Image processing for binarization enhancement via fuzzy reasoning
First Claim
1. A computer-based method for converting a gray-scale multiple pixel image to a binary image, said method comprising the steps of:
- a) loading a gray-scale image to be converted to a binary image into an operating memory of a computer, said gray-scale image being formed of a plurality of pixels, each pixel having a value that defines a relative brightness level of said pixel;
b) enhancing said gray-scale image by executing an application with said computer that adjusts the brightness level of pixels in said image, said application carrying out the steps of;
1) determining a difference value by which a value of a selected pixel in said image differs in brightness from an average brightness value of a plurality of pixels surrounding said selected pixel;
2) employing said difference value as input to a plurality of overlapping triangular fuzzy set type membership input functions to determine a corresponding plurality of membership values for said selected pixel;
3) applying said plurality of membership values as inputs to a corresponding plurality of single valued singleton type fuzzy membership output functions, each of which generates a single valued output;
4) generating a crisp output value from said single valued outputs from said membership output functions that identifies a relative amount by which said selected pixel'"'"'s brightness should be adjusted5) adjusting the brightness of said selected pixel by adding said crisp output value to said selected pixel value to obtain an adjusted pixel value for said selected pixel;
6) repeating steps (1)-(5) for additional pixels in said image, thereby generating an enhanced gray-scale image; and
c) converting said enhanced gray-scale image to a binary image.
1 Assignment
0 Petitions
Accused Products
Abstract
A technique for enhancing a gray-scale image to improve conversions of the image to binary employs fuzzy reasoning. In the technique, pixels in the image are analyzed by comparing the pixel'"'"'s gray scale value, which is indicative of its relative brightness, to the values of pixels immediately surrounding the selected pixel. The degree to which each pixel in the image differs in value from the values of surrounding pixels is employed as the variable in a fuzzy reasoning-based analysis that determines an appropriate amount by which the selected pixel'"'"'s value should be adjusted to reduce vagueness and ambiguity in the image and improve retention of information during binarization of the enhanced gray-scale image.
38 Citations
20 Claims
-
1. A computer-based method for converting a gray-scale multiple pixel image to a binary image, said method comprising the steps of:
-
a) loading a gray-scale image to be converted to a binary image into an operating memory of a computer, said gray-scale image being formed of a plurality of pixels, each pixel having a value that defines a relative brightness level of said pixel; b) enhancing said gray-scale image by executing an application with said computer that adjusts the brightness level of pixels in said image, said application carrying out the steps of; 1) determining a difference value by which a value of a selected pixel in said image differs in brightness from an average brightness value of a plurality of pixels surrounding said selected pixel; 2) employing said difference value as input to a plurality of overlapping triangular fuzzy set type membership input functions to determine a corresponding plurality of membership values for said selected pixel; 3) applying said plurality of membership values as inputs to a corresponding plurality of single valued singleton type fuzzy membership output functions, each of which generates a single valued output; 4) generating a crisp output value from said single valued outputs from said membership output functions that identifies a relative amount by which said selected pixel'"'"'s brightness should be adjusted 5) adjusting the brightness of said selected pixel by adding said crisp output value to said selected pixel value to obtain an adjusted pixel value for said selected pixel; 6) repeating steps (1)-(5) for additional pixels in said image, thereby generating an enhanced gray-scale image; and c) converting said enhanced gray-scale image to a binary image. - View Dependent Claims (2, 3, 4, 5, 6, 7)
-
-
8. A computer-based method for converting a gray-scale multiple pixel image to a binary image, said method comprising the steps of:
-
a) loading a gray-scale image to be converted to a binary image into an operating memory of a computer, said gray-scale image being formed of a plurality of pixels, each pixel having a value that defines a relative brightness level of said pixel; b) enhancing said gray-scale image by executing an application with said computer that adjusts the brightness level of pixels in said image, said application carrying out the steps of; 1) determining a difference value by which a value of a selected pixel in said image differs in brightness from an average brightness value of a plurality of pixels surrounding said selected pixel wherein an n×
n pixel window is employed to determine the average value of pixels surrounding said selected pixel, with the selected pixel located at the center of the window and n being an odd number of 3 or greater;2) employing said difference value as input to a plurality of overlapping triangular fuzzy set type membership input functions to determine a corresponding plurality of membership values for said selected pixel; 3) applying said plurality of membership values as inputs to a corresponding plurality of single valued singleton type fuzzy membership output functions, each of which generates a single valued output; 4) generating a crisp output value from said single valued outputs from said membership output functions that identifies a relative amount by which said selected pixel'"'"'s brightness should be adjusted; 5) adjusting the brightness of said selected pixel by adding said crisp output value to said selected pixel value to obtain an adjusted pixel value for said selected pixel; 6) repeating steps (1)-(5) for additional pixels in said image, thereby generating an enhanced gray-scale image; and c) converting said enhanced gray-scale image to a binary image. - View Dependent Claims (9, 10, 11, 12, 13)
-
-
14. A computer system for converting for converting a gray-scale multiple pixel image to a binary image comprising:
-
a) a processor; b) an operating memory; c) a source of multiple pixel gray-scale digital images; and d) a program for performing a conversion of said gray-scale multiple pixel images into binary images, said program including a method for enhancing a gray-scale image before converting said enhanced image to binary, said method comprising the steps of; 1) loading a gray-scale image to be converted to a binary image into said operating memory, said gray-scale image being formed by a plurality of pixels, each pixel having a value that defines a relative brightness level of said pixel; 2) enhancing said gray-scale image by executing an application with said processor that adjusts the brightness level of pixels in said image, said application carrying out the steps of; i) determining a difference value by which a value of a selected pixel in said image differs in brightness from an average brightness value of a plurality of pixels surrounding said selected pixel; ii) employing said difference value as input to a plurality of overlapping triangular fuzzy set type membership input functions to determine a corresponding plurality of membership values for said selected pixel; iii) applying said plurality of membership values as inputs to a corresponding plurality of single valued singleton type fuzzy membership output functions, each of which generates a single valued output; iv) generating a crisp output value from said single valued outputs from said membership output functions that identifies a relative amount by which said selected pixel'"'"'s brightness should be adjusted; v) adjusting the brightness of said selected pixel by adding said crisp output value to said selected pixel value to obtain an adjusted pixel value for said selected pixel; and vi) repeating steps (i)-(v) for additional pixels in said image, thereby generating an enhanced gray-scale image; and 3) converting said enhanced gray-scale image to a binary image. - View Dependent Claims (15, 16, 17, 18, 19, 20)
-
Specification