一种改进的模拟超声图像算法的实现  

Improved simulating ultrasonic image algorithm and its implementation

作　　者：陈露兰 孙尽尧[1] CHEN Lu-lan;SUN Jin-yao(School of Physics and Technology,Wuhan University,Wuhan 430072,China)

机构地区：[1]武汉大学物理科学与技术学院

出　　处：《信息技术》2019年第8期58-61,共4页Information Technology

基　　金：国家重点基础研究发展计划(973)(CB707900)

摘　　要：为了评估各种去噪算法的性能,给理想无噪声图像添加可控噪声则成为一个重点也是难点。在合成超声图像的过程中,对模拟算法中最为成熟的Burckhardt散斑形成模型进行改进,同时结合双线性插值将一个理想的无噪声几何图像模拟成超声图像。通过双线性插值的提前加入去除后期采样造成的几何图像边缘的奇异点,使得模拟而成的图像在视觉和理论上都非常接近真实的超声波图。In order to evaluate the performance of various denoising algorithms,it is important and difficult to add controllable noise to ideal noise-free image.In the process of synthesizing ultrasonic image,Burckhardt speckle formation model,which is the most mature in the simulation algorithm,is improved,and an ideal noise-free geometric image is simulated into ultrasonic image with bilinear interpolation.The edge singularities of the geometric image caused by the late sampling are removed by adding the bilinear interpolation in advance so that the simulated image is very close to the real ultrasonic image visually and theoretically.

关 键 词：合成图像 散斑模拟 奇异点 双线性插值 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]