计算机科学 ›› 2012, Vol. 39 ›› Issue (Z6): 485-488.

• • 上一篇    下一篇

两种超声颈动脉血管斑块图像分割方法比较与改进

金娇英,王龙会,丁明跃   

  1. (华中科技大学生命科学与技术学院 图像信息处理与智能控制教育部重点实验室 武汉430074)
  • 出版日期:2018-11-16 发布日期:2018-11-16

Comparing and Improvement of Two Carotid Ultrasound Image Vascular Plaque Segmentation Methods

  • Online:2018-11-16 Published:2018-11-16

摘要: 针对颈动脉超声图像,实现了两种颈动脉血管斑块的分割方法—活动形状模型(Active Shape Models,ASM)和活动表观模型(Active Appearance Models,AAM),对38组颈动脉超声图像进行了内外轮廓分割,并比较了两类算法对颈动脉内外轮廓分割的有效性。在综合分析实验结果的基础上,结合颈动脉超声图像的特点,通过引入比例不变性改进了ASM算法。统计结果表明,在运行时间上,ASM和改进ASM的运行时间相近,AAM大约为ASM和改进ASM的16倍。同时,采用FOM和RAY两种方法对分割效果进行评价,结果表明,改进ASM算法的分割效果较ASM有了很大的提高,是最适合颈动脉血管斑块超声图像分割的算法。

关键词: 图像分割,颈动脉超声图像,活动形状模型,活动表观模型

Abstract: This article presented two image segmentation methods on carotid ultrasound images of vascular plactues,i. e. , active shape models and active appearance models, and compares their effectiveness in segmenting the internal and external contours of carotid ultrasound images after segmentation of 38 groups of carotid ultrasound images. Based on comprehensive analysis of experimental results and considering the characteristics of carotid ultrasound images,the ASM method was improved by adding the scale invariant. Experimental results showed that in running time, ASM was close to the improved ASM, while AAM was about 16 times longer than ASM and improved ASM. By evaluating their operating efficiency with FOM and RAY methods, it was demonstrated that the improved ASM was much better than ASM and the improved ASM was the most effective algorithm for carotid ultrasound images segmentation.

Key words: Image segmentation,Carotid ultrasound images,Active shape models,Active appearance models

No related articles found!
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!