Image processing apparatus, image processing method, and storage medium

Abstract

An image processing apparatus to execute noise reduction processing on image data includes a setting unit, a determination unit, and an output unit. The setting unit sets a pixel group from among a plurality of pixel group candidates. The plurality of pixel group candidates includes at least a first pixel group having a plurality of pixels being a first number of pixels or a second pixel group having a plurality of pixels being a second number of pixels which is different from the first number of pixels. The determination unit determines, based on a similarity between a target pixel and a reference pixel that is obtained according to the set pixel group, a weight corresponding to the reference pixel. The output unit outputs a value, calculated based on a pixel value of the reference pixel and the weight, as a noise-reduced pixel value of the target pixel.

What is claimed is: 
     
         1 . An image processing apparatus configured to execute noise reduction processing on image data, the image processing apparatus comprising:
 a setting unit configured to set a pixel group from among a plurality of pixel group candidates, wherein the plurality of pixel group candidates includes at least a first pixel group having a plurality of pixels being a first number of pixels or a second pixel group having a plurality of pixels being a second number of pixels which is different from the first number of pixels;   a determination unit configured to determine, based on a similarity between a target pixel and a reference pixel that is obtained according to the pixel group set by the setting unit, a weight corresponding to the reference pixel; and   an output unit configured to output a value, calculated based on a pixel value of the reference pixel and the weight, as a noise-reduced pixel value of the target pixel.   
     
     
         2 . The image processing apparatus according to  claim 1 , wherein the setting unit is configured to acquire information related to noise in the image data and to set a pixel group based on the acquired information related to noise. 
     
     
         3 . The image processing apparatus according to  claim 1 , wherein the determination unit is configured to obtain a similarity corresponding to the reference pixel based on a target area, wherein the target area includes the target pixel and a reference area including the reference pixel based on the set pixel group. 
     
     
         4 . The image processing apparatus according to  claim 1 , wherein the determination unit is configured to set a plurality of pixels in the image data as a plurality of reference pixels and determine a plurality of weights respectively corresponding to the plurality of reference pixels. 
     
     
         5 . The image processing apparatus according to  claim 1 , wherein the setting unit is configured to set a pixel group for each piece of the image data. 
     
     
         6 . The image processing apparatus according to  claim 1 , wherein the setting unit is configured to set a pixel group for each pixel in the image data. 
     
     
         7 . The image processing apparatus according to  claim 1 , wherein, in a first case where information related to noise indicates that a noise variance in the image data is larger than a predetermined value, the setting unit is configured to set the set pixel group to include a number of pixels that is larger than a number of pixels configured to be set by the setting unit to the set pixel group in a second case where the information related to noise indicates that a noise variance in the image data is smaller than in the first case. 
     
     
         8 . The image processing apparatus according to  claim 1 ,
 wherein the plurality of pixel group candidates further includes a third pixel group having a plurality of pixels being a third number of pixels which is different from the first number of pixels and the second number of pixels, and   wherein the setting unit is configured to set, as the set pixel group, one pixel group from among the first pixel group, the second pixel group, and the third pixel group.   
     
     
         9 . An image processing apparatus comprising:
 a first noise reduction processing unit configured to obtain, for a target pixel and a reference pixel in image data, a first noise-reduced output value of the target pixel by performing weighting based on a similarity of the reference pixel to the target pixel, wherein the similarity of the reference pixel to the target pixel is obtained based on a target area including the target pixel and a reference area including the reference pixel that correspond to a first pixel group;   a second noise reduction processing unit configured to obtain a second noise-reduced output value of the target pixel by performing weighting based on a similarity of the reference pixel to the target pixel, wherein the similarity of the reference pixel to the target pixel is obtained based on a target area including the target pixel and a reference area including the reference pixel that correspond to a second pixel group, wherein the second pixel group includes a number of pixels that is different from a number of pixels in the first pixel group; and   a selection unit configured to select, based on information related to noise in the image data, either one of the first noise-reduced output value and the second noise-reduced output value as a noise-reduced output value of the target pixel.   
     
     
         10 . The image processing apparatus according to  claim 9 ,
 wherein the number of pixels included in the first pixel group is larger than the number of pixels included in the second pixel group, and   wherein, in a case where the information related to noise indicates that noise in the image data is larger than a predetermined value, the selection unit is configured to select the first noise-reduced output value.   
     
     
         11 . The image processing apparatus according to  claim 9 , wherein the information related to noise in the image data is information indicating an International Organization for Standardization (ISO) sensitivity of when the image data is obtained. 
     
     
         12 . The image processing apparatus according to  claim 9 , wherein the information related to noise in the image data is information indicating an exposure time of when the image data is obtained. 
     
     
         13 . The image processing apparatus according to  claim 9 , wherein the information related to noise in the image data is information indicating a temperature of when the image data is obtained. 
     
     
         14 . The image processing apparatus according to  claim 9 , wherein the information related to noise in the image data is a pixel value of the target pixel. 
     
     
         15 . An image processing apparatus comprising:
 an acquisition unit configured to acquire image data; and   a processing unit configured to execute, for each pixel in the acquired image data, noise reduction processing on the image data using non-local means (NLM) of calculating a similarity based on a pixel group including a number of pixels which is set based on information related to noise in the image data.   
     
     
         16 . An imaging apparatus comprising the image processing apparatus according to  claim 1 . 
     
     
         17 . An image processing method for an image processing apparatus configured to execute noise reduction processing on image data, the image processing method comprising:
 setting a pixel group from among a plurality of pixel group candidates, wherein the plurality of pixel group candidates includes at least a first pixel group having a plurality of pixels being a first number of pixels or a second pixel group having a plurality of pixels being a second number of pixels which is different from the first number of pixels;   determining, based on a similarity between a target pixel and a reference pixel that is obtained according to the set pixel group, a weight corresponding to the reference pixel; and   outputting a value, calculated based on a pixel value of the reference pixel and the weight, as a noise-reduced pixel value of the target pixel.   
     
     
         18 . A computer-readable storage medium storing a program causing an image processing apparatus, configured to execute noise reduction processing on image data, to perform an image processing method according to  claim 17 . 
     
     
         19 . An image processing method for an image processing apparatus, the image processing method comprising:
 obtaining, via a first noise reduction processing and for a target pixel and a reference pixel in image data, a first noise-reduced output value of the target pixel by performing weighting based on a similarity of the reference pixel to the target pixel, wherein the similarity of the reference pixel to the target pixel is obtained based on a target area including the target pixel and a reference area including the reference pixel that correspond to a first pixel group;   obtaining, via a second noise reduction processing, a second noise-reduced output value of the target pixel by performing weighting based on a similarity of the reference pixel to the target pixel, wherein the similarity of the reference pixel to the target pixel is obtained based on a target area including the target pixel and a reference area including the reference pixel that correspond to a second pixel group, wherein the second pixel group includes a number of pixels that is different from a number of pixels in the first pixel group; and   selecting, based on information related to noise in the image data, either one of the first noise-reduced output value and the second noise-reduced output value as a noise-reduced output value of the target pixel.   
     
     
         20 . An image processing method for an image processing apparatus, the image processing method comprising:
 acquiring image data; and   executing, for each pixel in the acquired image data, noise reduction processing on the image data using non-local means (NLM) of calculating a similarity based on a pixel group including a number of pixels which is set based on information related to noise in the image data.