US2014363090A1PendingUtilityA1

Methods for Performing Fast Detail-Preserving Image Filtering

Assignee: FOVEON INCPriority: Feb 24, 2011Filed: Aug 22, 2014Published: Dec 11, 2014
Est. expiryFeb 24, 2031(~4.5 yrs left)· nominal 20-yr term from priority
Inventors:Rastislav Lukac
G06T 2207/20192G06T 5/20G06T 5/002G06T 2207/20012G06T 5/70
54
PatentIndex Score
0
Cited by
0
References
0
Claims

Abstract

A method for performing fast detail-preserving filtering of an input digital image includes, for each pixel in the image, calculating the difference between the selected pixel and each of its four neighboring pixels located above, left, right, and below the selected pixel, calculating a scaled weighted sum of differences between the actual pixel and its four neighbors, where for each neighboring pixel the weight is an edge-sensing function having a data-adaptive scaling parameter function, and adding the weighted sum of differences to the value of the selected pixel.

Claims

exact text as granted — not AI-modified
1 - 18 . (canceled) 
     
     
         19 . A method for creating an output digital image by performing fast detail-preserving filtering of an input digital image comprising:
 a) selecting a pixel a from the input digital image, pixel a not being in a row or column at the boundary of the image;   b) calculating the difference between the selected pixel and each of its four neighboring pixels located above, left, right, and below the selected pixel;   c) calculating a scaled weighted sum of differences between the selected pixel and its four neighbors, where for each neighboring pixel b the weight is an edge-sensing function having a data-adaptive scaling parameter function;   d) adding the weighted sum of differences to the value of the selected pixel; and   e) performing a) through d) for each pixel in the input digital image that is not in a row or column at the boundary of the image.   
     
     
         20 . The method of  claim 19  wherein:
 the weight is 
 
       
         
           
             
               
                 
                   f 
                    
                   
                     ( 
                     
                       a 
                       , 
                       b 
                     
                     ) 
                   
                 
                 = 
                 
                   1 
                   
                     1 
                     + 
                     
                       
                         
                           ( 
                           
                             a 
                             - 
                             b 
                           
                           ) 
                         
                         2 
                       
                       / 
                       
                         
                           κ 
                            
                           
                             ( 
                             
                               a 
                               , 
                               b 
                             
                             ) 
                           
                         
                         2 
                       
                     
                   
                 
               
               ; 
             
           
         
          and 
         κ(·) is the data-adaptive scaling function. 
       
     
     
         21 . The method of  claim 20  wherein κ(·) is a function using the pixel under consideration to estimate noise variance. 
     
     
         22 . The method of  claim 21  wherein κ(a,b)=θ(Θ 0 +Θ 1 a+Θ 2 a 2 ), where Θ 0 , Θ 1 a, and Θ 2 a 2  are quadratic model parameters. 
     
     
         23 . The method of  claim 20  wherein κ(·) is a function using the neighboring pixels to estimate noise variance. 
     
     
         24 . The method of  claim 23  wherein κ(a,b)=θ(Θ 0 +Θ 1 b+Θ 2 b 2 ), where θ is a scaling factor, and where Θ 0 , Θ 1 a, and Θ 2 a 2  are quadratic model parameters. 
     
     
         25 . The method of  claim 21  wherein κ(·) is a function using a combination of the pixel under consideration and the neighboring pixels to estimate noise variance. 
     
     
         26 . The method of  claim 25  wherein κ(a,b)=θ(Θ 0 +Θ 1 c+Θ 2 c 2 ), where θ is a scaling factor, and where Θ 0 , Θ 1 a, and Θ 2 a 2  are quadratic model parameters, where the term c=(w a a+w b b)/(w a +w b ) and where w a  and w b  are weights associated with pixels a and b, respectively. 
     
     
         27 . A method for creating an output color digital image by performing fast detail-preserving filtering of an input color digital image comprising:
 a) selecting a pixel a from the input color digital image, pixel a not being in a row or column at the boundary of the image;   b) calculating separately for each color channel in the selected pixel the difference between the selected pixel and each of its four neighboring pixels located above, left, right, and below the selected pixel;   c) calculating separately for each color channel in the selected pixel a scaled weighted sum of differences between the selected pixel and its four neighbors, where for each neighboring pixel b the weight is an edge-sensing function having a data-adaptive scaling parameter function;   d) adding separately for each color channel in the selected pixel the weighted sum of differences to the value of the selected pixel; and   e) performing a) through d) for each pixel in the input color digital image that is not in a row or column at the boundary of the image.   
     
     
         28 . The method of  claim 27  wherein:
 the weight is 
 
       
         
           
             
               
                 
                   f 
                    
                   
                     ( 
                     
                       a 
                       , 
                       b 
                     
                     ) 
                   
                 
                 = 
                 
                   1 
                   
                     1 
                     + 
                     
                       
                         
                           ( 
                           
                             a 
                             - 
                             b 
                           
                           ) 
                         
                         2 
                       
                       / 
                       
                         
                           κ 
                            
                           
                             ( 
                             
                               a 
                               , 
                               b 
                             
                             ) 
                           
                         
                         2 
                       
                     
                   
                 
               
               ; 
             
           
         
          and 
         κ(·) is the data-adaptive scaling function. 
       
     
     
         29 . The method of  claim 28  wherein κ(·) is a function using the pixel under consideration to estimate noise variance. 
     
     
         30 . The method of  claim 29  wherein κ(a,b)=θ(Θ 0 +Θ 1 a+Θ 2 a 2 ), where θ is a scaling factor, and where Θ 0 , Θ 1 a, and Θ 2 a 2  are quadratic model parameters 
     
     
         31 . The method of  claim 28  wherein κ(·) is a function using the neighboring pixels to estimate noise variance. 
     
     
         32 . The method of  claim 31  wherein κ(a,b)=θ(Θ 0 +Θ 1 b+Θ 2 b 2 ), where θ is a scaling factor, and where Θ 0 , Θ 1 a, and Θ 2 a 2  are quadratic model parameters. 
     
     
         33 . The method of  claim 28  wherein κ(·) is a function using a combination of the pixel under consideration and the neighboring pixels to estimate noise variance. 
     
     
         34 . The method of  claim 33  wherein:
 κ(a,b)=θ(Θ 0 +Θ 1 c+Θ 2 c 2 ) , where θ is a scaling factor, and where Θ 0 , Θ 1 a, and Θ 2 a 2  are quadratic model parameters, where the term c=(w a a+w b b)/(w a +w b ) and where w a  and w b  are weights associated with pixels a and b, respectively. 
 
     
     
         35 . The method of  claim 27  wherein:
 the weight 
 
       
         
           
             
               
                 
                   f 
                    
                   
                     ( 
                     
                       a 
                       , 
                       b 
                     
                     ) 
                   
                 
                 = 
                 
                   1 
                   
                     1 
                     + 
                     η 
                   
                 
               
               ; 
             
           
         
         η is one of the sum or average values of (a−b) 2 /κ(a,b) 2  values obtained for each color channel; and 
         κ(·) is the data-adaptive scaling function. 
       
     
     
         36 . The method of  claim 27  wherein:
 the weight 
 
       
         
           
             
               
                 
                   f 
                    
                   
                     ( 
                     
                       a 
                       , 
                       b 
                     
                     ) 
                   
                 
                 = 
                 
                   1 
                   
                     1 
                     + 
                     
                       
                         η 
                         1 
                       
                       / 
                       
                         η 
                         2 
                       
                     
                   
                 
               
               ; 
             
           
         
         η 1  is one of a distance or similarity measure of a−b values obtained for each color; 
         η 2  is one of the sum or average values of κ(a,b) 2  values obtained for each color; and 
         κ(·) is the data-adaptive scaling function.

Join the waitlist — get patent alerts

Track US2014363090A1 — get alerts on status changes and closely related new filings.

We store only your email — no account needed. See our privacy policy.