Adaptive noise removal for video images with less correction of current pixel for higher variations between surrounding pixels

• US 5,768,440 A
• Filed: 12/18/1996
• Issued: 06/16/1998
• Est. Priority Date: 12/18/1996
• Status: Expired due to Term

• Current Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Original Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Inventors: Harrington, Steven J., Campanelli, Michael R.
• Primary Examiner(s): ROGERS, SCOTT A
• Application Number:
  US08/767,352
• Time in Patent Office: 545 Days
• Field of Search: 382/261, 382/260, 382/262, 382/264, 382/275, 382/254, 358/463, 358/447
• US Class Current: 382/261
• CPC Class Codes: G06T 5/20 using local operators

Abstract

A method of applying noise removal to an image so that the areas having the greatest variance have the least correction. First, the variance between each image pixel and its surrounding pixels is determined. Next, for the neighborhood around each pixel, a density of variance is determined. Finally the density of variance for the neighborhood around each pixel is used to correct the original pixel, the direction of correction being toward the mean value in the surrounding pixels. The result is that relatively uniform areas of the image, within which noise is most obvious, will get the most correction and areas that are textured will get the least.

Claims

1. An adaptive noise removal method for pixels of an original digital image comprising the steps of:

generating from the original image a first numerical value which is a function of the difference between the current pixel and other pixels in a first neighborhood surrounding the current pixel,

generating from the first numerical value a second numerical value which is a function of the amount of variation between pixel values in a second neighborhood surrounding the current pixel, and

using the second numerical value of each pixel to correct the current pixel of the original image by decreasing the difference between the value of the current pixel and the other pixels in the first neighborhood, the higher second numerical values of each pixel resulting in less correction to each pixel in the original image.

2. The method of claim 1 wherein the first numerical value of the current pixel is equal to the variance of the current pixel in the original image compared to the other pixels in the first neighborhood.

3. The method of claim 1 wherein the calculated second numerical value of the current pixel is equal to the root mean square of the first numerical values.

4. The method of claim 1 wherein the correction decreases the difference between the value of the current pixel and the median value of the other pixels in the original image.