节点文献

形状自由可调的隶属函数及其模糊图像增强应用

A New Method of Adjusting Freely Shape of Fuzzy Membership Function and Its Application to Enhancing Image Better

  • 推荐 CAJ下载
  • PDF下载
  • 不支持迅雷等下载工具,请取消加速工具后下载。

【作者】 张大奇曲仕茹何力

【Author】 Zhang Daqi,Qu Shiru,He Li (Department of Automatic Control,Northwestern Polytechnical University,Xi′an 710072,China)

【机构】 西北工业大学自动化学院

【摘要】 借助非均匀有理B样条在自由曲线设计方面的优势,给出了一种能够根据实际需要设计生成模糊隶属度函数的新方法。利用该方法生成了目前常用的5种类型模糊隶属函数,最后以设计生成模糊图像增强算法中的"S"型隶属度函数为例说明了新方法的便易和有效性。

【Abstract】 Section 3 of the full paper applies the powerful NURBS(Non-Uniform Rational B-Spline) method to adjusting freely the shape of each of the five types of membership function:(1) exponential,(2) S-shape,(3) tangential,(4) trapezoidal,(5) Gaussian.Figs.1 through 5 of the full paper,giving the results of application corresponding respectively to the five types of membership function,show that our new method does succeed in achieving free adjustment.Section 4 applies the S-shape membership function to image enhancement.Fig.7 compares the image enhancement obtained with S-shape membership function by our new method with those obtained by previous methods;the comparison shows preliminarily that our new method enhances image better.

【基金】 陕西省科学技术研究发展计划项目(2008K07-14)资助
  • 【文献出处】 西北工业大学学报 ,Journal of Northwestern Polytechnical University , 编辑部邮箱 ,2009年03期
  • 【分类号】TP391.41
  • 【被引频次】8
  • 【下载频次】215
节点文献中: 

本文链接的文献网络图示:

本文的引文网络