Correction of shading effects in video images

• US 4,695,884 A
• Filed: 02/01/1985
• Issued: 09/22/1987
• Est. Priority Date: 12/30/1982
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Anastassiou, Dimitris, Pennebaker, William B.
• Primary Examiner(s): Groody, James J.
• Assistant Examiner(s): Parker, Michael D.
• Application Number:
  US06/697,300
• Time in Patent Office: 963 Days
• Field of Search: 358/163, 358/166, 358/167, 358/282, 358/284, 382/51, 382/54
• US Class Current: 348/615
• CPC Class Codes: G06T 5/20 using local operators
• CPC Class Codes: H04N 1/403 Discrimination between the two tones in the picture signal of a two-tone original

Abstract

Video images are corrected for shading effects by generating a histogram of picture element (pel) gray scale intensity values and calculating the median and the black and white extremes thereof. A range of permitted background gray scale values is empirically selected such as by calculating the distance between the median and the closest color extreme and setting the range as twice that distance centered at the median. A background image is then created by sampling the intensity values of the original image pels and using a look-up table to remap the image. In one embodiment, pel intensity values are compared with the median value and only corrections lying within the range are carried out. In another embodiment the original image pels are remapped from 8 bits to 5 bits with all values outside the permitted range set to that of the nearest color extreme. A duplicate image is created from the remapped image and the intensity values of a number of pels in a predetermined region about a first pel in this image are sampled and the value of the corresponding pel in the background image is changed to the largest value among those of the first pel and the sampled pels. This sampling and changing is continued using a look-up table with a 7 bit output until all of the pels in the background image have been changed accordingly. The duplicate image is then changed to conform to the changed background image and the sampling procedure is carried out over a larger surrounding region, or neighborhood, and a larger number of pels with the corresponding changes being made in the background image. The foregoing conforming, sampling, and changing operations may be repeated until a background image with the desired level of shading correction is achieved.

Claims

1. A method for correcting shading effects in a video image using a histogram of gray scale intensity values of picture elements or pels making up the image, which histogram contains a median and black and white color extremes, comprising the steps of:

generating a histogram of distribution of gray scale intensity values of the pels in an original image to be corrected;

determining an acceptable range of background intensity values by;

calculating the median and the black and white color extremes in the histogram; and

calculating the distance between the histogram median and the color extreme on the side of the median having the majority of gray scale intensity values and setting said range as twice that distance centered at the median;

creating a first background image by sampling the gray scale intensity value of each pel to be corrected and setting all values outside of said acceptable range to that of the color extreme least distant therefrom;

creating a duplicate background image from said first background image;

sampling the gray scale values of a number of pels in a predetermined region around a first pel in said duplicate background image;

changing the value of the pel in said first background image corresponding to said first pel, to the largest value among those of said first pel and the sampled pels;

continuing the foregoing sampling and changing steps until all of the pel values in said first background image have been changed accordingly;

changing the duplicate background image in accordance with the changes in said first background image;

repeating the sampling step over a larger region and a larger number of pels with the corresponding changing in the first background image; and

repeating the foregoing changing until a desired corrected background image is obtained which becomes the shading correction for the original image.

2. A method as in claim 1 further comprising the steps of carrying out a further correction while sampling by detecting changed pel values below a given value and storing indications thereof;

and using said indications to cause only said below value pels to be corrected during subsequent sampling and changing.

3. A method as in claim 2 wherein said method is carried out in a computer and said indications are stored in an overlay plane, which indications are changed during successive samplings accordingly to provide a check as to the adequacy of the corrected image.

4. A method for correcting shading effects in a video image using

a histogram of gray scale intensity values of picture elements or pels making up the image, which histogram contains a median and black and white color extremes, comprising the steps of:

generating a histogram of distribution of gray scale intensity values of the pels in an original image to be corrected;

determining an acceptable range of background intensity values by;

calculating the median and the black and white color extremes in the histogram; and

calculating the distance between the histogram median and the color extreme on the side of the median having the majority of gray scale intensity values and setting said range as twice that distance centered at the median; and

creating a correction image by;

comparing the gray scale intensity value of each pel to be corrected in the original image with the median value and calculating respective difference values;

correcting the value of each said pel by its respective difference value when the latter is within said range; and

not correcting the value of each said pel when its respective difference value is outside of said range.

5. A method as in claim 4 wherein said original image is remapped to said correction image using a lookup table, the output of which is said respective difference values.

6. A method as in claim 4 comprising the further steps of:

sampling the gray scale values of a number of pels in a predetermined region around a first pel in said correction image;

summing said sampled values and dividing the sum by the number of samples to obtain an average value and using said average value to correct the value of said first pel; and

continuing the foregoing summing and correcting steps until all of the pels in said correction image have been corrected accordingly;

7. A system for correcting shading effects in a video image using a histogram of gray scale intensity values of picture elements or pels making up the image, which histogram contains a median and black and white color extremes, comprising:

means for generating a histogram of distribution of gray scale intensity values of the pels in an original image to be corrected;

means for determining an acceptable image of background intensity values comprising;

means for calculating the median and the black and white color extremes in the histogram; and

means for calculating the distance between the histogram median and the color extreme on the side of the median having the majority of gray scale intensity values and setting said range as twice that distance centered at the median;

means for creating a first background image by sampling the gray scale intensity value of each pel to be corrected and setting all values outside of said acceptable range to that of the color extreme least distant therefrom;

means for creating a duplicate background image from said first background image;

first means for sampling the gray scale values of a number of pels in a predetermined region around a first pel in said duplicate background image;

first means for changing the value of the pel in said first background image corresponding to said first pel, to the largest value among those of said first pel and the sampled pels;

means for reactivating said first sampling and changing means until all of the pel values in said first background image have been changed accordingly;

second means for changing the duplicate background image in accordance with the changes in said first background image;

second means for sampling the gray scale values in said duplicate image over a larger region and a larger number of pels than said first sampling means and activating said first changing means to make the corresponding changes in the first background image; and

means for activating said second changing means until a desired corrected background image is obtained which becomes the shading correction for the original image.

8. A system as in claim 7 further comprising:

means for carrying out a further correction while sampling by detecting changed pel values below a given value and storing indications thereof; and

means for using said indications to cause only said below value pels to be corrected during subsequent sampling and changing.

9. A system as in claim 8 wherein said carrying out means comprises:

computer means for storing said indications in an overlay plane, and for changing said indications during successive samplings accordingly to provide a check as to the adequacy of the corrected image.

10. A system for correcting shading effects in a video image using a histogram of gray scale intensity values of picture elements or pels making up the image, which histogram contains a median and black and white color extremes, comprising:

means for generating a histogram of distribution of gray scale intensity values of the pels in an original image to be corrected;

means for determining an acceptable range of background intensity values comprising;

means for calculating the median and the black and white color extremes in the histogram; and

means for calculating the distance between the histogram median and the color extreme on the side of the median having the majority of gray scale intensity values and setting said range as twice that distance centered at the median; and

means for creating a correction image comprising;

means for comparing the gray scale intensity value of each pel to be corrected in the original image with the median value and calculating respective difference values;

means for correcting the value of each said pel by its respective difference value when the latter is within said range; and

means for not correcting the value of each said pel when its respective difference value is outside of said range.

11. A system as in claim 10 further comprising means for remapping said original image to said correction image using a lookup table, the output of which is said respective difference values.

12. A system as in claim 10 further comprising:

means for sampling the gray scale values of a number of pels in a predetermined region around a first pel in said correction image;

means for summing said sampled values and dividing the sum by the number of samples to obtain an average value;

means using said average value correcting the value of said first pel; and

means for reactivating said summing means and said correcting means until all of the pels in said correction image have been corrected accordingly.