Poisson噪音在小波变换域内的分布特性及去除方法

DISTRIBUTION PROPERTIES OF POISSON NOISE IN WAVELET TRANSFORMING DOMAIN AND THE CORRESPONDING REMOVAL METHOD

本文推导出 Poisson噪音在小波变换下随尺度变化的变化公式 ,得出 Poisson噪音经小波变换之后其方差随尺度增大而减小的结论 ,从理论上保证了应用小波变换多尺度方法去除 Poisson噪音的可行性与有效性 .针对 Poisson噪音不具有一致分布性 ,即它在各点的方差与信号的强度有关 ,本文提出了一种基于小波变换的变阈值滤波算法 ,文末的仿真试验结果说明本文提出的算法对去除 Poisson噪音是十分有效的 .

This paper derives a formula of Poisson noise with the decomposition scale in wavelet-domain from which authors can find that the variation of Poisson noise in wavelet-domain decreases as the scale increases.It theoretically ensures that the wavelet multiscale method can be used for Poisson noise removal.According to the unevenly distributed property of Poisson noise,a threshold-variant wavelet multiple filtering algorithm is presented.The experimental results show that the algorithm presented in this paper is very efficient for Poisson noise reduction.