摘要
在分析简单线性插值的基础上,使用简单线性插值、二次多项式插值和四次多项式对采集来的含有金属伪影的CT图像进行校正并对校正结果从定量和定性两个角度进行比较.实验结果表明,简单线性插值算法运行速度最快,二次多项式插值次之,四次多项式最慢.均方根差(RMSE)方面,原始含有金属伪影的CT图像最大,一次线性插值校正方法次之、二次线性插值校正方法再次之,四次线性插值校正方法最小.
In this paper,based on the analysis of simple linear interpolation,simple linear interpolation,quadratic polynomial interpolation and quartic polynomial were used to correct the collected CT images containing metal artifacts.The correction results were compared from quantitative and qualitative perspectives.The experimental results showed that the speed of simple linear interpolation algorithm was the fastest,followed by quadratic polynomial interpolation and quartic polynomial interpolation.In terms of root mean square error(RMSE),the original CT image with metal artifacts was the largest,followed by the first-order linear interpolation correction method,the second-order linear interpolation correction method,and the fourth-order linear interpolation correction method.
作者
李长春
李元金
LI Chang-chun;LI Yuan-jin(Information Engineering Institute,Chuzhou Polytechnic,Chuzhou 239000,China;Computer and Information Engineering Institute,Chuzhou University,Chuzhou 239000,China)
出处
《哈尔滨商业大学学报（自然科学版）》
CAS
2021年第4期436-442,共7页
Journal of Harbin University of Commerce：Natural Sciences Edition
基金
国家自然科学基金项目(No.2017YFC0107900)
安徽省教育厅重点研究项目(KJ2019A1136,KJ2018A0431,KJ2020A0992,KJ2020A0998)'安徽省高校优秀青年骨干人才国内外访学研修项目(gxgwfx2018077)
安徽省质量工程项目(2019xqsxzx31,2017kfk144)
滁州学院科研启动资金项目(2020qd15).
关键词
CT图像
金属伪影
插值算法
性能比较
均方根差
多项式
CT image
metal artifacts
interpolation algorithm
performance comparison
RMSE
polynomial
