首页    期刊浏览 2024年12月05日 星期四
登录注册

文章基本信息

  • 标题:Quantitative Analysis Of Hrct Images Using Knn Classifier
  • 本地全文:下载
  • 作者:Sinduja.J ; S.Prathiba
  • 期刊名称:International Journal of Computer Trends and Technology
  • 电子版ISSN:2231-2803
  • 出版年度:2013
  • 卷号:4
  • 期号:6-4
  • 出版社:Seventh Sense Research Group
  • 摘要:A Diabetic Retinopathy is retinopathy caused by complications of diabetes, which can eventually lead to blindness. Early stage of vision loss diagnosis is one of the challenge processes. In this project prior Diabetic retinopathy diagnosis implement based on SVM classifier. To improve the classifier accuracy depends upon the retinal blood vessel features. For blood vessel extraction proposes a function based on the evaluation of measurable features describing retinal blood vasculature. This paper proposes a function for vessel segmentation assessment based on vascular tree descriptive features with the aim of supplementing the existing quality evaluation functions (QEFs). Specifically, this new proposed function, denoted by CAL, evaluates vessel connectivity, area and length in a segmented image in comparison with those in a referencestandard image. Thus, this function is sensitive to vasculature features such as connectivity, area and length, and supplements widelyused metrics based on contingency tables.
  • 关键词:Emphysema; Local binary patterns; Quantitative computed tomography; Texture analysis; Tissue classification
国家哲学社会科学文献中心版权所有