计算机工程与应用 ›› 2020, Vol. 56 ›› Issue (9): 221-227.DOI: 10.3778/j.issn.1002-8331.1902-0192

• 图形图像处理 • 上一篇    下一篇

Fisher-Tippet分布拟合的超声图像联合双边滤波方法

李蒙蒙,邵良志,崔文超,孙水发   

  1. 三峡大学 计算机与信息学院,湖北 宜昌 443002
  • 出版日期:2020-05-01 发布日期:2020-04-29

Joint Bilateral Filtering Method for Ultrasound Image by Fisher-Tippett Distribution Fitting

LI Mengmeng, SHAO Liangzhi, CUI Wenchao, SUN Shuifa   

  1. College of Computer and Information Technology, China Three Gorges University, Yichang, Hubei 443002, China
  • Online:2020-05-01 Published:2020-04-29

摘要:

由于双边滤波的固有不足以及Rayleigh分布散斑拟合存在的较大误差,致使SRBF方法对超声图像的散斑消减不够理想。针对该问题,采用联合双边滤波以及Fisher-Tippet分布散斑拟合来有效改进SRBF方法。联合双边滤波利用导向图像计算权值,有效减小双边滤波利用噪声图像计算权值产生的偏差。由散斑包络信号满足Rayleigh分布导出散斑图像满足Fisher-Tippet分布,更精确描述散斑统计特性。合成图像和真实图像的实验结果通过多种定量指标评估,验证了提出算法对SRBF在两个方面的改进都能有效提高散斑消减性能。

关键词: 超声图像, 散斑消减双边滤波(SRBF), 联合双边滤波, Fisher-Tippet分布

Abstract:

The SRBF method often gets poor performance on speckle reduction due to the inherent defect of bilateral filtering and the large bias from fitting speck lewith Rayleigh distribution. Aiming at the problem, two modifications namely joint bilateral filtering and Fisher-Tippet distribution fitting are used to improve the SRBF method effectively. Joint bilateral filtering leads to smaller bias on weights than that of traditional bilateral filtering. The Fisher-Tippet distribution that derives from the Rayleigh distribution of speckle envelope describes the statistical character of speckle image better than the Rayleigh distribution. It can be verified by quantitative evaluation on synthetic images and real images that the proposed algorithm enhances the despeckling performance based on either of these two modifications of the SRBF method.

Key words: ultrasound image, Speakle Reduction Bilateral Filtering(SRBF), joint bilateral filtering, Fisher-Tippet distribution