US 2011/0234823 A1Published application

Image processing apparatus and method of controlling the same

AssigneeCanon Kabushiki Kaisha

InventorKen Terasawa

Life of this patent

Estimating the term…

Loading patent…

Abstract

The correction amount of an aberration caused in each pixel of the region by the optical system is calculated. A predetermined number of peripheral pixels center on a position apart from the pixel by the distance corresponding to the calculated aberration correction amount are multiplied by interpolation coefficients obtained from an interpolation function and added, thereby deriving the pixel value at the pixel position after correction. If the predetermined number of peripheral pixels around the position apart by the distance corresponding to the calculated aberration correction amount are not present in the readout region, aberration correction is implemented by changing the interpolation function. More specifically, the interpolation function is changed so as to make the frequency response behavior of the interpolation function more moderate than that of the interpolation function that gives the interpolation coefficients to multiply the predetermined number of peripheral pixels and used in the correction processing.

Claims7

Dependent claims

An image processing apparatus arranged to correct, for each pixel of an image, for the influence of aberrations in an optical system in the image captured using the optical system, comprising:

storage unit configured to read out and store an image of a partial region of the image;
calculation unit configured to calculate a distance, on the image of the partial region, between a pixel of interest of the image of the partial region stored in said storage unit and an image which should be formed at a position of the pixel of interest but has moved due to the aberrations of the optical system, as a correction amount using information of an image height of the pixel of interest and information of the optical system;
and
correction unit configured to derive a pixel value of the pixel of interest after correction by adding pixel values obtained by multiplying a pixel value of each of a predetermined number of peripheral pixels, center on a position apart from the pixel of interest by the distance calculated by said calculation unit, by an interpolation coefficient of each of the peripheral pixels obtained from a predetermined interpolation function that is a function of the distance from the position apart from the pixel of interest by the distance calculated by said calculation unit;
wherein if the peripheral pixels are stored in said storage unit only in a number smaller than the predetermined number, said correction unit changes the interpolation function so as to make a frequency response behavior of the predetermined interpolation function more moderate than that when the predetermined number of peripheral pixels are stored in said storage unit, and derives the pixel value of the pixel of interest after correction by adding pixel values obtained by multiplying a pixel value of each of the peripheral pixels in the number smaller than the predetermined number by the interpolation coefficient of each of the peripheral pixels obtained from the interpolation function.

2 more elements

A method of controlling an image processing apparatus arranged to correct, for each pixel of an image, for the influence of aberrations in an optical system in the image captured using the optical system, comprising the steps of:

reading out and storing an image of a partial region of the image in storage unit;
calculating a distance, on the image of the partial region, between a pixel of interest of the image of the partial region stored in the storage unit and an image which should be formed at a position of the pixel of interest but has moved due to the aberrations of the optical system, as a correction amount using information of an image height of the pixel of interest and information of the optical system;
and
deriving a pixel value of the pixel of interest after correction by adding pixel values obtained by multiplying a pixel value of each of a predetermined number of peripheral pixels, center on a position apart from the pixel of interest by the distance calculated in the step of calculating the distance, by an interpolation coefficient of each of the peripheral pixels obtained from a predetermined interpolation function that is a function of the distance from the position apart from the pixel of interest by the distance calculated at the step of calculating the distance;
wherein in the step of deriving the pixel value, if the peripheral pixels are stored in the storage unit only in a number smaller than the predetermined number, the interpolation function is changed so as to make a frequency response behavior of the predetermined interpolation function more moderate than that when the predetermined number of peripheral pixels are stored in the storage unit, and the pixel value of the pixel of interest after correction is derived by adding pixel values obtained by multiplying a pixel value of each of the peripheral pixels in the number smaller than the predetermined number by the interpolation coefficient of each of the peripheral pixels obtained from the interpolation function.

2 more elements

Dependencies are inferred from the claim text.

Facts

Application: 13/045,053Filed 10 Mar 2011
Priority: JP 2010-07620129 Mar 2010
Published: 29 Sep 2011
Term ends: Not estimatedThe source records don’t support an estimate.

Classification7 codes

G06T 5/80
Image enhancement or restoration: geometric correction
H04N 23/843
For processing colour signals: demosaicing, e.g. interpolating colour pixel values
H04N 25/611
The noise originating only from the lens unit, e.g. flare, shading, vignetting or "cos4": correction of chromatic aberration
H04N 25/615
The noise originating only from the lens unit, e.g. flare, shading, vignetting or "cos4": involving a transfer function modelling the optical…