Computationally efficient method for image segmentation with intensity and texture discrimination

Abstract

An automatic method for depicting texture in a digital image is provided. A category is automatically assigned to each pixel in the digital image based on one of a set of brightness values associated with the pixel. Each of the brightness values corresponds to a range of intensities associated with the pixels in the digital image, and one of a set of texture values associated with the pixel. Each of the texture values corresponds to a range of differences in the value between the intensity of the pixel and the intensity of neighbors of the pixel. A shading level is automatically assigned to each of the pixels based on the categories. A processed image is automatically produced based on the shading levels, the processed image indicating the texture in the image.

What is claimed is: 
     
         1 . An automatic method for depicting texture in a digital image comprising:
 automatically assigning a category to each pixel in the digital image based on one of a set of brightness values, associated with the pixel, each, of the brightness values corresponding to a range of intensities associated with the pixels in the digital image, and one of a set of texture values associated with the pixel, each of the texture values corresponding to a range of differences in the value between the intensity of the pixel and the intensity of neighbors of the pixel;   automatically assigning a shading level to each of the pixels based on the categories;   automatically producing a processed image based on the shading levels, the processed image indicating the texture in the image.   
     
     
         2 . The method as in  claim 1  wherein the brightness value comprises dark, medium, and bright. 
     
     
         3 . The method as in  claim 1  wherein the texture value comprises a first texture, a second texture, and a third texture. 
     
     
         4 . The method as in  claim 1  further comprising;
 dividing the image into a plurality of sections; 
 processing each of the plurality of sections separately; and 
 combining the processed sections to produce the processed image. 
 
     
     
         5 . The method, as in  claim 4  further comprising:
 processing the plurality of sections in parallel. 
 
     
     
         6 . The method as in  claim 1  further comprising:
 despeckeling the processed image. 
 
     
     
         7 . The method as in  claim 1  comprising:
 despeckeling the processed image using a smoothing filter.