System, method, and computer program product for implementing a spatially varying unsharp mask noise reduction filter
A system, method, and computer program product for applying a spatially varying unsharp mask noise reduction filter is disclosed. The spatially varying unsharp mask noise reduction filter generates a low-pass filtered image by applying a low-pass filter to a digital image, generates a high-pass filtered image of the digital image, and generates an unsharp masked image based on the low-pass filtered image and the high-pass filtered image. The filter also blends the low-pass filtered image with the unsharp masked image based on a shaping function.
1. A method comprising:
generating a low-pass filtered image by applying a low-pass filter to a digital image;
generating a high-pass filtered image of the digital image;
generating an unsharp masked image based on the low-pass filtered image and the high-pass filtered image; and
blending the low-pass filtered image with the unsharp masked image based on a shaping function,
wherein blending the low-pass filtered image with the unsharp masked image comprises performing a linear interpolation between the low-pass filtered image and the unsharp masked image based on the shaping function,
wherein an input to the shaping function comprises a function of a pixel value of the high-pass filtered image, and
wherein the input to the shaping function is based on a noise amplitude cutoff parameter.
2. The method of claim 1 , wherein generating the high-pass filtered image comprises calculating the difference between the digital image and the low-pass filtered image.
3. The method of claim 1 , wherein generating the unsharp masked image comprises combining the low-pass filtered image and a scaled version of the high-pass filtered image.
4. The method of claim 3 , wherein generating the unsharp masked image comprises scaling the high-pass filtered image based on a coring function.
5. The method of claim 1 , wherein the low-pass filtered image is generated by applying a convolution kernel to each pixel of the digital image.
6. The method of claim 5 , wherein the convolution kernel is implemented in software configured to be executed on a graphics processing unit.
7. The method of claim 1 , wherein a value returned by the shaping function is generated using a look-up table (LUT).
8. The method of claim 7 , wherein an index into the LUT comprises a function of a pixel value of the high-pass filtered image.
9. The method of claim 7 , wherein the LUT table represents a Gaussian function.
10. A non-transitory computer-readable storage medium storing instructions that, when executed by a processor, cause the processor to perform steps comprising:
generating a low-pass filtered image by applying a low-pass filter to a digital image;
generating a high-pass filtered image of the digital image;
generating an unsharp masked image based on the low-pass filtered image and the high-pass filtered image; and
blending the low-pass filtered image with the unsharp masked image based on a shaping function,
wherein blending the low-pass filtered image with the unsharp masked image comprises performing a linear interpolation between the low-pass filtered image and the unsharp masked image based on the shaping function,
wherein an input to the shaping function comprises a function of a pixel value of the high-pass filtered image, and
wherein the input to the shaping function is based on a noise amplitude cutoff parameter.
11. The non-transitory computer-readable storage medium of claim 10 , wherein blending the low-pass filtered image with the unsharp masked image comprises performing a linear interpolation between the low-pass filtered image and the unsharp masked image based on the shaping function.
12. The non-transitory computer-readable storage medium of claim 11 , wherein an input to the shaping function comprises a function of a pixel value of the high-pass filtered image.
13. The non-transitory computer-readable storage medium of claim 10 , wherein a value returned by the shaping function is generated using a look-up table (LUT).
14. A system, comprising:
a memory storing a digital image; and
a processing unit configured to:
generate a low-pass filtered image by applying a low-pass filter to a digital image,
generate a high-pass filtered image of the digital image,
generate an unsharp masked image based on the low-pass filtered image and the high-pass filtered image, and
blend the low-pass filtered image with the unsharp masked image based on a shaping function,
wherein blending the low-pass filtered image with the unsharp masked image comprises performing a linear interpolation between the low-pass filtered image and the unsharp masked image based on the shaping function,
wherein an input to the shaping function comprises a function of a pixel value of the high-pass filtered image, and
wherein the input to the shaping function is based on a noise amplitude cutoff parameter.
15. The system of claim 14 , wherein blending the low-pass filtered image with the unsharp masked image comprises performing a linear interpolation between the low-pass filtered image and the unsharp masked image based on the shaping function.
16. The system of claim 14 , wherein the processing unit is a graphics processing unit.
17. The system of claim 16 , wherein the processing unit is included in a system-on-chip (SoC) that further comprises a central processing unit.