超声图像中的speckle噪声是如何产生的？

A tissue is often modeled as an aggregate of small sub wavelength point scatters. The scattered signal
from such an object is then the sum of many small reflections whose phase is uniformly distributed
between 0 and 2 pi . By central limit theorem, the resulting signal can be viewed as a complex Gaussian
random variable. Note that the detection of the signal is done non-coherently, only the envelope of the
signal is used for imaging. The pixel intensity follows a Raleigh distribution. In the presence of a strong
scattering object (specular scattering), the pixel intensity follows a Rician distribution. The impact of this
shows up as 2D alternating regions of high and low intensity in the image. This effect is called speckle .

 

注：超声波scattering的方式共有三种：specular（镜面反射），diffusive（散射），diffractive（衍射）。

 

Raleigh distribution：

http://en.wikipedia.org/wiki/Rayleigh_distribution

 

Rician distribution：

http://en.wikipedia.org/wiki/Rice_distribution