Method and system arranged for filtering an image
View Patent ↗A method and a system, the system includes: a memory unit, an image region definer, a spectral analyzer, a determination module, and an image processor. The spectral analyzer is arranged to divide a frequency spectrum of each image region into at least three image region frequency range components (IRFRCs); and calculate an energy level for each IRFRC. The comparator configured to compare the energy level of each IRFRC to at least one frequency region threshold to provide a comparison result for each IRFRC; wherein each frequency region threshold is calculated based on an intensity parameter of an image section and based on an energy parameter of at least one reference image section; a determination module arranged to determine a processing operation for each IRFRC, based on a comparison result associated with the IRFRC and based on a configurable decision rule.
1. A computer implemented method for filtering image data, the method comprising:
receiving image data at the computer, wherein the image comprises a plurality of pixels;
defining, for each image pixel out of multiple image pixels of the plurality of pixels, an image region;
dividing a frequency spectrum of each image region into at least three image region frequency range components;
calculating an energy level for each of the image region frequency range components;
comparing the energy level of each image region frequency range component to at least one frequency region threshold to provide a comparison result for each image region frequency range component, wherein each frequency region threshold is calculated based on an intensity parameter of an image section and based on an energy parameter of at least one reference image section;
determining, a processing operation for each image region frequency range component, based on the comparison result associated with the image region frequency range component and based on a configurable decision rule, wherein the processing operation is selected from a filtering operation and a bypass operation;
processing each image region frequency range component according to said determining to provide processed image data; and
displaying a processed image based on the processed image data.
2. The method according to claim 1 comprising:
performing vertical processing of image data to provide vertical processed image data; and
performing horizontal processing of the vertical processed image data to provide horizontal processed image data;
wherein the performing of vertical processing comprises:
dividing a vertical frequency spectrum of each image region into at least three image region vertical frequency range components;
calculating an energy level for each of the vertical image region frequency range component;
comparing the energy level of each vertical image region frequency range component to at least one vertical frequency region threshold to provide a comparison result for each image region vertical frequency range component;
determining, a processing operation for each image region vertical frequency range component; and
processing each image region vertical frequency range component according to the determination.
3. The method according to claim 1 comprising calculating a frequency region threshold associated with an image region, wherein the image section is the image region and the at least one reference image section comprises image regions of previous images that are located at a same location as the image region.
4. The method according to claim 1 comprising calculating a frequency region threshold associated with a image region, wherein the image section is an entire image and the at least one reference image section comprises the image region and image regions of previous images that are located at a same location as the image region.
5. The method according to claim 1 comprising calculating a frequency region threshold associated with a image region, wherein the image comprises a video portion and a graphic portion and the image section is defined so as not to include the graphic portion.
6. The method according to claim 1 comprising dividing a frequency spectrum of each image region into at least five image region frequency range components.
7. The method according to claim 1 comprising dividing a vertical frequency spectrum of each image region into at least five image region vertical frequency range components.
8. The method according to claim 1 wherein the image data represents a Luma component of the image.
9. The method according to claim 1 comprising:
calculating the intensity parameter of the image section as an average gray level of pixels of the image section; and
calculating the energy parameter of at least one reference image section as an average value of reference image section first derivative values.
10. The method according to claim 1 comprising:
calculating, in parallel, energy levels for different image region frequency range components; and
comparing, in parallel, energy levels of different image region frequency components to different frequency region thresholds to provide comparison results.
11. The method according to claim 1 comprising:
defining, for an image region frequency range component multiple frequency region threshold sets; and
selecting a set of frequency region thresholds based on an intensity parameter of an image section associated with the image region.
12. The method according to claim 1 comprising:
calculating, for an image region frequency range component multiple frequency region thresholds that are proportional to an energy parameter of at least one reference image section associated with the image region.
13. The method according to claim 1 , wherein
the determining comprises determining whether to apply deflicker filtering to an image region high frequency range component; and
the processing comprises performing deflicker filtering to the image region high frequency range component.
14. A system arranged to filter an image, the system comprising:
a memory unit arranged to receive image data wherein the image comprises a plurality of pixels;
a image region definer arranged to define, for each image pixel out of multiple image pixels, an image region;
a spectral analyzer arranged to
divide a frequency spectrum of each image region into at least three image region frequency range components, and
calculate an energy level for each of the image region frequency range component;
a comparator configured to compare the energy level of each image region frequency range component to at least one frequency region threshold to provide a comparison result for each image region frequency range component, wherein each frequency region threshold is calculated based on an intensity parameter of an image section and based on an energy parameter of at least one reference image section;
a determination module arranged to determine a processing operation for each image region frequency range component, said determining based on a comparison result associated with the image region frequency range component and based on a configurable decision rule, wherein the processing operation is selected from a filtering operation and a bypass operation; and
an image processor arranged to process each image region frequency range component according to the determination to provide processed image data and send display instructions, based on the processed image data, that cause the display to display the processed image data.
15. The system according to claim 14 wherein the spectral analyzer comprises:
a vertical spectral analyzer and a horizontal spectral analyzer;
wherein the determination module comprises a vertical determination module and a horizontal determination module; and
wherein the comparator comprises a vertical comparator and a horizontal comparator.
16. The system according to claim 14 further comprising:
a frequency region threshold calculator arranged to calculate at least one frequency region threshold for each image region frequency component, wherein
the image section is the image region, and
the at least one reference image section comprises image regions of previous images that are located at a same location as the image region.
17. The system according to claim 14 further comprising:
a frequency region threshold calculator arranged to calculate at least one frequency region threshold for each image region frequency component, wherein
the image section is an entire image, and
the at least one reference image section comprises the image region and image regions of previous images that are located at a same location as the image region.
18. The system according to claim 14 further comprising:
a frequency region threshold calculator arranged to
calculate the intensity parameter of the image section as an average gray level of pixels of the image section, and
calculate the energy parameter of at least one reference image section as an average value of reference image section first derivative values.
19. The system according to claim 14 , wherein
the spectral analyzer is arranged to calculate, in parallel, energy levels for different image region frequency range components, and
the comparator is arranged to compare, in parallel, energy levels of different image region frequency components to different frequency region thresholds to provide comparison results.
20. The system according to claim 14 further comprising:
a frequency region threshold calculator arranged to
define, for an image region frequency range component multiple frequency region threshold sets, and
select a set of frequency region thresholds based on an intensity parameter of an image section associated with the image region.