摘要
双源CT已被广泛应用于泌尿系统疾病的诊断。然而,双源CT中的高辐射剂量风险引起了人们广泛的关注。本文利用向量全变分模型对双源CT图像进行去噪,向量全变分方法不仅保持了二维全变分方法的边缘保持特性,而且利用了双源CT能量间的梯度关系。针对此去噪模型,本文提出了一种优化策略,此策略利用了全变分模型的对偶形式。仿真数据实验表明,向量全变分方法不但可以消除图像噪声,还可以更好地保持图像的边缘和结构信息。
Dual source CT （DSCT） has been widely applied in the diagnosis of urinary system diseases. However, the high radiation dose risk in DSCT has raised growing concerns. A vector total variation model was applied to denoise the DSCT image. The vector total variation method kept the characteristic on edge preservation of two-dimensional total variation method and made use of gradient among the DSCT energy. Based on the denoising model, an optimized strategy, taking the advantage of dual form of total variation model, was proposed in this paper. The experimental results showed that the total variation method eliminated image noise and properly preserved the image edges and structural information.
出处
《中国医学物理学杂志》
CSCD
2015年第5期721-723,732,共4页
Chinese Journal of Medical Physics
基金
广东省科技计划项目(412040701040)
关键词
双源CT
向量
全变分
泌尿系统疾病
图像去噪
dual source CT
vector
total variation
urinary system disease
image denoising
