基于Ridgedet变换和极小极大估计的图像去噪  

Image denoising based on finite ridgelet transform and minimax estimation

作　　者：赵振磊[1] 耿则勋[1] 张亚新[1] 

机构地区：[1]信息工程大学测绘学院,郑州450052

出　　处：《测绘科学》2009年第3期31-33,共3页Science of Surveying and Mapping

基　　金：国家863计划项目(2007AA703418A;2006AA12Z110);国家自然科学基金(60778051)

摘　　要：Ridegdet是继小波变换后一种新的图像多尺度分析方法,它能有效地对图像进行多尺度、多方向的描述,对于图像中的直线状和超平面的奇异性问题,Ridgelet变换具有比小波变换更好的处理效果。极小化极大估计是对"最不利"先验分布的贝叶斯估计,本文利用极小化极大原理来获取阈值,实现了一种基于有限Ridgelet变换和极小极大估计的图像去噪算法,实验结果表明,对于含有较多线状目标的图像,该方法取得了比小波变换更好的效果。Ridgelet transform, which is suitable for describing the image in multi-scales and multi-directions, is one novel method of image multi-scale geometric analysis succeeding the wavelet transform, especially for the singularity problems of beeline and hyper- plane in image. It is more effective than wavelet transform. Minimax estimation is the Bayesian estimation of the ＂worst＂ apriori distribution. Using minimax estimator to obtain the thresholds, an image denoising algorithm based on Finite Ridgelet Transform and Minimax estimation is proposed. Experiments show that it is more effective than the normal wavelet transform especially for images with many obvious line boundaries.

关 键 词：小波变换 RADON变换 RIDGELET变换 图像去噪 极小极大估计 

分 类 号：TP751[自动化与计算机技术—检测技术与自动化装置]