Image processing apparatus, image processing method, and storage medium

A method for an image processing apparatus, the method comprising:

• setting a target patch in an input image;
• detecting a plurality of similar patches which are similar to the target patch set in the input image;
• calculating a covariance matrix representing correlation between pixels based on the plurality of similar patches;
• calculating a plurality of eigenvalues and a plurality of eigenvectors of the covariance matrix;
• obtaining a noise amount in the input image;
• deriving a correction matrix based on the plurality of eigenvalues, the plurality of eigenvectors, the noise amount, and the number of similar patches, wherein the correction matrix is different from the calculated covariance matrix;
• correcting values of pixels in at least one of the plurality of similar patches based on the correction matrix;
• and
• generating an output image by combining the corrected similar patches.

A non-transitory computer readable storage medium storing a program to cause a computer to perform a method for an image processing apparatus, the method comprising:

• setting a target patch in an input image;
• detecting a plurality of similar patches which are similar to the target patch set in the input image;
• calculating a covariance matrix representing correlation between pixels based on the plurality of similar patches;
• calculating a plurality of eigenvalues and a plurality of eigenvectors of the covariance matrix;
• obtaining a noise amount in the input image;
• deriving a correction matrix based on the plurality of eigenvalues, the plurality of eigenvectors, the noise amount, and the number of similar patches, wherein the correction matrix is different from the calculated covariance matrix;
• correcting values of pixels in at least one of the plurality of similar patches based on the correction matrix;
• and
• generating an output image by combining the corrected similar patches.

An image processing apparatus comprising:

• a setting unit configured to set a target patch in an input image;
• a detection unit configured to detect a plurality of similar patches which are similar to the target patch set in the input image;
• a calculation unit configured to calculate a plurality of texture components based on the plurality of similar patches;
• a correction unit configured to detect of appearance degrees of individual texture components in the plurality of similar patches and to correct at least a portion of the variation of the appearance degrees;
• a processing unit configured to execute a noise reduction process on at least one of the plurality of similar patches based on the plurality of texture components and the at least the portion of the variation of the appearance degrees corrected by the correction unit;
• and
• a generation unit configured to generate an output image by combining the similar patches processed by the processing unit.

A method for an image processing apparatus, the method comprising:

• setting a target patch in an input image;
• detecting a plurality of similar patches which are similar to the target patch set in the input image;
• calculating a plurality of texture components based on the plurality of similar patches;
• detecting variation of appearance degrees of individual texture components in the plurality of similar patches and correcting at least a portion of the variation of the appearance degrees;
• executing a noise reduction process on at least one of the plurality of similar patches based on the plurality of texture components and the corrected at least the portion of the variation;
• and
• generating an output image by combining the processed similar patches.

A non-transitory computer readable storage medium storing a program to cause a computer to perform a method for an image processing apparatus, the method comprising:

• setting a target patch in an input image;
• detecting a plurality of similar patches which are similar to the target patch set in the input image;
• calculating a plurality of texture components based on the plurality of similar patches;
• detecting variation of appearance degrees of individual texture components in the plurality of similar patches and correcting at least a portion of the variation of the appearance degrees;
• executing a noise reduction process on at least one of the plurality of similar patches based on the plurality of texture components and the corrected at least the portion of the variation;
• and
• generating an output image by combining the processed similar patches.