J. Yu and Y. Wang (PRC)
Ultrasonic speckle reduction, contourlet transform based complex diffusion, nonsubsampled contourlet transform, nonlinear complex diffusion.
A novel multiscale anisotropic diffusion method is proposed for ultrasonic speckle reduction. This method, which extends the nonlinear complex diffusion to a nonsubsampled contourlet domain, is called a contourlet transform based complex diffusion (CTCD). The nonsubsampled contourlet transform provides a flexible, multiscale, multidirection, and shift-invariant image decomposition with a good contour capturing and representation. Then, the nonlinear complex diffusion removes noise from the contourlet coefficients with the imaginary values serving as a robust edge-detector. The directionality and multiresolution features of the nonsubsampled contourlet transform and the robust anisotropic smoothing feature of the complex diffusion make the hybrid filter produce satisfactory despeckling results. Both synthetic and real ultrasound images are used to evaluate the performance of the filter, and denoising results are also compared with other three competitive speckle reduction algorithms. It is shown that the proposed method is superior to other methods in both noise reduction and structure preservation.
Important Links:
Go Back
