Image denoising and contrast enhancement based on nonsubsampled contourlet transform

A method aimed at minimizing image noise while optimizing contrast of image subtle features based on nonsubsampled contourlet transform is presented in this paper. Nonsubsampled contourlet transform, which is a shift-invariant version of the contourlet transform, has better performance in representing image edges than separable wavelet for its anisotropy, directionality and shift-invariance, and is therefore well-suited for multi-scale edge enhancement. We modify the nonsubsampled contourlet coefficients of images in corresponding subbands via a new and operable nonlinear mapping function and take the noise into account for more precise reconstruction and better visualization. Experimental results on some medical images show that the proposed enhancement method effectively highlights subtle features while suppressing noise. A comparison with other enhancement algorithms, such as histogram equalization and contourlet-based enhancement approach, is also discussed.