Method for determining noise level

Abstract

The present invention relates to a method for determining noise levels in a subband of an image. The method comprises receiving the subband of the image, defining block regions in the at least two space domains of the subband, for each defined block region, identifying first wavelet coefficients associated with coordinate values in the at least two space domains in the defined block region, computing a correlation matrix between identified wavelet coefficients to determine the correlation between first wavelet coefficients according to the at least one color domain, computing second wavelet coefficients, the computation of second wavelet coefficients being based on the correlation matrix and the first wavelet coefficients, computing at least one noise level, the noise level computation being based on at least one second wavelet coefficient and providing the at least one noise level.

What is claimed is:  
     
         1 - 10 . (canceled) 
     
     
         11 . A method for determining a noise level in a sub-band of an image, the method comprising:
 receiving the sub-band of the image, the image being expressed by a color domain comprising a plurality of color channels;   dividing the sub-band spatially into a plurality of blocks;   determining a first wavelet coefficient for each of the blocks;   computing a correlation matrix between each of the color channels for each of the blocks;   computing a plurality of second wavelet coefficients based on the correlation matrix and the first wavelet coefficients for each of the blocks; and   determining the noise level based on the second wavelet coefficients.   
     
     
         12 . The method of  claim 11 , wherein each block comprises a plurality of points and wherein each value of the correlation matrix represents a correlation between two color channels for a corresponding point of the block. 
     
     
         13 . The method of  claim 11 , wherein the computing the second wavelet coefficients comprises:
 determining an orthogonal transformation matrix based on the correlation matrix; and   generating the second wavelet coefficients by applying the orthogonal transformation matrix to the first wavelet coefficients.   
     
     
         14 . The method of  claim 13 , wherein the orthogonal transformation matrix is selected such that the noise in the second wavelet coefficients is substantially uncorrelated. 
     
     
         15 . The method of  claim 11 , wherein each of the blocks at least partially overlaps an adjacent one of the blocks. 
     
     
         16 . The method of  claim 11 , wherein the determining the noise level determining the minimum median of the second wavelet coefficients. 
     
     
         17 . The method of  claim 11 , wherein the computing the second wavelet coefficients comprises transforming the color domain into a new color domain and wherein the noise in each color channel of the new color domain is substantially uncorrelated. 
     
     
         18 . A device for determining a noise level in a sub-band of an image, the device comprising:
 an input interface configured to receive the sub-band of the image, the image being expressed by a color domain comprising a plurality of color channels;   a memory configured to store the sub-band; and   a processor operatively coupled to the memory and the input interface and configured to:
 divide the sub-band spatially into a plurality of blocks; 
 determine a first wavelet coefficient for each of the blocks; 
 compute a correlation matrix between each of the color channels for each of the blocks; 
 compute a plurality of second wavelet coefficients based on the correlation matrix and the first wavelet coefficients for each of the blocks; and 
 determine the noise level based on the second wavelet coefficients. 
   
     
     
         19 . The device of  claim 18 , wherein each block comprises a plurality of points and wherein each value of the correlation matrix represents a correlation between two color channels for a corresponding point of the block. 
     
     
         20 . The device of  claim 18 , wherein the processor is further configured to compute the second wavelet coefficients based on:
 determining an orthogonal transformation matrix based on the correlation matrix; and   generating the second wavelet coefficients by applying the orthogonal transformation matrix to the first wavelet coefficients.   
     
     
         21 . The device of  claim 20 , wherein the processor is further configured to select orthogonal transformation matrix such that the noise in the second wavelet coefficients is substantially uncorrelated. 
     
     
         22 . The device of  claim 18 , wherein each of the blocks at least partially overlaps an adjacent one of the blocks. 
     
     
         23 . The device of  claim 18 , wherein the processor is further configured to determine the noise level based on determining the minimum median of the second wavelet coefficients. 
     
     
         24 . The device of  claim 18 , wherein the processor is further configured to compute the second wavelet coefficients based on transforming the color domain into a new color domain and wherein the noise in each channel of the new color domain is substantially uncorrelated. 
     
     
         25 . An apparatus, comprising:
 means for receiving a sub-band of an image, the image being expressed by a color domain comprising a plurality of color channels;   means for dividing the sub-band spatially into a plurality of blocks;   means for determining a first wavelet coefficient for each of the blocks;   means for computing a correlation matrix between each of the color channels for each of the blocks;   means for computing a plurality of second wavelet coefficients based on the correlation matrix and the first wavelet coefficients for each of the blocks; and   means for determining the noise level based on the second wavelet coefficients.   
     
     
         26 . The apparatus of  claim 25 , wherein the means for computing the second wavelet coefficients comprises:
 means for determining an orthogonal transformation matrix based on the correlation matrix; and   means for generating the second wavelet coefficients by applying the orthogonal transformation matrix to the first wavelet coefficients.   
     
     
         27 . The apparatus of  claim 26 , wherein the orthogonal transformation matrix is selected such that the noise in the second wavelet coefficients is substantially uncorrelated. 
     
     
         28 . A non-transitory computer readable medium comprising code that, when executed, causes an apparatus to perform a process comprising:
 receive a sub-band of an image, the image being expressed by a color domain comprising a plurality of color channels;   divide the sub-band spatially into a plurality of blocks;   determine a first wavelet coefficient for each of the blocks;   compute a correlation matrix between each of the color channels for each of the blocks;   compute a plurality of second wavelet coefficients based on the correlation matrix and the first wavelet coefficients for each of the blocks; and   determine the noise level based on the second wavelet coefficients.   
     
     
         29 . The non-transitory computer readable medium of  claim 28 , further comprising code that, when executed, causes the apparatus to compute the second wavelet coefficients based on:
 determining an orthogonal transformation matrix based on the correlation matrix; and   generating the second wavelet coefficients by applying the orthogonal transformation matrix to the first wavelet coefficients.   
     
     
         30 . The non-transitory computer readable medium of  claim 29 , further comprising code that, when executed, causes the apparatus to select the orthogonal transformation matrix such that the noise in the second wavelet coefficients is substantially uncorrelated.