Method and apparatus for blocking effect reduction
View Patent ↗By utilizing parameters derivable from a compressed bit-stream, measures are obtained of the subjective quality of a decoded picture, without the necessity for reference to source material. A blockiness measure is derived both locally and globally and used to control a de-blocking filter. An estimate of signal-to-noise ratio is derived from quantization values.
1. A method for estimating the signal to noise ratio of a picture signal decoded from a compressed bit-stream, comprising the steps of determining the quantization values employed in said compression, generating a measure of the bit rate of the compressed bit-stream and deriving said estimate by processing said quantization values and said measure of the bit rate.
2. A method according to claim 1 , wherein the step of determining the quantization values comprises the step of deriving an average quantizer value for a picture.
3. A method according to claim 2 , wherein said average is taken in a logarithmic domain.
4. A method according to claim 1 , further comprising the step of using said measure of the bit rate to form a measure of picture activity.
5. A method according to claim 4 , in which the compressed bit-stream includes predicted and non-predicted pictures, and wherein said measure of picture activity is formed of the most recent non-predicted picture.
6. A method according to claim 4 , wherein the measure of picture activity comprises a product of a number of compression bits and a further function of quantization.
7. A method according to claim 6 , wherein said further function of quantization is a quadratic function.
8. A method according to claim 6 , wherein said further function of quantization is modified to take into account deviations from a pre-defined quantization weighting matrix.
9. A method according to claim 1 , wherein a base signal to noise ratio is taken as an experimental value of signal to noise ratio employing the finest allowable quantization and a pre-determined quantization weighting matrix.
10. A method according to claim 1 , wherein the quantization values are quantizer level spacings.
11. A method according to claim 1 , wherein a noise measure taken at an upstream location is passed forward for comparison with a noise measure taken at the device under test.
12. A method for estimating the signal to noise ratio PSNR estimate of a picture signal decoded from a compressed bit-stream, comprising the steps of determining quantization values q employed in said compression; generating a measure c of the bit rate of the compressed bitstream and processing said values and said measure to derive:
PSNR
estimate
=
PSNR
0
-
∑
log
q
N
(
A
+
B
∑
c
·
f
(
Mq
)
)
where A is a parameter that can be zero and B is a parameter that can be 1.
13. A method according to claim 12 , wherein a base ratio PSNR 0 is taken as an experimental value of signal to noise ratio employing the finest allowable quantization and a pre-determined quantization weighting matrix.
14. A method according to claim 12 , wherein the function f(q) is a quadratic function of q.
15. A method according to claim 12 , wherein the quantization values are quantizer level spacing.
16. A new method according to claim 12 , wherein a noise measure taken at an upstream location is passed forward for comparison with a noise measure taken at the device under test.