基于相位重组近邻点建构的虚拟成像索引技术  

Virtual Image Indexing Technology Based on Neighbor Construction and Phase Reconstruction

作　　者：张杰[1] 丁玲[1] 宗克玄[1] 

机构地区：[1]沧州医学高等专科学校基础部,河北沧州061001

出　　处：《科技通报》2014年第6期182-184,共3页Bulletin of Science and Technology

基　　金：沧州市教育科学研究"十二五"规划课题(201206006)

摘　　要：提出一种考虑相位重组近邻点建构的虚拟成像索引技术。针对强脉冲噪声污染下的虚拟成图像进行图像降噪滤波模型构建,求取噪声像点的近邻点,通过对数字图像的像点的近邻点噪声序列进行高斯伪随机数排序,以近邻点为中心进行相位重新建构,实现对虚拟成像的降噪滤波和检测索引。仿真实验证明,新的图像索引技术能有效强脉冲噪声污染虚拟成像进行高精度低均方根误差索引,滤波效果卓越,比传统算法性能有大幅提高,其优越的图像降噪处理和索引性能将在医学成像处理、空间探测等虚拟成像处理领域有较好的工程实践价值。An improved virtual image indexing technology was proposed based on neighbor construction and phase recon-struction. The strong pulse noise pollution of the virtual image noise filtering model construction was got, the neighbor points of image was calculated. The mean square error of the virtual pixel index imaging as PSNR and MSE was constructed as the quantitative evaluation index for analyzing the quality of the algorithm. Simulation result shows that the new method can get the best indexing and filtering effect for the strong polluted images, and the indexing precision and the MSE are per-fect with great improvement compare to the traditional algorithm. Its superior performance will have great value in practice such as in medical imaging processing, virtual imaging space exploration etc.

关 键 词：相位重组 近邻点 图像索引 降噪滤波 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]