Methods and systems for noise reduction and image enhancement
Aspects of the present invention are related to systems and methods for image enhancement, wherein an input image may be separated into two images: a first image comprising image content, from the input image, which may be sharpened; and a second image comprising image content, from the input image, which may be attenuated. One aspect of the present invention relates to determining a noise estimate associated with the input image and using the noise estimate to control the separation of the input image into the two images. Another aspect of the present invention relates to controlling the image separation based on the level of sharpening being applied to the first image.
1. A system for enhancing an input image, said system comprising:
an edge-preserving filter for generating, from an input image, a first image comprising image content, from said input image, to sharpen and a second image comprising image content, from said input image, to attenuate;
a noise estimator for determining a noise estimate associated with said input image, wherein said noise estimator comprises:
an image-noise statistic estimator for estimating at least one of grain noise, sensor noise and thermal noise in said input image, thereby producing an image-noise statistic;
a codec-noise statistic estimator for estimating compression noise in said input image, thereby producing a codec-noise statistic;
a combiner for combining said image-noise statistic and said codec-noise statistic; and
wherein said noise estimate is based on said combined said image-noise statistic and said codec-noise statistic; and
wherein a parameter value associated with said edge-preserving filter is based on said noise estimate.
2. The system as described in claim 1 , wherein said edge-preserving filter comprises a bi-lateral filter.
3. The system as described in claim 1 , wherein said parameter value associated with said edge-preserving filter is proportional to said noise estimate.
4. The system as described in claim 1 , wherein said parameter value is a range sigma value.
5. The system as described in claim 4 , wherein said range sigma value is proportional to said noise estimate.
6. The system as described in claim 1 , wherein said combiner comprises selecting the maximum value of said image-noise statistic and said codec-noise statistic for said combined said image-noise statistic and said codec-noise statistic.
7. The system as described in claim 1 , wherein said codec-noise statistic estimator estimates said codec noise statistic from said input image.
8. The system as described in claim 1 , wherein said codec-noise statistic estimator estimates said codec-noise statistic from coding parameters associated with said input image.
9. The system as described in claim 8 , wherein said codec-noise statistic estimator comprises a plurality of look-up tables, wherein each look-up table in said plurality of look-up tables is associated with a codec and at least one codec coding parameter.
10. The system as described in claim 1 further comprising a decoder for decoding a bitstream associated with said input image.
11. The system as described in claim 1 further comprising:
a sharpener for sharpening said first image;
an attenuator for attenuating said second image; and
an image combiner for combining said sharpened first image and said attenuated second image.
12. The system as described in claim 11 further comprising an up-sampler for up-sampling said combined image.
13. The system as described in claim 11 , wherein said sharpener comprises an un-sharp masking filter.
14. The system as described in claim 11 further comprising a sharpening compensator for compensating said combined said image-noise statistic and said codec-noise statistic to account for a level of sharpening associated with said sharpener, and wherein said parameter value associated with said edge-preserving filter is based on said compensated, combined said image-noise statistic and said codec-noise statistic.
15. The system as described in claim 14 , wherein:
said sharpener comprises an un-sharp masking filter; and
said compensating said combined said image-noise statistic and said codec-noise statistic to account for a level of sharpening associated with said sharpener is based on an un-sharp masking alpha parameter value.
16. The system as described in claim 15 , wherein said compensating said combined said image-noise statistic and said codec-noise statistic to account for said level of sharpening associated with said sharpener comprises accessing a sharpener look-up table using said un-sharp masking alpha parameter value as an index.
17. The system as described in claim 14 further comprising a temporal filter for combining said compensated, combined said image-noise statistic and said codec-noise statistic with a previous parameter value associated with said edge-preserving filter.
18. A system for enhancing an input image, said system comprising:
an edge-preserving filter for generating, from an input image, a first image comprising image content, from said input image, to sharpen and a second image comprising image content, from said input image, to attenuate;
a noise estimator for determining a noise estimate associated with said input image, wherein said noise estimate is based on a combination of an image-noise statistic and a codec-noise statistic;
a sharpener for sharpening said first image;
an attenuator for attenuating said second image;
a combiner for combining said sharpened first image and said attenuated second image;
a sharpening compensator for compensating said combination of said image-noise statistic and said codec-noise statistic to account for a level of sharpening associated with said sharpener; and
wherein a parameter value associated with said edge-preserving filter is based on said compensated combination of said image-noise statistic and said codec-noise statistic.
19. The system as described in claim 18 , wherein said sharpener comprises an un-sharp masking filter.
20. The system as described in claim 18 further comprising an up-sampler for up-sampling said combined said sharpened first image and said attenuated second image.
21. The system as described in claim 18 , wherein:
said sharpener comprises an un-sharp masking filter; and
said compensating said combination of said image-noise statistic and said codec-noise statistic to account for a level of sharpening associated with said sharpener is based on an un-sharp masking alpha parameter value.
22. The system as described in claim 21 , wherein said compensating said combination of said image-noise statistic and said codec-noise statistic to account for said level of sharpening associated with said sharpener comprises accessing a sharpener look-up table using said un-sharp masking alpha parameter value as an index.
23. The system as described in claim 18 further comprising a temporal filter for combining said compensated combination of said image-noise statistic and said codec-noise statistic with a previous parameter value associated with said edge-preserving filter.