From 11599f772a82207e51b73cc928f11f0e4123f415 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Fran=C3=A7ois=20Orieux?= Date: Mon, 28 Oct 2013 15:00:39 +0100 Subject: [PATCH] Typo fix. --- skimage/deconvolution/wiener.py | 23 ++++++++++++++--------- 1 file changed, 14 insertions(+), 9 deletions(-) diff --git a/skimage/deconvolution/wiener.py b/skimage/deconvolution/wiener.py index 05b0c854..b789d026 100644 --- a/skimage/deconvolution/wiener.py +++ b/skimage/deconvolution/wiener.py @@ -74,7 +74,7 @@ def wiener(data, psf, reg_val, reg=None, real=True): Returns ------- im_deconv : (M, N) ndarray - The deconvolued data + The deconvolved data Examples -------- @@ -94,7 +94,7 @@ def wiener(data, psf, reg_val, reg=None, real=True): .. math:: y = Hx + n - where :math:`n is the noise`, :math:`H` the psf and :math:`x` the + where :math:`n` is noise, :math:`H` the PSF and :math:`x` the unknown original image, the wiener filter is .. math:: \hat x = F^\dag (|\Lambda_H|^2 + \lambda |\Lambda_D|^2) \Lambda_H^\dag F y @@ -102,13 +102,18 @@ def wiener(data, psf, reg_val, reg=None, real=True): where :math:`F` and :math:`F^\dag` is the Fourier and inverse Fourier transfrom, :math:`\Lambda_H` the transfert function (or the Fourier transfrom of the PSF, see [2]) and :math:`\Lambda_D` - the filter to penalized the restored image frequency (laplacian by - default, that is penalization of high frequency). The parameter - :math:`\lambda` tune the balance between the data (that tends to - increase high frequency, even the noise), and the regularization. + the filter to penalize the restored image frequencies (laplacian + by default, that is penalization of high frequency). The parameter + :math:`\lambda` tunes the balance between the data (that tends to + increase high frequency, even those coming from noise), and the + regularization. - These methods are then specifique to a prior model that must be - adequate. They could be refered to bayesian approaches. + These methods are then specifique to a prior model that must match + the application (smoothness by default). They could be refered to + bayesian approaches. + + The use of Fourier space implies a circulant property of + :math:`H`, see [2]. References ---------- @@ -175,7 +180,7 @@ def unsupervised_wiener(data, psf, reg=None, user_params=None): Returns ------- x_postmean : (M, N) ndarray - The deconvolued data (the posterior mean) + The deconvolved data (the posterior mean) chains : dict The keys 'noise' and 'prior' contains the chain list of noise and