Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
1760526 | Ultrasound in Medicine & Biology | 2013 | 14 Pages |
Abstract
A new adaptive wavelet packet-based approach to minimize speckle noise in ultrasound images is proposed. This method combines wavelet packet thresholding with a bilateral filter. Here, the best bases after wavelet packet decomposition are selected by comparing the first singular value of all sub-bands, and the noisy coefficients are thresholded using a modified NeighShrink technique. The algorithm is tested with various ultrasound images, and the results, in terms of peak signal-to-noise ratio and mean structural similarity values, are compared with those for some well-known de-speckling techniques. The simulation results indicate that the proposed method has better potential to minimize speckle noise and retain fine details of the ultrasound image.
Keywords
Related Topics
Physical Sciences and Engineering
Physics and Astronomy
Acoustics and Ultrasonics
Authors
Sankaralingam Esakkirajan, Chinna Thambi Vimalraj, Rashad Muhammed, Ganapathi Subramanian,