Reduction of banding artifacts in image processing

1. A method comprising:

detecting a sinusoidal light waveform in a frame that includes a banding artifact;

obtaining a gain map based on the sinusoidal light waveform;

obtaining an inverted gain map based on the gain map via a sinusoidal regression based on a timing of the frame and a frequency of the sinusoidal light waveform;

applying the inverted gain map to the frame to repair the banding artifact to obtain a repaired frame; and

outputting the repaired frame.

2. The method of claim 1 , further comprising:

applying a subsequent inverted gain map to a subsequent frame, wherein the subsequent inverted gain map is based on the frequency of the sinusoidal light waveform.

3. The method of claim 1 , wherein the sinusoidal light waveform is detected based on image content data.

4. The method of claim 1 , wherein the sinusoidal light waveform is detected based on luminance data of the frame.

5. The method of claim 1 , wherein the sinusoidal light waveform is detected based on a timing of each frame based on a sensor configuration.

6. The method of claim 5 , wherein the sensor configuration includes a clock.

7. The method of claim 5 , wherein the sensor configuration includes a PLL divider.

8. The method of claim 1 further comprising:

performing a temporal subtraction to remove noise data from the frame.

9. The method of claim 1 , wherein the sinusoidal light waveform is detected based on infrared (IR) sensor data.

10. An image capture device comprising:

an image sensor configured to capture a frame that includes a sinusoidal light waveform banding artifact; and

an image processor configured to:

detect a sinusoidal light waveform in the frame;

perform a sinusoidal regression based on a timing of the frame and a frequency of the sinusoidal light waveform;

obtain an inverted gain map;

apply the inverted gain map to the frame; and

output the frame.

11. The image capture device of claim 10 , wherein the image processor is further configured to match an ambient light to a known light pulse frequency based on a location of the image capture device.

12. The image capture device of claim 10 , wherein the image processor is further configured to apply a subsequent inverted gain map to a subsequent frame, wherein the subsequent inverted gain map is based on the frequency of the sinusoidal light waveform.

13. The image capture device of claim 10 , wherein the frequency is 50 Hz or 60 Hz.

14. The image capture device of claim 10 , wherein the image processor is configured to perform the sinusoidal regression to synchronize a respective inverted gain map with each subsequent frame based on a timing of each subsequent frame and the frequency of the sinusoidal light waveform.

15. The image capture device of claim 10 , wherein the image processor is configured to apply each respective inverted gain map to a respective subsequent frame.

16. A method comprising:

detecting a sinusoidal light waveform in a first frame of a plurality of frames;

determining a frequency of the sinusoidal light waveform;

performing a sinusoidal regression to obtain a first inverted gain map for the first frame, wherein the sinusoidal regression is based on a timing of the first frame and the frequency of the sinusoidal light waveform;

applying the first inverted gain map to the first frame; and

applying a respective inverted gain map to each of the plurality of frames, wherein each respective gain map is based on a timing of each respective frame of the plurality of frames and the frequency of the sinusoidal light waveform.

17. The method of claim 16 , wherein the sinusoidal light waveform is detected based on luminance data of the first frame.

18. The method of claim 16 further comprising:

performing a temporal subtraction to remove noise data from the first frame.

19. The method of claim 16 , wherein the sinusoidal light waveform is detected based on infrared (IR) sensor data.

20. The method of claim 16 , wherein the frequency is 50 Hz or 60 Hz.