Research On Medical Ultrasound Images De-noising Method Based On Alpha Stable Distribution

Ultrasonic detecting technology is a diagnostic method to detect disease and to give some guiding information about human body based on the data and shape of the physiological or organic structure. In clinical diagnostics, Ultrasound, X ray, Computerized Tomography and Nuclear Magnetic Resonance have been called the four medical imageology technologies, which are very important for doctors to diagnose. And ultrasound images diagnoses is an intuitionistic, convenient and effective examining measure without cut and pain, which is widely used. But the multiplicative speckle noises make the quality of ultrasound images decline and force doctors to make out a decision with other detecting methods.With respect to de-noising ultrasound images, there are two means. The first method is to scan the same object with different frequency, in different time and directions, and then get the final composed image. Another one is to filter the noise. Classical ultrasound image filtering measuresinclude mean filtering median filtering, Wiener filtering and wavelet filtering--softthresholding and hard thresholding, which lots of researchers now are advocated in. However, we are interested in the parts of images caused by pathological changes, those methods: composing and median filtering, Wiener filtering could not get the good de-noising effects. Although wavelet thresholding is better than the methods talked before, it still has some drawbacks such as how to select the optimal threshold etc. So in this paper we propose a new method to de-noise ultrasound...