Adaptive dithering during image processing

Abstract

Systems and method are provided for adjusting certain pixel values in an image. In an embodiment of the invention, an average pixel value of pixels at given distances from a selected pixel are examined to determine if the pixel is in a high-contrast area. If the pixel is in a smooth color gradient transition area, the pixel value may be adjusted in some embodiments using an additional dither or dither pattern to reduce differences between the pixel values of the selected pixel and the additional average pixel values exceeding the lower threshold.

1 . An image processing method, comprising:
 filtering an input image at a plurality of filter sizes, and   modifying each pixel in the input image having a non-zero difference between the respective input pixel and a corresponding pixel of a filtered image at a first filter size that is less than a predetermined threshold, the modification based on the difference and a dither component.   
     
     
         2 . The method of  claim 1 , further comprising leaving the respective input pixel unchanged when the difference is either zero or greater than a predetermined threshold. 
     
     
         3 . The method of  claim 1 , wherein the modification is further based on a non-zero difference between the respective input pixel and a corresponding pixel of a filtered image at a second filter size that is less than a predetermined threshold. 
     
     
         4 . The method of  claim 1 , wherein the modification is further based on a plurality of non-zero differences between the respective input pixel and corresponding pixels of a filtered image at increasing filter sizes that are less than a predetermined threshold. 
     
     
         5 . An image processing system, comprising:
 a transition detector, having an input for input image data, to:
 filter image data at a plurality of filter sizes and 
 for each pixel in the input image data:
 compare the pixel of the input image data to a pixel of the filtered image data, and 
 when a difference between the respective input pixel and a filtered pixel at a first filter size is non-zero and less than a predetermined threshold, generate a control output for the input pixel having a value dependent on a difference between the input pixel and at least one filtered pixel; 
 
   a mixer to adjust pixels in a dither image pattern according to the control output of the transition detector, and   an adder, having a first input for the image data and a second input for an output of the mixer.   
     
     
         6 . The system of  claim 5 , wherein when the difference between the respective input pixel and a filtered pixel at a first filter size is either zero or greater than a predetermined threshold, the respective input pixel remains unchanged, and an output of the adder contains original input pixel data for the respective pixel. 
     
     
         7 . An image processing method comprising:
 calculating an average pixel value of pixels at a plurality of distances from a selected pixel of an image;   for each of the plurality of distances:
 if a difference between the calculated average value of pixels at the distance and a value of the selected pixel does not exceed a threshold, calculating a weighted pixel value for the distance, wherein the weighted pixel value is proportional to the difference, 
 otherwise, setting the weighted pixel value for the distance and a weighted pixel value for any greater distances to zero; 
   averaging the weighted pixel values; and   adjusting the selected pixel value based on the average and a dither when the average is non-zero.   
     
     
         8 . The computer-implemented method of  claim 7 , further comprising repeating the method for each pixel in the image. 
     
     
         9 . The computer-implemented method of  claim 7 , further comprising:
 retrieving a dither value for the selected pixel;   scaling the dither value proportionate to the average weighted pixel values; and   adjusting the selected pixel value based on the scaled dither value.   
     
     
         10 . The computer-implemented method of  claim 9 , wherein the dither value is retrieved from a value corresponding to the selected pixel in a dither pattern. 
     
     
         11 . The computer-implemented method of  claim 9 , wherein the selected pixel value is adjusted by either adding or subtracting the scaled dither value. 
     
     
         12 . The computer-implemented method of  claim 10 , wherein the dither pattern is selected based on at least one of: a size, a resolution, a position of the selected pixel, a color space, or a color model of the image. 
     
     
         13 . The computer-implemented method of  claim 10 , wherein the dither pattern is selected based on the direction of a color gradient change in the image. 
     
     
         14 . The computer-implemented method of  claim 10 , wherein the dither pattern is selected to increase a color palette of the image. 
     
     
         15 . The computer-implemented method of  claim 10 , wherein the dither pattern is a repeating pattern applying a dither to a first pixel in the image and not applying a dither to a second pixel in the image. 
     
     
         16 . An image processing method comprising:
 calculating an average pixel value of pixels at a given distance from a selected pixel;   when a difference between the calculated average pixel value and a value of the selected pixel is less than an upper threshold, modifying the value of the selected pixel to reduce the difference between the value of the selected pixel and the calculated average pixel value using a dither; and   otherwise, not modifying the value of the selected pixel.   
     
     
         17 . An image processor, comprising:
 an input interface to receive an image for processing;   a memory to store image processing data;   a processor coupled to the input/output interface and the memory, the processor performing the steps of:
 calculating an average pixel value of pixels at a plurality of distances from a selected pixel of an image; 
 for each of the plurality of distances:
 if a difference between the calculated average value of pixels at the distance and a value of the selected pixel does not exceed a threshold, calculating a weighted pixel value for the distance, wherein the weighted pixel value is proportional to the difference, 
 otherwise, setting the weighted pixel value for the distance and a weighted pixel value for any remaining greater distances to zero; 
 
 averaging the weighted pixel values; and 
 adjusting the selected pixel value based on the average and a dither; and 
   a output interface to output the adjusted selected pixel value.   
     
     
         18 . The device of  claim 17 , the processor performing the additional steps comprising:
 retrieving a dither value for the selected pixel;   scaling the dither value proportionate to the average weighted pixel values; and   adjusting the selected pixel value based on the scaled dither value.   
     
     
         19 . The device of  claim 17 , wherein the dither value is retrieved from a value corresponding to the selected pixel in a dither pattern. 
     
     
         20 . The device of  claim 17 , wherein the selected pixel value is adjusted by either adding or subtracting the scaled dither value. 
     
     
         21 . The device of  claim 17 , wherein the device further comprises a computing device, the input interface, memory, processor, and output interface incorporated in the computing device. 
     
     
         22 . The device of  claim 21 , wherein the computing device is portable. 
     
     
         23 . The device of  claim 17 , wherein the device further comprises a display device, the input interface, memory, processor, and output interface incorporated in the display device. 
     
     
         24 . The device of  claim 23 , wherein the display device is a monitor. 
     
     
         25 . The device of  claim 17 , wherein the device further comprises an electronics device, the input interface, memory, processor, and output interface incorporated in the electronics device. 
     
     
         26 . The device of  claim 25 , wherein the electronics device is a digital video player. 
     
     
         27 . An article of manufacture including a computer readable medium storing instructions that, when executed by a processor, cause the processor to:
 filter an input image at a plurality of filter sizes, and   modify each pixel in the input image having a non-zero difference between the respective input pixel and a corresponding pixel of a filtered image at a first filter size that is less than a predetermined threshold, the modification based on the difference and a dither component.   
     
     
         28 . An article of manufacture including a computer readable medium storing instructions that, when executed by a processor, cause the processor to:
 calculate an average pixel value of pixels at a given distance from a selected pixel;   when a difference between the calculated average pixel value and a value of the selected pixel is less than an upper threshold, modify the value of the selected pixel to reduce the difference between the value of the selected pixel and the calculated average pixel value using a dither; and otherwise, leave the value of the selected pixel unchanged.   
     
     
         29 . An article of manufacture including a computer readable medium storing instructions that, when executed by a processor, cause the processor to:
 calculate an average pixel value of pixels at a plurality of distances from a selected pixel of an image;   for each of the plurality of distances:
 if a difference between the calculated average value of pixels at the distance and a value of the selected pixel does not exceed a threshold, calculate a weighted pixel value for the distance, wherein the weighted pixel value is proportional to the difference, 
 otherwise, set the weighted pixel value for the distance and a weighted pixel value for any greater distances to zero; 
   average the weighted pixel values; and   adjust the selected pixel value based on the average and a dither when the average is non-zero.