restoration

Module: restoration

Image restoration module.

References

denoise_bilateral

skimage.restoration.denoise_bilateral(image, win_size=None, sigma_color=None, sigma_spatial=1, bins=10000, mode='constant', cval=0, multichannel=True, sigma_range=None) [source]

Denoise image using bilateral filter.

This is an edge-preserving and noise reducing denoising filter. It averages pixels based on their spatial closeness and radiometric similarity.

Spatial closeness is measured by the gaussian function of the euclidian distance between two pixels and a certain standard deviation (sigma_spatial).

Radiometric similarity is measured by the gaussian function of the euclidian distance between two color values and a certain standard deviation (sigma_color).

Parameters:

image : ndarray, shape (M, N[, 3]) Input image, 2D grayscale or RGB. win_size : int Window size for filtering. If win_size is not specified, it is calculated as max(5, 2*ceil(3*sigma_spatial)+1) sigma_color : float Standard deviation for grayvalue/color distance (radiometric similarity). A larger value results in averaging of pixels with larger radiometric differences. Note, that the image will be converted using the img_as_float function and thus the standard deviation is in respect to the range [0, 1]. If the value is None the standard deviation of the image will be used. sigma_spatial : float Standard deviation for range distance. A larger value results in averaging of pixels with larger spatial differences. bins : int Number of discrete values for gaussian weights of color filtering. A larger value results in improved accuracy. mode : {‘constant’, ‘edge’, ‘symmetric’, ‘reflect’, ‘wrap’} How to handle values outside the image borders. See numpy.pad for detail. cval : string Used in conjunction with mode ‘constant’, the value outside the image boundaries. multichannel : bool Whether the last axis of the image is to be interpreted as multiple channels or another spatial dimension.

Returns:

denoised : ndarray Denoised image.

References

[R324]

http://users.soe.ucsc.edu/~manduchi/Papers/ICCV98.pdf

Examples

>>> from skimage import data, img_as_float
>>> astro = img_as_float(data.astronaut())
>>> astro = astro[220:300, 220:320]
>>> noisy = astro + 0.6 * astro.std() * np.random.random(astro.shape)
>>> noisy = np.clip(noisy, 0, 1)
>>> denoised = denoise_bilateral(noisy, sigma_color=0.05, sigma_spatial=15)

denoise_nl_means

skimage.restoration.denoise_nl_means(image, patch_size=7, patch_distance=11, h=0.1, multichannel=True, fast_mode=True) [source]

Perform non-local means denoising on 2-D or 3-D grayscale images, and 2-D RGB images.

Parameters:

image : 2D or 3D ndarray Input image to be denoised, which can be 2D or 3D, and grayscale or RGB (for 2D images only, see multichannel parameter). patch_size : int, optional Size of patches used for denoising. patch_distance : int, optional Maximal distance in pixels where to search patches used for denoising. h : float, optional Cut-off distance (in gray levels). The higher h, the more permissive one is in accepting patches. A higher h results in a smoother image, at the expense of blurring features. For a Gaussian noise of standard deviation sigma, a rule of thumb is to choose the value of h to be sigma of slightly less. multichannel : bool, optional Whether the last axis of the image is to be interpreted as multiple channels or another spatial dimension. Set to False for 3-D images. fast_mode : bool, optional If True (default value), a fast version of the non-local means algorithm is used. If False, the original version of non-local means is used. See the Notes section for more details about the algorithms.

Returns:

result : ndarray Denoised image, of same shape as image.

Notes

The non-local means algorithm is well suited for denoising images with specific textures. The principle of the algorithm is to average the value of a given pixel with values of other pixels in a limited neighbourhood, provided that the patches centered on the other pixels are similar enough to the patch centered on the pixel of interest.

In the original version of the algorithm [R325], corresponding to fast=False, the computational complexity is

image.size * patch_size ** image.ndim * patch_distance ** image.ndim

Hence, changing the size of patches or their maximal distance has a strong effect on computing times, especially for 3-D images.

However, the default behavior corresponds to fast_mode=True, for which another version of non-local means [R326] is used, corresponding to a complexity of

image.size * patch_distance ** image.ndim

The computing time depends only weakly on the patch size, thanks to the computation of the integral of patches distances for a given shift, that reduces the number of operations [R325]. Therefore, this algorithm executes faster than the classic algorith (fast_mode=False), at the expense of using twice as much memory. This implementation has been proven to be more efficient compared to other alternatives, see e.g. [R327].

Compared to the classic algorithm, all pixels of a patch contribute to the distance to another patch with the same weight, no matter their distance to the center of the patch. This coarser computation of the distance can result in a slightly poorer denoising performance. Moreover, for small images (images with a linear size that is only a few times the patch size), the classic algorithm can be faster due to boundary effects.

The image is padded using the reflect mode of skimage.util.pad before denoising.

References

[R325]

(1, 2, 3) Buades, A., Coll, B., & Morel, J. M. (2005, June). A non-local algorithm for image denoising. In CVPR 2005, Vol. 2, pp. 60-65, IEEE.

[R326]

(1, 2) J. Darbon, A. Cunha, T.F. Chan, S. Osher, and G.J. Jensen, Fast nonlocal filtering applied to electron cryomicroscopy, in 5th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, 2008, pp. 1331-1334.

[R327]

(1, 2) Jacques Froment. Parameter-Free Fast Pixelwise Non-Local Means Denoising. Image Processing On Line, 2014, vol. 4, p. 300-326.

Examples

>>> a = np.zeros((40, 40))
>>> a[10:-10, 10:-10] = 1.
>>> a += 0.3 * np.random.randn(*a.shape)
>>> denoised_a = denoise_nl_means(a, 7, 5, 0.1)

denoise_tv_bregman

skimage.restoration.denoise_tv_bregman(image, weight, max_iter=100, eps=0.001, isotropic=True) [source]

Perform total-variation denoising using split-Bregman optimization.

Total-variation denoising (also know as total-variation regularization) tries to find an image with less total-variation under the constraint of being similar to the input image, which is controlled by the regularization parameter.

Parameters:

image : ndarray Input data to be denoised (converted using img_as_float`). weight : float Denoising weight. The smaller the weight, the more denoising (at the expense of less similarity to the input). The regularization parameter lambda is chosen as 2 * weight. eps : float, optional Relative difference of the value of the cost function that determines the stop criterion. The algorithm stops when: SUM((u(n) - u(n-1))**2) < eps max_iter : int, optional Maximal number of iterations used for the optimization. isotropic : boolean, optional Switch between isotropic and anisotropic TV denoising.

Returns:

u : ndarray Denoised image.

References

[R328]

http://en.wikipedia.org/wiki/Total_variation_denoising

[R329]

Tom Goldstein and Stanley Osher, “The Split Bregman Method For L1 Regularized Problems”, ftp://ftp.math.ucla.edu/pub/camreport/cam08-29.pdf

[R330]

Pascal Getreuer, “Rudin–Osher–Fatemi Total Variation Denoising using Split Bregman” in Image Processing On Line on 2012–05–19, http://www.ipol.im/pub/art/2012/g-tvd/article_lr.pdf

[R331]

http://www.math.ucsb.edu/~cgarcia/UGProjects/BregmanAlgorithms_JacquelineBush.pdf

denoise_tv_chambolle

skimage.restoration.denoise_tv_chambolle(im, weight=0.1, eps=0.0002, n_iter_max=200, multichannel=False) [source]

Perform total-variation denoising on n-dimensional images.

Parameters:

im : ndarray of ints, uints or floats Input data to be denoised. im can be of any numeric type, but it is cast into an ndarray of floats for the computation of the denoised image. weight : float, optional Denoising weight. The greater weight, the more denoising (at the expense of fidelity to input). eps : float, optional Relative difference of the value of the cost function that determines the stop criterion. The algorithm stops when: (E_(n-1) - E_n) < eps * E_0 n_iter_max : int, optional Maximal number of iterations used for the optimization. multichannel : bool, optional Apply total-variation denoising separately for each channel. This option should be true for color images, otherwise the denoising is also applied in the channels dimension.

Returns:

out : ndarray Denoised image.

Notes

Make sure to set the multichannel parameter appropriately for color images.

The principle of total variation denoising is explained in http://en.wikipedia.org/wiki/Total_variation_denoising

The principle of total variation denoising is to minimize the total variation of the image, which can be roughly described as the integral of the norm of the image gradient. Total variation denoising tends to produce “cartoon-like” images, that is, piecewise-constant images.

This code is an implementation of the algorithm of Rudin, Fatemi and Osher that was proposed by Chambolle in [R332].

References

[R332]

(1, 2) A. Chambolle, An algorithm for total variation minimization and applications, Journal of Mathematical Imaging and Vision, Springer, 2004, 20, 89-97.

Examples

2D example on astronaut image:

>>> from skimage import color, data
>>> img = color.rgb2gray(data.astronaut())[:50, :50]
>>> img += 0.5 * img.std() * np.random.randn(*img.shape)
>>> denoised_img = denoise_tv_chambolle(img, weight=60)

3D example on synthetic data:

>>> x, y, z = np.ogrid[0:20, 0:20, 0:20]
>>> mask = (x - 22)**2 + (y - 20)**2 + (z - 17)**2 < 8**2
>>> mask = mask.astype(np.float)
>>> mask += 0.2*np.random.randn(*mask.shape)
>>> res = denoise_tv_chambolle(mask, weight=100)

nl_means_denoising

skimage.restoration.nl_means_denoising(*args, **kwargs) [source]

Deprecated function. Use skimage.restoration.denoise_nl_means instead.

Perform non-local means denoising on 2-D or 3-D grayscale images, and 2-D RGB images.

Parameters:

image : 2D or 3D ndarray Input image to be denoised, which can be 2D or 3D, and grayscale or RGB (for 2D images only, see multichannel parameter). patch_size : int, optional Size of patches used for denoising. patch_distance : int, optional Maximal distance in pixels where to search patches used for denoising. h : float, optional Cut-off distance (in gray levels). The higher h, the more permissive one is in accepting patches. A higher h results in a smoother image, at the expense of blurring features. For a Gaussian noise of standard deviation sigma, a rule of thumb is to choose the value of h to be sigma of slightly less. multichannel : bool, optional Whether the last axis of the image is to be interpreted as multiple channels or another spatial dimension. Set to False for 3-D images. fast_mode : bool, optional If True (default value), a fast version of the non-local means algorithm is used. If False, the original version of non-local means is used. See the Notes section for more details about the algorithms.

Returns:

result : ndarray Denoised image, of same shape as image.

Notes

The non-local means algorithm is well suited for denoising images with specific textures. The principle of the algorithm is to average the value of a given pixel with values of other pixels in a limited neighbourhood, provided that the patches centered on the other pixels are similar enough to the patch centered on the pixel of interest.

In the original version of the algorithm [R333], corresponding to fast=False, the computational complexity is

image.size * patch_size ** image.ndim * patch_distance ** image.ndim

Hence, changing the size of patches or their maximal distance has a strong effect on computing times, especially for 3-D images.

However, the default behavior corresponds to fast_mode=True, for which another version of non-local means [R334] is used, corresponding to a complexity of

image.size * patch_distance ** image.ndim

The computing time depends only weakly on the patch size, thanks to the computation of the integral of patches distances for a given shift, that reduces the number of operations [R333]. Therefore, this algorithm executes faster than the classic algorith (fast_mode=False), at the expense of using twice as much memory. This implementation has been proven to be more efficient compared to other alternatives, see e.g. [R335].

Compared to the classic algorithm, all pixels of a patch contribute to the distance to another patch with the same weight, no matter their distance to the center of the patch. This coarser computation of the distance can result in a slightly poorer denoising performance. Moreover, for small images (images with a linear size that is only a few times the patch size), the classic algorithm can be faster due to boundary effects.

The image is padded using the reflect mode of skimage.util.pad before denoising.

References

[R333]

(1, 2, 3) Buades, A., Coll, B., & Morel, J. M. (2005, June). A non-local algorithm for image denoising. In CVPR 2005, Vol. 2, pp. 60-65, IEEE.

[R334]

(1, 2) J. Darbon, A. Cunha, T.F. Chan, S. Osher, and G.J. Jensen, Fast nonlocal filtering applied to electron cryomicroscopy, in 5th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, 2008, pp. 1331-1334.

[R335]

(1, 2) Jacques Froment. Parameter-Free Fast Pixelwise Non-Local Means Denoising. Image Processing On Line, 2014, vol. 4, p. 300-326.

Examples

>>> a = np.zeros((40, 40))
>>> a[10:-10, 10:-10] = 1.
>>> a += 0.3 * np.random.randn(*a.shape)
>>> denoised_a = denoise_nl_means(a, 7, 5, 0.1)

richardson_lucy

skimage.restoration.richardson_lucy(image, psf, iterations=50, clip=True) [source]

Richardson-Lucy deconvolution.

Parameters:	image : ndarray Input degraded image (can be N dimensional). psf : ndarray The point spread function. iterations : int Number of iterations. This parameter plays the role of regularisation. clip : boolean, optional True by default. If true, pixel value of the result above 1 or under -1 are thresholded for skimage pipeline compatibility.
Returns:	im_deconv : ndarray The deconvolved image.

References

[R336]

http://en.wikipedia.org/wiki/Richardson%E2%80%93Lucy_deconvolution

Examples

>>> from skimage import color, data, restoration
>>> camera = color.rgb2gray(data.camera())
>>> from scipy.signal import convolve2d
>>> psf = np.ones((5, 5)) / 25
>>> camera = convolve2d(camera, psf, 'same')
>>> camera += 0.1 * camera.std() * np.random.standard_normal(camera.shape)
>>> deconvolved = restoration.richardson_lucy(camera, psf, 5)

unsupervised_wiener

skimage.restoration.unsupervised_wiener(image, psf, reg=None, user_params=None, is_real=True, clip=True) [source]

Unsupervised Wiener-Hunt deconvolution.

Return the deconvolution with a Wiener-Hunt approach, where the hyperparameters are automatically estimated. The algorithm is a stochastic iterative process (Gibbs sampler) described in the reference below. See also wiener function.

Parameters:	image : (M, N) ndarray The input degraded image. psf : ndarray The impulse response (input image’s space) or the transfer function (Fourier space). Both are accepted. The transfer function is automatically recognized as being complex (np.iscomplexobj(psf)). reg : ndarray, optional The regularisation operator. The Laplacian by default. It can be an impulse response or a transfer function, as for the psf. user_params : dict Dictionary of parameters for the Gibbs sampler. See below. clip : boolean, optional True by default. If true, pixel values of the result above 1 or under -1 are thresholded for skimage pipeline compatibility.
Returns:	x_postmean : (M, N) ndarray The deconvolved image (the posterior mean). chains : dict The keys noise and prior contain the chain list of noise and prior precision respectively.
Other Parameters:
	The keys of ``user_params`` are: threshold : float The stopping criterion: the norm of the difference between to successive approximated solution (empirical mean of object samples, see Notes section). 1e-4 by default. burnin : int The number of sample to ignore to start computation of the mean. 100 by default. min_iter : int The minimum number of iterations. 30 by default. max_iter : int The maximum number of iterations if threshold is not satisfied. 150 by default. callback : callable (None by default) A user provided callable to which is passed, if the function exists, the current image sample for whatever purpose. The user can store the sample, or compute other moments than the mean. It has no influence on the algorithm execution and is only for inspection.

Notes

The estimated image is design as the posterior mean of a probability law (from a Bayesian analysis). The mean is defined as a sum over all the possible images weighted by their respective probability. Given the size of the problem, the exact sum is not tractable. This algorithm use of MCMC to draw image under the posterior law. The practical idea is to only draw highly probable images since they have the biggest contribution to the mean. At the opposite, the less probable images are drawn less often since their contribution is low. Finally the empirical mean of these samples give us an estimation of the mean, and an exact computation with an infinite sample set.

References

[R337]

François Orieux, Jean-François Giovannelli, and Thomas Rodet, “Bayesian estimation of regularization and point spread function parameters for Wiener-Hunt deconvolution”, J. Opt. Soc. Am. A 27, 1593-1607 (2010) http://www.opticsinfobase.org/josaa/abstract.cfm?URI=josaa-27-7-1593 http://research.orieux.fr/files/papers/OGR-JOSA10.pdf

Examples

>>> from skimage import color, data, restoration
>>> img = color.rgb2gray(data.astronaut())
>>> from scipy.signal import convolve2d
>>> psf = np.ones((5, 5)) / 25
>>> img = convolve2d(img, psf, 'same')
>>> img += 0.1 * img.std() * np.random.standard_normal(img.shape)
>>> deconvolved_img = restoration.unsupervised_wiener(img, psf)

unwrap_phase

skimage.restoration.unwrap_phase(image, wrap_around=False, seed=None) [source]

Recover the original from a wrapped phase image.

From an image wrapped to lie in the interval [-pi, pi), recover the original, unwrapped image.

Parameters:	image : 1D, 2D or 3D ndarray of floats, optionally a masked array The values should be in the range [-pi, pi). If a masked array is provided, the masked entries will not be changed, and their values will not be used to guide the unwrapping of neighboring, unmasked values. Masked 1D arrays are not allowed, and will raise a ValueError. wrap_around : bool or sequence of bool, optional When an element of the sequence is True, the unwrapping process will regard the edges along the corresponding axis of the image to be connected and use this connectivity to guide the phase unwrapping process. If only a single boolean is given, it will apply to all axes. Wrap around is not supported for 1D arrays. seed : int, optional Unwrapping 2D or 3D images uses random initialization. This sets the seed of the PRNG to achieve deterministic behavior.
Returns:	image_unwrapped : array_like, double Unwrapped image of the same shape as the input. If the input image was a masked array, the mask will be preserved.
Raises:	ValueError If called with a masked 1D array or called with a 1D array and wrap_around=True.

References

[R338]

Miguel Arevallilo Herraez, David R. Burton, Michael J. Lalor, and Munther A. Gdeisat, “Fast two-dimensional phase-unwrapping algorithm based on sorting by reliability following a noncontinuous path”, Journal Applied Optics, Vol. 41, No. 35 (2002) 7437,

[R339]

Abdul-Rahman, H., Gdeisat, M., Burton, D., & Lalor, M., “Fast three-dimensional phase-unwrapping algorithm based on sorting by reliability following a non-continuous path. In W. Osten, C. Gorecki, & E. L. Novak (Eds.), Optical Metrology (2005) 32–40, International Society for Optics and Photonics.

Examples

>>> c0, c1 = np.ogrid[-1:1:128j, -1:1:128j]
>>> image = 12 * np.pi * np.exp(-(c0**2 + c1**2))
>>> image_wrapped = np.angle(np.exp(1j * image))
>>> image_unwrapped = unwrap_phase(image_wrapped)
>>> np.std(image_unwrapped - image) < 1e-6   # A constant offset is normal
True

wiener

skimage.restoration.wiener(image, psf, balance, reg=None, is_real=True, clip=True) [source]

Wiener-Hunt deconvolution

Return the deconvolution with a Wiener-Hunt approach (i.e. with Fourier diagonalisation).

Parameters:

image : (M, N) ndarray Input degraded image psf : ndarray Point Spread Function. This is assumed to be the impulse response (input image space) if the data-type is real, or the transfer function (Fourier space) if the data-type is complex. There is no constraints on the shape of the impulse response. The transfer function must be of shape (M, N) if is_real is True, (M, N // 2 + 1) otherwise (see np.fft.rfftn). balance : float The regularisation parameter value that tunes the balance between the data adequacy that improve frequency restoration and the prior adequacy that reduce frequency restoration (to avoid noise artifacts). reg : ndarray, optional The regularisation operator. The Laplacian by default. It can be an impulse response or a transfer function, as for the psf. Shape constraint is the same as for the psf parameter. is_real : boolean, optional True by default. Specify if psf and reg are provided with hermitian hypothesis, that is only half of the frequency plane is provided (due to the redundancy of Fourier transform of real signal). It’s apply only if psf and/or reg are provided as transfer function. For the hermitian property see uft module or np.fft.rfftn. clip : boolean, optional True by default. If True, pixel values of the result above 1 or under -1 are thresholded for skimage pipeline compatibility.

Returns:

im_deconv : (M, N) ndarray The deconvolved image.

Notes

This function applies the Wiener filter to a noisy and degraded image by an impulse response (or PSF). If the data model is

where is noise, the PSF and the unknown original image, the Wiener filter is

where and are the Fourier and inverse Fourier transfroms respectively, the transfer function (or the Fourier transfrom of the PSF, see [Hunt] below) and the filter to penalize the restored image frequencies (Laplacian by default, that is penalization of high frequency). The parameter tunes the balance between the data (that tends to increase high frequency, even those coming from noise), and the regularization.

These methods are then specific to a prior model. Consequently, the application or the true image nature must corresponds to the prior model. By default, the prior model (Laplacian) introduce image smoothness or pixel correlation. It can also be interpreted as high-frequency penalization to compensate the instability of the solution with respect to the data (sometimes called noise amplification or “explosive” solution).

Finally, the use of Fourier space implies a circulant property of , see [Hunt].

References

[R340]

François Orieux, Jean-François Giovannelli, and Thomas Rodet, “Bayesian estimation of regularization and point spread function parameters for Wiener-Hunt deconvolution”, J. Opt. Soc. Am. A 27, 1593-1607 (2010) http://www.opticsinfobase.org/josaa/abstract.cfm?URI=josaa-27-7-1593 http://research.orieux.fr/files/papers/OGR-JOSA10.pdf

[R341]

B. R. Hunt “A matrix theory proof of the discrete convolution theorem”, IEEE Trans. on Audio and Electroacoustics, vol. au-19, no. 4, pp. 285-288, dec. 1971

Examples

>>> from skimage import color, data, restoration
>>> img = color.rgb2gray(data.astronaut())
>>> from scipy.signal import convolve2d
>>> psf = np.ones((5, 5)) / 25
>>> img = convolve2d(img, psf, 'same')
>>> img += 0.1 * img.std() * np.random.standard_normal(img.shape)
>>> deconvolved_img = restoration.wiener(img, psf, 1100)