1984年生于河北邯郸.现为安徽农业大学副教授、硕士生导师.主要研究方向为物联网和机器学习.E-mail:qiaoyan101@gmail.com;孙水发 男,1977年生于江西黎川.现为三峡大学教授、硕士生导师.主要研究方向为图像处理和计算机视觉.E-mail:sunshuifa1977@yeah.net;臧兆祥 男,1985年生于云南曲靖.现为三峡大学副教授、硕士生导师.主要研究方向为机器学习和进化计算.E-mail:ambingo@zoho.com;夏平 男,1967年生于湖北麻城.现为三峡大学教授、硕士生导师.主要研究方向为计算机视觉、智能信息处理、多尺度几何分析及应用.E-mail:xiaping1967@163.com;王俊英 女,1971年生于湖北黄梅.现为三峡大学教授、硕士生导师.主要研究方向为模式识别与人工智能.E-mail:jywang1971@21cn.com;董方敏 男,1965年生于湖北荆门.现为三峡大学教授、博士生导师.主要研究方向为智能信息处理和图形图像处理.E-mail:fmdong_ctgu@126.com;龚国强 男,1976年生于湖北荆门.现为三峡大学副教授、硕士生导师.主要研究方向为无线通信和大数据处理.E-mail:gqianggong@21cn.com
    • 基金资助:
    • 国家重点研发计划资助项目 (No.2016YFB0800403); 国家自然科学基金 (No.61871258,No.61502274); 湖北省水电工程智能视觉监测重点实验室开放基金项目 (No.2017SDSJ04)

Automatic Threshold Selection Guided by Maximizing Cross-Region Shannon Entropy Under Edge Guidance and Contour Constraints

ZOU Yao-bin1, QIAO Yan2, SUN Shui-fa1, ZANG Zhao-xiang1, XIA Ping1, WANG Jun-ying1, DONG Fang-min1, GONG Guo-qiang1   

  1. 1. College of Computer and Information Technology, China Three Gorges University, Yichang, Hubei 443002, China;
    2. School of Information and Computer, Anhui Agricultural University, Hefei, Anhui 230036, China
  • Received:2019-01-18 Revised:2019-06-28 Online:2019-12-25 Published:2019-12-25
    • Corresponding author:
    • ZOU Yao-bin
    • Supported by:
    • Program supported by National Key Research and Development Program of China (No.2016YFB0800403); National Natural Science Foundation of China (No.61871258, No.61502274); Open Fund Project of Hubei Key Laboratory of Intelligent Vision Monitoring for Hydropower Engineering (No.2017SDSJ04)

摘要: 为了处理诸如高斯、伽马、极值、瑞利、均匀或贝塔等基本灰度分布情形下的阈值选取难题,本文提出了一种跨域香农熵最大化导向的自动阈值选取方法.该方法利用不变的引导边缘图像和变化的约束轮廓图像共同构造出一系列持续变化的一维灰度直方图,并采用香农熵作为熵计算模型,从而得以跨越图像中若干局部区域去计算跨域香农熵,并以最大跨域香农熵对应的阈值作为最终阈值.在40幅合成图像和50幅真实世界图像上的实验结果表明,该方法虽然在计算效率方面不优于Masi熵阈值方法、Tsallis熵阈值方法、局部香农熵阈值方法和迭代三类阈值方法,但在分割适应性方面有显著增强,且在误分割率方面有显著下降.

 

关键词: 阈值分割, 最大熵原理, 跨域香农熵, 香农熵差, 全局熵方法, 局部熵方法

Abstract: When the basic distribution constituting one gray level histogram is presented as a non-Gaussian distribution, such as gamma,  extreme value, Rayleigh, uniform or beta distribution, how to automatically select the best possible segmentation threshold is still quite challenging. To deal with the issue of threshold selection in the above-mentioned different gray level distributions, we propose an automatic method of threshold selection that is guided by maximizing cross-region Shannon entropy under edge guidance and contour constraints. This method utilizes constant guiding edges and dynamically changing contours to construct a series of continuously changing one-dimensional gray level histograms,and adopts Shannon entropy as the entropy calculation model. Therefore, it can calculate the cross-region Shannon entropy across several local regions in the image, and it takes the threshold corresponding to the maximum cross-region Shannon entropy as the final segmentation threshold. The proposed method is compared with Masi entropy thresholding, Tsallis entropy thresholding, Shannon entropy thresholding, and iterative triclass thresholding on 40 synthetic images and 50 real-world images. The results show that the proposed method is not superior to the 4 compared methods in computational efficiency, but it has significant enhancement in segmentation adaptability and a significant decrease in the mis-segmentation rate.

 

Key words: image thresholding, maximum entropy principle, cross-region Shannon entropy, Shannon entropy difference, global entropy method, local entropy method

中图分类号: