Image enhancing method using mean-separate histogram equalization and a circuit therefor
First Claim
1. A method for enhancing an image by histogram-equalizing an input image signal represented by a predetermined number of gray levels, said method comprising:
- calculating a mean level of the input image;
grouping a plurality of pixels of said input image signal into a plurality of groups of image in accordance with the gray level of each of said pixels, wherein division of the input image signal into a plurality of groups of image occurs in accordance with the mean level;
calculating a cumulative density function for each group of image; and
histogram-equalizing each group of image independent of the other groups of image by mapping an input sample of the group of image into a new gray level based on the cumulative density functions;
wherein the number of groups of images per each frame of the input image signal is predetermined to be two; and
wherein said histogram-equalizing comprises;
mapping the input sample into a first gray level by using a first transform function, wherein the first transform function is based on one of the cumulative density functions and the first gray level is in a first range;
mapping the input sample into a second gray level by using a second transform function, wherein the second transform function is based on another of the cumulative density functions and the second gray level is in a second range;
comparing the input sample with the mean level; and
selecting one of said first and second gray levels according to the comparison of he input sample with the mean level.
1 Assignment
0 Petitions
Accused Products
Abstract
An image enhancing method using a mean-separate histogram equalization method is disclosed. An input image signal is divided into two sub-images in a picture unit according to a mean level of the input image signal. Then, cumulative density functions for each of the sub-images are calculated. Afterwards, the input sample is mapped into two gray levels, each of which belongs to a first range and a second range, by using first and second transform functions defined by use of the cumulative density functions, respectively. Finally, one of two mapped levels is selected depending on the magnitude of the input sample. As a result, the brightness of the given image can be preserved while the contrast is enhanced.
-
Citations
45 Claims
-
1. A method for enhancing an image by histogram-equalizing an input image signal represented by a predetermined number of gray levels, said method comprising:
-
calculating a mean level of the input image; grouping a plurality of pixels of said input image signal into a plurality of groups of image in accordance with the gray level of each of said pixels, wherein division of the input image signal into a plurality of groups of image occurs in accordance with the mean level; calculating a cumulative density function for each group of image; and histogram-equalizing each group of image independent of the other groups of image by mapping an input sample of the group of image into a new gray level based on the cumulative density functions; wherein the number of groups of images per each frame of the input image signal is predetermined to be two; and wherein said histogram-equalizing comprises; mapping the input sample into a first gray level by using a first transform function, wherein the first transform function is based on one of the cumulative density functions and the first gray level is in a first range; mapping the input sample into a second gray level by using a second transform function, wherein the second transform function is based on another of the cumulative density functions and the second gray level is in a second range; comparing the input sample with the mean level; and selecting one of said first and second gray levels according to the comparison of he input sample with the mean level.
-
-
2. A method for enhancing an image by histogram-equalizing an input image signal represented by a predetermined number of gray levels, said method comprising:
-
calculating a mean level in a picture unit of the input image signal; obtaining a gray level distribution in the picture unit of the input image signal; dividing the gray level distribution into a predetermined number of sub-images; calculating a cumulative density function for each sub-image based on the gray level distribution; and histogram-equalizing each sub-image independent of one another by mapping a sample of the sub-image into a new gray level based on the cumulative density functions. - View Dependent Claims (3, 4, 5)
-
-
6. A method for enhancing an image by histogram-equalizing an input image signal represented by a predetermined number of grays levels, said method comprising:
-
calculating a mean level in a picture unit of the input image signal; obtaining a gray level distribution in the picture unit of the input image signal; dividing the gray level distribution into a predetermined number of sub-images; calculating a cumulative density function for each sub-image based on the gray level distribution; and obtaining a compensated mean level by adding to said mean level a brightness compensation value determined according to a predetermined correction function based on a mean brightness of the input image signal; and histogram-equalizing each sub-image independent of one another by mapping an input sample of the sub-image into a new gray level based on the cumulative density function and the compensated mean level. - View Dependent Claims (7, 8, 9)
-
-
10. A circuit for enhancing an image by histogram-equalizing an image signal represented by a predetermined number of gray levels in a picture unit, said circuit comprising:
-
first calculating means for calculating a gray level distribution of the image signal; second calculating means for calculating a mean level of the image signal; third calculating means for dividing the gray level distribution into a predetermined number of sub-images based on the mean level and calculating a cumulative density function for each sub-image; a picture memory for storing samples of the image signal and providing an input sample delayed by a picture unit; mapping means for mapping the input sample output by said picture memory into mapped gray levels based on the sub-images and the cumulative density functions, and outputting the mapped gray levels; and outputting means for comparing the input sample with the mean level, selecting one of the mapped gray levels output by said mapping means and outputting the selected level. - View Dependent Claims (11, 12, 13, 14)
-
-
15. A circuit for enhancing an image by histogram-equalizing an image signal represented by a predetermined number of gray levels in a picture unit, said circuit comprising:
-
first calculating means for calculating a gray level distribution of the image signal; second calculating means for calculating a mean level of the image signal; third calculating means for dividing the gray level distribution into a predetermined number of sub-images based on the mean level and calculating a cumulative density function for each sub-image; mapping means for mapping an input sample of the image signal into mapped gray levels based on the sub-images and the cumulative density functions calculated for a picture just before a current picture, and outputting the mapped gray levels; and outputting means for comparing the input sample with the mean level, selecting one of the mapped gray levels output by said mapping means and outputting the selected gray level. - View Dependent Claims (16, 17, 18, 19)
-
-
20. A circuit for enhancing an image by histogram-equalizing an image signal represented by a predetermined number of gray levels in a picture unit, said circuit comprising:
-
first calculating means for calculating a gray level distribution of the image signal; second calculating means for calculating a mean level of the image signal; compensation means for calculating a compensated mean level by adding a brightness compensation value according to a predetermined correction function to the mean level based on the mean brightness of the input image; third calculating means for dividing the gray level distribution into a predetermined number of sub-images on the mean level and calculating a cumulative density function for each sub-image; a picture memory for delaying the image signal by a picture unit and providing an input sample; mapping means for mapping the input sample output by said picture memory into mapped gray levels based on the sub-images, the cumulative density functions and the compensated mean level, and outputting the mapped gray levels; and outputting means for comparing the input sample with the mean level and selecting one of the mapped gray levels output by said mapping means and outputting the selected level. - View Dependent Claims (21, 22, 23, 24, 25)
-
-
26. A circuit for enhancing an image by histogram-equalizing an image signal represented by a predetermined number of gray levels in a picture unit, said circuit comprising:
-
first calculating means for calculating a gray level distribution of the image signal; second calculating means for calculating a mean level of the image signal; compensation means for calculating a compensated mean level by adding a brightness compensation value according to a predetermined correction function to the mean level based on the mean brightness of the input image; third calculating means for dividing the gray level distribution into a predetermined number of sub-images based on the mean level and calculating a cumulative density function for each sub-image; mapping means for mapping an input sample of the image signal into mapped gray levels based on the sub-images, the cumulative density functions and the compensated mean level, and outputting the mapped gray levels; and outputting means for comparing the input sample with the mean level and selecting one of the mapped gray levels output by said mapping means and outputting the selected level. - View Dependent Claims (27, 28, 29, 30, 31)
-
-
32. An image enhancing circuit in which an input image signal expressed as a predetermined number (L) of gray levels is histogram-equalized, said circuit comprising:
-
first calculating means for calculating a gray level distribution in a picture unit of the input image signal; second calculating means for calculating a mean level in the picture unit of the input image signal; compensation means for adding a brightness compensation value according to a predetermined correction function to the mean level based on a mean brightness of the input image, and outputting a compensated mean level; third calculating means for dividing the gray level distribution into a predetermined number of sub-images according to the mean level and calculating a cumulative density function for each sub-image; mapping means for mapping a sample of the input image to mapped gray levels according to the cumulative density functions and the compensated mean level, and outputting the mapped gray levels; and outputting means for selectively outputting one of the mapped gray levels output from said mapping means as an enhanced signal, according to a result obtained by comparing the input image signal and the mean level. - View Dependent Claims (33, 34, 35, 36, 37)
-
-
38. A circuit for enhancing an image by histogram-equalizing an image signal represented by a predetermined number of gray levels in a picture unit, said circuit comprising:
-
a first calculator receiving the image signal, calculating a gray level distribution of the image signal and outputting a gray level distribution signal; a second calculator receiving the image signal, calculating a mean level of the image signal and outputting a mean level signal; a third calculator which receives the gray level distribution signal and the mean level signal, divides the gray level distribution into a predetermined number of sub-images based on the mean level signal and calculates a cumulative density function for each sub-image and outputs cumulative density function signals representing the calculated cumulative density functions; a mapping unit receiving the mean level signal, the cumulative density function signals and the image signal, mapping an input sample of the image signal to a plurality of gray levels based on the cumulative density functions, and outputting mapped gray levels corresponding to the input sample; and an output unit receiving the mapped gray levels from said mapping unit, comparing the input sample with the mean level, selecting one of the mapped gray levels output by said mapping unit based on the comparison and outputting the selected gray level. - View Dependent Claims (39, 40, 41, 42, 43, 44, 45)
-
Specification