noise.wavelet

supreme.noise.wavelet.dwt_denoise(X[, ...])

Denoise an image using the Discrete Wavelet Transform.

dwt_denoise

supreme.noise.wavelet.dwt_denoise(X, wavelet='db8', levels=4, alpha=2)

Denoise an image using the Discrete Wavelet Transform.

Parameters:

X : ndarray of uint8

Image to denoise.

wavelet : str

Wavlet family to use. See supreme.lib.pywt.wavelist() for a complete list.

levels : int

Number of levels to use in the decomposition.

alpha : float

Parameter used to tweak the Wiener estimator. A larger value of alpha results in a smoother output.

Returns:

Y : ndarray of float64

Denoised image.

Notes

Implemented according to the overview of [R12] given in [R11].

References

[R11]

(1, 2) J. Fridrich, “Digital Image Forensics,” IEEE Signal Processing Magazine, vol. 26, 2009, pp. 26-37.

[R12]

(1, 2) M. Mihcak, I. Kozintsev, K. Ramchandran, and P. Moulin, “Low-complexity image denoising based on statistical modeling of wavelet coefficients,” IEEE Signal Processing Letters, vol. 6, 1999, pp. 300-303.