Spatially adaptive wavelet thresholding with context modeling for image denoising
IEEE Transactions on Image Processing
The contourlet transform: an efficient directional multiresolution image representation
IEEE Transactions on Image Processing
The Nonsubsampled Contourlet Transform: Theory, Design, and Applications
IEEE Transactions on Image Processing
Hi-index | 0.01 |
A new algorithm of crack image enhancement of straddle-type monorail track beam surface based on nonsubsampled contourlet transform (NSCT) is presented. It is according to the characteristics of different domain in NSCT, and the fractional differentiation can enhance the mid-frequency and retain the low frequency nonlinearly. Then a new enhancement method has been proposed that the smooth sub-band texture of NSCT domain can be enhanced by the fractional differentiation; while in the high-frequency sub-bands of NSCT domain, each pixel of high-frequency sub-bands is divided into strong edge, weak edge and noise on the basis of direction sensitivity characteristics, then the weak edge can be enhanced nonlinearly, strong edge be retained, and noise be removed. Experimental results show that the method proposed in this paper have greatly improved visual effects and larger contrast improve index (CII) as compared with wavelet transform, and the enhancement effect is very good.