Deconvolution

The blur and speckle noise of the medical ultrasound image significantly reduces the diagnostic value of the images. They are caused by the representation of the true acoustic tissue reflectance. Many approaches have been proposed to solve this problem. One of them is the 2-D deconvolution. Deconvolution is used to restore the original image from the spacial response and the additive noise. Because of the limited bandwidth of the spacial response and the additive noise, perfect deconvolution is impossible. In stead, statistical restoration is used to remove the effect of the degradation by estimating the true reflectance image, such as commonly used Wiener filtering. Recent research shows that a number of choices other than Wiener filter gives better image quality.