Adaptive noise filter
A method for applying a filter to data to improve data quality and/or reduce file size. In one example, a region of interest of an image is identified. A histogram is generated of pixel intensity values in the region of interest. The histogram is iteratively updated to focus (zoom) in on the highest peak in the histogram. A Gaussian curve is fitted to the updated histogram. A bilateral filter is applied to the images, where parameters of the bilateral filter are based on the parameters of the Gaussian curve.
1. A processor-implemented method for applying a filter to an image, the method comprising the steps, implemented in one or more processors, of:
selecting a first region of the image;
generating an initial histogram of pixel intensity values in the first region, wherein the histogram includes a plurality of bins;
identifying a first bin as a bin having a greatest count of pixel intensity values;
identifying a left bin as a closest bin to the left of the first bin in the histogram that has a count that is less than a first count percentage of the greatest count;
identifying a right bin as a closest bin to the right of the first bin in the histogram that has a count that is less than a second count percentage of the greatest count;
generating an updated histogram of pixel intensity values having a range between the identified left bin and identified right bin;
fitting a Gaussian curve onto the updated histogram; and
applying a filter to the image based on parameters of the Gaussian curve.
2. A method according to claim 1 , wherein the first count percentage and the second count percentage are 2%.
3. A method according to claim 1 , wherein the filter is a bilateral filter.
4. A method according to claim 1 , wherein edges of the updated histogram are within a threshold percentage of the range of the pixel intensity values between the identified left bin and identified right bin.
5. A method according to claim 4 , wherein the threshold percentage is 80%.
6. A method according to claim 1 , wherein selecting the first region includes:
receiving the image;
dividing the image into a plurality of regions;
identifying one or more of the plurality of regions on which to perform local analysis;
performing local analysis on each of the one or more identified regions; and
selecting one of the one or more identified regions as the first region based on results of the local analysis.
7. A non-transitory computer readable medium storing code, which when executed by one or more processors cause the one or more processors to implement a method of applying a filter to an image, the code including instructions to:
select a first region of the image;
generate an initial histogram of pixel intensity values in the first region, wherein the histogram includes a plurality of bins;
identify a first bin as a bin having a greatest count of pixel intensity values;
identify a left bin as a closest bin to the left of the first bin in the histogram that has a count that is less than a first count percentage of the greatest count;
identify a right bin as a closest bin to the right of the first bin in the histogram that has a count that is less than a second count percentage of the greatest count;
generate an updated histogram of pixel intensity values having a range between the identified left bin and identified right bin;
fit a Gaussian curve onto the updated histogram; and
apply a filter to the image based on parameters of the Gaussian curve.
8. The computer-readable medium of claim 7 , wherein the instructions to select the first region of the image include instructions to:
receive the image;
divide the image into a plurality of regions;
identify one or more of the plurality of regions on which to perform local analysis;
perform local analysis on each of the one or more identified regions; and
select one of the one or more identified regions as the first region based on results of the local analysis.
9. A processor-implemented method for applying a filter to an image, the method comprising the steps, implemented in one or more processors, of:
selecting a first region of the image;
generating an initial histogram of pixel intensity values in the first region, wherein the histogram includes a plurality of bins; and
a) identifying a first bin as a bin having a greatest count of pixel intensity values;
b) identifying a left bin as a closest bin to the left of the first bin in the histogram that has a count that is less than a first count percentage of the greatest count;
c) identifying a right bin as a closest bin to the right of the first bin in the histogram that has a count that is less than a second count percentage of the greatest count; and
d) determining whether edges of the histogram are within a threshold percentage of the range of pixel intensity values between the identified left bin and identified right bin, and
i) if edges of the histogram are within a threshold percentage of the range of pixel intensity values between the identified left bin and identified right bin fitting a Gaussian curve onto the histogram, applying a filter to the image based on parameters of the Gaussian curve; and
ii) if edges of the histogram are not within a threshold percentage of the range of pixel intensity values between the identified left bin and identified right bin, generating an updated histogram of pixel intensity values having a range between the identified left bin and identified right bin, and repeating steps a)-d) using the updated histogram.