ENHANCEMENT OF ULTRASOUND LIVER IMAGES

Abstract

Ultrasound imaging has become a popular modality because it is safe, non-invasive. However, the fundamental problem of ultrasound images is the poor quality, mainly caused by multiplicative speckle noise. Speckle is mainly caused by interference between coherent waves which are backscattered by targeted surfaces and arrives out of phase at the sensor. Speckle can be modelled as random noise (irregular pattern), which degrades the detection of low-contrast lesions and also reduces the ability of a human observer to resolve fine detail. So the quality of the image can be improved by using speckle reduction (filtering) techniques. In this Thesis we have implemented the Adaptive filtering techniques. We have compared Peak Valley filter with median, mean, morphological and Adaptive median filter by adding salt & pepper noise to the images. Peak valley filters performance better in the sense of both noise reduction and preserving image information. Finally smoothening has been done by processing image through windows adaptive threshold. The main drawbacks in adaptive techniques are "edge preserving" and "feature preserving". First, the filters are sensitive to the size and, shape of the filter window. Given a filter window that is too large (compared to the scale of interest), over-smoothing will occur and edges will be blurred. A small window will decrease the smoothing capability of the filter and will leave speckle. We have implemented the SRAD and Curvelet filter (recently proposed method for image Denoising) to overcome the above drawbacks. We have compared both Lee and frost filters with SRAD and Curvelet filters [l][