IP Library › Granted Patent US 9,888,240
Granted Patent B2
US 9,888,240 · App. 14/056,043 · Granted Feb 6, 2018

Video processors for preserving detail in low-light scenes

Inventors: Yeping Su (Sunnyvale, CA); Chris Y. Chung (Sunnyvale, CA); Hao Pan (Sunnyvale, CA); Hsi-Jung Wu (San Jose, CA); Ke Zhang (Leuven, BE)
Assignee: APPLE INC.
H04N19/117H04N19/136
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 9,888,240
App. No.
14/056,043
Granted
Feb 6, 2018
Kind
B2
Abstract

Video coding systems and methods are provided that protect against various artifacts (e.g., banding artifacts) in decoded frame content. The embodiments to protect against introducing artifacts include pre-processing the input video frame, controlling the encoding parameters and/or post-processing the decoded video frame. A method for encoding a frame may include calculating a dark-scene score for the frame based on measured statistics of the frame. The dark-scene score may be compared to a threshold to determine if the banding artifacts may be present in the frame. Based on the comparison, filtering parameters for the frame may be altered to reduce the amount of noise filtered from the frame. The frame may be filtered based on the altered filtering parameters and encoded.

Claims (58)

1. A method for encoding a frame in a sequence of frames, comprising:

measuring statistics of the frame;

based on the measured statistics, estimating a likelihood that coding of the frame will cause banding artifacts on decoding based on comparing an anti-banding noise threshold to a number of pixel blocks in the frame that have a noise level below a selected noise threshold;

adding noise to the frame based on the estimating;

calculating a dark-scene score for the frame based on the measured statistics;

comparing the dark-scene score to a threshold;

when the comparison indicates that the frame is dark, altering filtering parameters for the frame from a default level to reduce an amount of noise filtered from the frame;

filtering the frame based on the altered filtering parameters; and

encoding the filtered frame.

2. The method of claim 1 , wherein measuring statistics of the frame includes measuring a noise level of the frame.

3. The method of claim 1 , wherein measuring statistics of the frame includes measuring a brightness level of the frame.

4. The method of claim 1 , wherein calculating the dark-scene score includes determining a region of the frame having motion below a predetermined motion threshold and calculating the dark scene score for the region having motion below the predetermined motion threshold.

5. The method of claim 1 , wherein the dark-scene score represents a number of pixel blocks in the frame that have a brightness level below a brightness threshold and the noise level below the selected noise threshold.

6. The method of claim 1 , wherein the filtering parameters are altered in proportion to the value of the dark-scene score.

7. The method of claim 1 , wherein altering the filtering parameters comprises reducing a strength of the filter as compared to the default level.

8. The method of claim 1 , wherein estimating the likelihood that coding of the frame will cause the banding artifacts on decoding includes determining a number of pixel blocks in the frame that have a brightness level below a predetermined brightness level.

9. The method of claim 1 , wherein estimating the likelihood that coding of the frame will cause the banding artifacts on decoding includes determining a number of pixel blocks in the frame that have the noise level below the selected noise threshold.

10. The method of claim 1 , wherein dither noise is added to the frame and the amount of dither noise that is added is determined based on the statistics of the frame.

11. The method of claim 1 , wherein estimating the likelihood that coding of the frame will cause the banding artifacts on decoding includes determining regions of the frame that have likelihood to cause the banding artifacts on decoding and wherein noise is added to the regions based on the estimate.

12. The method of claim 1 , further comprising:

determining masking values for the frame, the masking value determining an allocation of bits in regions of the frame;

based on the estimating the likelihood, altering the masking values to allocate more bits to areas of the frame that are candidate for introducing banding artifacts;

determining quantization parameters for encoding the frame based on the masking values and frame dependencies; and

encoding the frame using the quantization parameters.

13. The method of claim 12 , wherein the masking values are altered by applying a sublinear function transform.

14. A non-transitory computer-readable medium embodied with computer-executable instructions for causing a computer to execute instructions, the computer instructions comprising:

measuring statistics of a frame;

based on the measured statistics, estimating a likelihood that coding of the frame will cause banding artifacts on decoding based on deter mining, a number of pixel blocks in the frame that have a brightness level below a predetermined brightness level;

adding noise to the frame based on the estimating;

calculating a dark-scene score for the frame based on the measured statistics;

comparing the dark-scene score to a threshold;

when the comparison indicates that the frame is dark, altering filtering parameters for the frame from a default level to reduce an amount of noise filtered from the frame;

filtering the frame based on the altered filtering parameters; and

encoding the filtered frame.

15. The non-transitory computer-readable medium of claim 14 , wherein measuring statistics of the frame includes measuring a noise level of the frame.

16. The non-transitory computer-readable medium of claim 14 , wherein measuring statistics of the frame includes a brightness level of the frame.

17. The non-transitory computer-readable medium of claim 14 , wherein calculating the dark-scene score includes determining a region of the frame having motion below a predetermined motion threshold and calculating the dark scene score for the region having motion below the predetermined motion threshold.

18. The non-transitory computer-readable medium of claim 14 , wherein the dark-scene score represents a number of pixel blocks in the frame that have the brightness level below the predetermined brightness threshold and a noise level below a noise threshold.

19. The non-transitory computer-readable medium of claim 14 , wherein the filtering parameters are altered in proportion to the value of the dark-scene score.

20. The non-transitory computer-readable medium of claim 14 , wherein altering the filtering parameters comprises reducing a strength of the filter as compared to the default level.

21. The method of claim 14 , wherein estimating the likelihood that the coding of the frame will cause the banding artifacts on decoding includes comparing an anti-banding noise threshold to a number of pixel blocks in the frame that have a noise level below a selected noise threshold.

22. An apparatus, comprising:

a video coder to:

measure statistics of a frame,

based on the measured statistics, estimate a likelihood that coding of the frame will cause banding artifacts on decoding based on a comparison of an anti-banding noise threshold to a number of pixel blocks in the frame that have a noise level below a selected noise threshold;

add noise to the frame based on the estimate;

calculate a dark-scene score for the frame based on the measured statistics;

compare the dark-scene score to a threshold,

when the comparison indicates that the frame is dark, alter filtering parameters from for the frame from a default level to reduce an amount of noise filtered from the frame,

filter the frame based on the altered filtering parameters, and

encode the filtered frame; and

a transmitter to transmit the encoded frame.

23. The apparatus of claim 22 , wherein measuring the statistics of the frame includes measuring a noise level of the frame.

24. The apparatus of claim 22 , wherein measuring the statistics of the frame includes measuring a brightness level of the frame.

25. The apparatus of claim 22 , wherein calculating the dark-scene score includes determining a region of the frame having motion below a predetermined motion threshold and calculating the dark scene score for the region having motion below the predetermined motion threshold.

26. The apparatus of claim 22 , wherein the dark-scene score represents a number of pixel blocks in the frame that have a brightness level below a brightness threshold and the noise level below the selected noise threshold.

27. The apparatus of claim 22 , wherein the filtering parameters are altered in proportion to the value of the dark-scene score.

28. The apparatus of claim 22 , wherein altering the filtering parameters comprises reducing a strength of the filter as compared to the default level.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Oct 17, 2013
From: SU, YEPING; CHUNG, CHRIS Y.; PAN, HAO; WU, HSI-JUNG; ZHANG, KE
To: APPLE INC.
Reel/Frame 031423/0852 →
Continuity (2)
Provisional Application 61817212 · Apr 29, 2013
Related Publication 20140321534A1 · Oct 30, 2014