Image Binarization

Abstract

Systems and methods convert to binary an input image having pixels defining text and background. Thresholds are determined by which pixels in the input image and a corresponding blurred image will be defined as either binary black or binary white. Thresholds derive from grouped together neighboring pixels having pixels separated out that correspond to the background. For pixels of the input image defined as binary black and having corresponding pixels in the blurred image defined as binary black relative to their thresholds, those are set to black in the binary image, else they are set white. Techniques for devising thresholds, blurring images, grouping together pixels, statistical analysis, etc., typify the embodiments.

1 . A method of converting to a binary image an input image having pluralities of pixels defining text and background each with an intensity value defined on a scale of multiple intensity values, comprising:
 blurring the input image to achieve a blurred image so that the pixels of the input image have corresponding blurred pluralities of pixels each with blurred intensity values defined on said scale of multiple intensity values;   for a pixel under consideration in either the input image or the blurred image, devising a neighborhood of adjacent pixels;   for the neighborhood of adjacent pixels, identifying a pixel having the intensity value or blurred intensity value with a greatest frequency of occurrence;   establishing a zone of background pixels nearby the pixel with the greatest frequency of occurrence; and   determining whether or not other pixels outside the zone of background pixels are sufficiently many in order to calculate a unique binarization threshold of said intensity value by which the pixel under consideration will later become identified as a black pixel or a white pixel in the binary image or are insufficiently many in order to apply a predetermined binarization threshold of said intensity value.   
     
     
         2 . The method of  claim 1 , further including advancing to a next pixel under consideration in either the input image or the blurred image and devising a next neighborhood of adjacent pixels. 
     
     
         3 . The method of  claim 2 , for the next neighborhood of adjacent pixels, identifying a next pixel having the intensity value or blurred intensity value with the greatest frequency of occurrence. 
     
     
         4 . The method of  claim 3 , establishing a next zone of background pixels nearby the next pixel with the greatest frequency of occurrence. 
     
     
         5 . The method of  claim 4 , further including determining whether or not other pixels outside the next zone of background pixels are sufficiently many in order to calculate a next unique binarization threshold of said intensity value by which the next pixel under consideration will become identified as said black pixel or said white pixel in the binary image or are insufficiently many in order to apply said predetermined binarization threshold of said intensity value. 
     
     
         6 . The method of  claim 2 , wherein the advancing to the next pixel under consideration in either the input image or the blurred image further includes advancing away three pixels in a same row or column of pixels from the pixel under consideration. 
     
     
         7 . The method of  claim 6 , wherein the devising the next neighborhood of adjacent pixels further includes establishing a square bounding box of pixels surrounding the next pixel under consideration. 
     
     
         8 . The method of  claim 1 , wherein said determining whether or not other pixels outside the zone of background pixels are sufficiently or insufficiently many includes determining whether or not more than four pixels exist outside the zone of background pixels. 
     
     
         9 . The method of  claim 1 , wherein the devising said neighborhood of adjacent pixels further includes creating a distribution of data in the form of a histogram noting frequency of occurrence versus pixel value intensity. 
     
     
         10 . The method of  claim 1 , wherein if the pixel under consideration in the input image and the blurred image corresponding thereto are defined as binary black relative to said unique binarization threshold or the predetermined binarization threshold, setting the pixel under consideration to black in the binary image, else setting the pixel under consideration to white in the binary image.