Apparatus and method for encoding and decoding image data with selected quantization based on pixel bit depth
An apparatus and method for encoding and decoding image data. The image data encoding apparatus includes: a transformer transforming pixel values of an image in a time domain into pixel values in a frequency domain; a quantization coefficient determiner determining a quantization coefficient proportional to a number of bits per pixel of the image; a quantization unit quantizing the pixel values transformed by the transformer based on the quantization coefficient determined by the quantization coefficient determiner; and an entropy encoder generating a bitstream of the quantized pixel values.
1. An image data encoding apparatus comprising:
a transformer including at least one processing device transforming pixel values of an image in a time domain into pixel values in a frequency domain;
a quantization coefficient determiner determining a quantization coefficient proportional to a number of bits per pixel of the image, with the quantization coefficient being selectively different in a low frequency domain than in a high frequency domain based on a same quantization variable according to the determining;
a quantization unit quantizing the pixel values transformed by the transformer based on the quantization coefficient determined by the quantization coefficient determiner; and
an entropy encoder generating a bitstream of the quantized pixel values.
2. The image data encoding apparatus of claim 1 , further comprising:
a temporal predictor temporally predicting pixel values of a current block using a previous block in the time domain;
a dequantization unit dequantizing the pixel values quantized by the quantization unit based on the quantization coefficient determined by the quantization coefficient determiner;
a detransformer detransforming the pixel values dequantized by the dequantization unit; and
a temporal prediction compensator compensating for the pixel values predicted by the temporal predictor.
3. The image data encoding apparatus of claim 2 ,
wherein the dequantization unit dequantizes the quantized pixel values using the following equation when performing dequantization of a low frequency domain or uniform dequantization of a high frequency domain
c′=q×QS , and
wherein c′ is a dequantized pixel value, q is a quantized pixel value, and QS is the quantization coefficient.
4. The image data encoding apparatus of claim 2 , wherein the dequantization unit dequantizes the quantized pixel values using the following equation when performing non-uniform dequantization of a high frequency domain
c′=q×QS +{sign( c )× QP× 2 (N−8) }, and
wherein c′ is a dequantized pixel value, q is a quantized pixel value, QS is the quantization coefficient, {sign(c)×QP×2 (N−8) } is an offset for the non-uniform dequantization.
5. The image data encoding apparatus of claim 1 , wherein the quantization coefficient determiner determines a quantization coefficient in a low frequency domain using the following equation
{
QS
=
QP
×
2
(
N
-
8
)
×
2
(
QP
<
3
)
QS
=
2
(
N
-
8
)
×
8
(
QP
=
3
or
QP
=
4
)
QS
=
(
[
QP
×
2
(
N
-
8
)
]
/
2
)
+
6
(
3
<
QP
<
4
or
QP
>
4
)
,
and
wherein QS is the quantization coefficient, QP is a quantization variable defined by a user, and N is the number of bits per pixel.
6. The image data encoding apparatus of claim 1 , wherein the quantization coefficient determiner determines a quantization coefficient in a high frequency domain using the following equation
QS=QP× 2 (N−8) ×2, and
wherein QS is the quantization coefficient and QP is a quantization variable defined by a user.
7. The image data encoding apparatus of claim 6 , wherein the quantization unit quantizes the transformed pixel values using the following equation when performing quantization of a low frequency domain or uniform quantization of a high frequency domain
q=c/QS , and
wherein q is a quantized pixel value, c is a pixel value transformed by the transformer, and QS is the quantization coefficient.
8. The image data encoding apparatus of claim 6 , wherein the quantization unit quantizes the transformed pixel values using the following equation when performing non-uniform quantization of a high frequency domain
q =( c +{sign( c )× QP× 2 (N−8) ×r })/ QS , and
wherein q is a quantized pixel value, c is a pixel value transformed by the transformer, QS is the quantization coefficient, and {sign(c)×QP×2 (N−8) ×r} is an offset for the non-uniform quantization, wherein sign(c)=1 when c>0, sign(c)=0 when c=0, and sign(c)=−1 when c<0, and r is a constant within a range 0≦r≦1.
9. An image data encoding apparatus comprising:
a transformer including at least one processing device transforming pixel values of an image in a time domain into pixel values in a frequency domain;
a quantization coefficient determiner determining a quantization coefficient proportional to a number of bits per pixel of the image;
a quantization unit quantizing the pixel values transformed by the transformer based on the quantization coefficient determined by the quantization coefficient determiner; and
an entropy encoder generating a bitstream of the quantized pixel values,
wherein the quantization coefficient determiner determines a quantization coefficient in a low frequency domain using the following equation
{
QS
=
QP
×
2
(
N
-
8
)
×
2
(
QP
<
3
)
QS
=
2
(
N
-
8
)
×
8
(
QP
=
3
or
QP
=
4
)
QS
=
(
[
QP
×
2
(
N
-
8
)
]
/
2
)
+
6
(
3
<
QP
<
4
or
QP
>
4
)
,
and
wherein QS is the quantization coefficient, QP is a quantization variable defined by a user, and N is the number of bits per pixel.
10. An image data encoding apparatus comprising:
a transformer including at least one processing device transforming pixel values of an image in a time domain into pixel values in a frequency domain;
a quantization coefficient determiner determining a quantization coefficient proportional to a number of bits per pixel of the image;
a quantization unit quantizing the pixel values transformed by the transformer based on the quantization coefficient determined by the quantization coefficient determiner; and
an entropy encoder generating a bitstream of the quantized pixel values,
wherein the quantization coefficient determiner determines a quantization coefficient in a high frequency domain using the following equation
QS=QP× 2 (N−8) ×2, and
wherein QS is the quantization coefficient and QP is a quantization variable defined by a user.
11. The image data encoding apparatus of claim 10 , wherein the quantization unit quantizes the transformed pixel values using the following equation when performing quantization of a low frequency domain or uniform quantization of a high frequency domain
q=c/QS , and
wherein q is a quantized pixel value, c is a pixel value transformed by the transformer, and QS is the quantization coefficient.
12. The image data encoding apparatus of claim 10 , wherein the quantization unit quantizes the transformed pixel values using the following equation when performing non-uniform quantization of a high frequency domain
q =( c +{sign( c )× QP× 2 (N−8) ×r })/ QS , and
wherein q is a quantized pixel value, c is a pixel value transformed by the transformer, QS is the quantization coefficient, and {sign(c)×QP×2 (N−8) ×r} is an offset for the non-uniform quantization, wherein sign(c)=1 when c>0, sign(c)=0 when c=0, and sign(c)=−1 when c<0, and r is a constant within a range 0≦r≦1.
13. An image data encoding apparatus comprising:
a transformer including at least one processing device transforming pixel values of an image in a time domain into pixel values in a frequency domain;
a quantization coefficient determiner determining a quantization coefficient proportional to a number of bits per pixel of the image;
a quantization unit quantizing the pixel values transformed by the transformer based on the quantization coefficient determined by the quantization coefficient determiner;
an entropy encoder generating a bitstream of the quantized pixel values;
a temporal predictor temporally predicting pixel values of a current block using a previous block in the time domain;
a dequantization unit dequantizing the pixel values quantized by the quantization unit based on the quantization coefficient determined by the quantization coefficient determiner;
a detransformer detransforming the pixel values dequantized by the dequantization unit; and
a temporal prediction compensator compensating for the pixel values predicted by the temporal predictor,
wherein the dequantization unit dequantizes the quantized pixel values using the following equation when performing non-uniform dequantization of a high frequency domain
c′=q×QS +{sign( c )× QP× 2 (N−8) }, and
wherein c′ is a dequantized pixel value, q is a quantized pixel value, QS is the quantization coefficient, {sign(c)×QP×2 (N−8) } is an offset for the non-uniform dequantization.
14. An image data decoding apparatus comprising:
an entropy decoder including at least one processing device decoding a bitstream of pixel values of an image;
a quantization coefficient determiner calculating a quantization coefficient proportional to a number of bits per pixel of the image from the bitstream, with the quantization coefficient being selectively different in a low frequency domain than in a high frequency domain based on a same quantization variable according to the determining;
a dequantization unit dequantizing the pixel values decoded by the entropy decoder based on the quantization coefficient calculated by the quantization coefficient determiner; and
a detransformer detransforming the dequantized pixel values having the number of bits per pixel.
15. The image data decoding apparatus of claim 14 , wherein the quantization coefficient determiner calculates a quantization coefficient in a high frequency domain using the following equation
QS=QP× 2 (N−8) ×2, and
wherein QS is the quantization coefficient and QP is a quantization variable defined by a user.
16. The image data decoding apparatus of claim 14 , wherein the dequantization unit dequantizes the quantized pixel values using the following equation when performing dequantization of a low frequency domain or uniform dequantization of a high frequency domain
c′=q×QS , and
wherein c′ is a dequantized pixel value, q is a quantized pixel value, and QS is the quantization coefficient.
17. The image data decoding apparatus of claim 14 , further comprising a temporal prediction compensator compensating for the temporally predicted pixel values of the detransformed pixel values.
18. The image data decoding apparatus of claim 14 , wherein the quantization coefficient determiner calculates a quantization coefficient in a low frequency domain using the following equation
{
QS
=
QP
×
2
(
N
-
8
)
×
2
(
QP
<
3
)
QS
=
2
(
N
-
8
)
×
8
(
QP
=
3
or
QP
=
4
)
QS
=
(
[
QP
×
2
(
N
-
8
)
]
/
2
)
+
6
(
3
<
QP
<
4
or
QP
>
4
)
,
and
wherein QS is the quantization coefficient, QP is a quantization variable defined by a user, and N is the number of bits per pixel.
19. The image data decoding apparatus of claim 14 , wherein the dequantization unit dequantizes the quantized pixel values using the following equation when performing non-uniform dequantization of a high frequency domain
c′=q×QS +{sign( c )× QP× 2 (N−8) }, and
wherein c′ is a dequantized pixel value, q is a quantized pixel value, QS is the quantization coefficient, and {sign(c)×QP×2 (N−8) } is an offset for the non-uniform dequantization.
20. An image data decoding apparatus comprising:
an entropy decoder including at least one processing device decoding a bitstream of pixel values of an image;
a quantization coefficient determiner calculating a quantization coefficient proportional to a number of bits per pixel of the image from the bitstream;
a dequantization unit dequantizing the pixel values decoded by the entropy decoder based on the quantization coefficient calculated by the quantization coefficient determiner; and
a detransformer detransforming the dequantized pixel values having the number of bits per pixel,
wherein the quantization coefficient determiner calculates a quantization coefficient in a low frequency domain using the following equation
{
QS
=
QP
×
2
(
N
-
8
)
×
2
(
QP
<
3
)
QS
=
2
(
N
-
8
)
×
8
(
QP
=
3
or
QP
=
4
)
QS
=
(
[
QP
×
2
(
N
-
8
)
]
/
2
)
+
6
(
3
<
QP
<
4
or
QP
>
4
)
,
and
wherein QS is the quantization coefficient, QP is a quantization variable defined by a user, and N is the number of bits per pixel.
21. An image data decoding apparatus comprising:
an entropy decoder including at least one processing device decoding a bitstream of pixel values of an image;
a quantization coefficient determiner calculating a quantization coefficient proportional to a number of bits per pixel of the image from the bitstream;
a dequantization unit dequantizing the pixel values decoded by the entropy decoder based on the quantization coefficient calculated by the quantization coefficient determiner; and
a detransformer detransforming the dequantized pixel values having the number of bits per pixel,
wherein the dequantization unit dequantizes the quantized pixel values using the following equation when performing non-uniform dequantization of a high frequency domain
c′=q×QS +{sign( c )× QP× 2 (N−8) }, and
wherein c′ is a dequantized pixel value, q is a quantized pixel value, QS is the quantization coefficient, and {sign(c)×QP×2 (N−8) } is an offset for the non-uniform dequantization.
22. An image data encoding method comprising:
transforming, using at least one processing device, pixel values of an image in a time domain into pixel values in a frequency domain;
determining a quantization coefficient proportional to the number of bits per pixel of the image, with the quantization coefficient being selectively different in a low frequency domain than in a high frequency domain based on a same quantization variable according to the determining;
quantizing the transformed pixel values based on the determined quantization coefficient; and
generating a bitstream of the quantized pixel values.
23. The image data encoding method of claim 22 , wherein, in the determination of the quantization coefficient, a quantization coefficient in a high frequency domain is determined using the following equation
QS=QP× 2 (N−8) ×2, and
wherein QS is the quantization coefficient and QP is a quantization variable defined by a user.
24. The image data encoding method of claim 23 , wherein, in the quantization, when non-uniform quantization of a high frequency domain is performed, the transformed pixel values are quantized using the following equation
q =( c +{sign( c )× QP× 2 (N−8) ×r })/ QS , and
wherein q is a quantized pixel value, c is a pixel value transformed by the transformer, QS is the quantization coefficient, and {sign(c)×QP×2 (N−8) ×r} is an offset for the non-uniform quantization, wherein sign(c)=1 when c>0, sign(c)=0 when c=0, and sign(c)=−1 when c<0, and r is a constant within a range 0≦r≦1.
25. The image data encoding method of claim 22 , wherein, in the quantization, when quantization of a low frequency domain or uniform quantization of a high frequency domain is performed, the transformed pixel values are quantized using the following equation
q=c/QS , and
wherein q is a quantized pixel value, c is a pixel value transformed by the transformer, and QS is the quantization coefficient.
26. The image data encoding method of claim 22 , further comprising temporally predicting pixel values of a current block using a previous block in the time domain and proceeding to the transforming.
27. The image data encoding method of claim 22 , wherein, in the determination of the quantization coefficient, a quantization coefficient in a low frequency domain is determined using the following equation
{
QS
=
QP
×
2
(
N
-
8
)
×
2
(
QP
<
3
)
QS
=
2
(
N
-
8
)
×
8
(
QP
=
3
or
QP
=
4
)
QS
=
(
[
QP
×
2
(
N
-
8
)
]
/
2
)
+
6
(
3
<
QP
<
4
or
QP
>
4
)
,
and
wherein QS is the quantization coefficient, QP is a quantization variable defined by a user, and N is the number of bits per pixel.
28. An image data encoding method comprising:
transforming, using at least one processing device, pixel values of an image in a time domain into pixel values in a frequency domain;
determining a quantization coefficient proportional to the number of bits per pixel of the image;
quantizing the transformed pixel values based on the determined quantization coefficient; and
generating a bitstream of the quantized pixel values,
wherein, in the determination of the quantization coefficient, a quantization coefficient in a low frequency domain is determined using the following equation
{
QS
=
QP
×
2
(
N
-
8
)
×
2
(
QP
<
3
)
QS
=
2
(
N
-
8
)
×
8
(
QP
=
3
or
QP
=
4
)
QS
=
(
[
QP
×
2
(
N
-
8
)
]
/
2
)
+
6
(
3
<
QP
<
4
or
QP
>
4
)
,
and
wherein QS is the quantization coefficient, QP is a quantization variable defined by a user, and N is the number of bits per pixel.
29. An image data encoding method comprising:
transforming, using at least one processing device, pixel values of an image in a time domain into pixel values in a frequency domain;
determining a quantization coefficient proportional to the number of bits per pixel of the image;
quantizing the transformed pixel values based on the determined quantization coefficient; and
generating a bitstream of the quantized pixel values,
wherein, in the determination of the quantization coefficient, a quantization coefficient in a high frequency domain is determined using the following equation
QS=QP× 2 (N−8) ×2, and
wherein QS is the quantization coefficient and QP is a quantization variable defined by a user, and
wherein, in the quantization, when non-uniform quantization of a high frequency domain is performed, the transformed pixel values are quantized using the following equation
q =( c +{sign( c )× QP× 2 (N−8) ×r })/ QS , and
wherein q is a quantized pixel value, c is a pixel value transformed by the transformer, QS is the quantization coefficient, and {sign(c)×QP×2 (N−8) ×r} is an offset for the non-uniform quantization, wherein sign(c)=1 when c>0, sign(c)=0 when c=0, and sign(c)=−1 when c<0, and r is a constant within a range 0≦r≦1.
30. An image data decoding method comprising:
decoding, using at least one processing device, a bitstream of pixel values of an image;
calculating a quantization coefficient proportional to a number of bits per pixel of the image from the bitstream, with the quantization coefficient being selectively different in a low frequency domain than in a high frequency domain based on a same quantization variable according to the determining;
dequantizing the decoded pixel values based on the calculated quantization coefficient; and
detransforming the dequantized pixel values having the number of bits per pixel.
31. The image data decoding method of claim 30 , wherein, in the calculation of the quantization coefficient, a quantization coefficient in a high frequency domain is calculated using the following equation
QS=QP× 2 (N−8) ×2, and
wherein QS is the quantization coefficient and QP is a quantization variable defined by a user.
32. The image data decoding method of claim 30 , wherein, in the dequantization, when dequantization of a low frequency domain or uniform dequantization of a high frequency domain is performed, the quantized pixel values are dequantized using the following equation
c′=q×QS , and
wherein c′ is a dequantized pixel value, q is a quantized pixel value, and QS is the quantization coefficient.
33. The image data decoding method of claim 30 , further comprising after the detransforming, compensating for the temporally predicted pixel values of the detransformed pixel values.
34. The image data decoding method of claim 30 , wherein, in the calculation of the quantization coefficient, a quantization coefficient in a low frequency domain is calculated using the following equation
{
QS
=
QP
×
2
(
N
-
8
)
×
2
(
QP
<
3
)
QS
=
2
(
N
-
8
)
×
8
(
QP
=
3
or
QP
=
4
)
QS
=
(
[
QP
×
2
(
N
-
8
)
]
/
2
)
+
6
(
3
<
QP
<
4
or
QP
>
4
)
,
and
wherein QS is the quantization coefficient, QP is a quantization variable defined by a user, and N is the number of bits per pixel.
35. The image data decoding method of claim 30 , wherein, in the dequantization, when non-uniform dequantization of a high frequency domain is performed, the quantized pixel values are dequantized using the following equation
c′=q×QS +{sign( c )× QP× 2 (N−8) }, and
wherein c′ is a dequantized pixel value, q is a quantized pixel value, QS is the quantization coefficient, and {sign(c)×QP×2 (N−8) } is an offset for the non-uniform dequantization.
36. An image data decoding method comprising:
decoding, using at least one processing device, a bitstream of pixel values of an image;
calculating a quantization coefficient proportional to a number of bits per pixel of the image from the bistream;
dequantizing the decoded pixel values based on the calculated quantization coefficient; and
detransforming the dequantized pixel values having the number of bits per pixel,
wherein, in the calculation of the quantization coefficient, a quantization coefficient in a low frequency domain is calculated using the following equation
{
QS
=
QP
×
2
(
N
-
8
)
×
2
(
QP
<
3
)
QS
=
2
(
N
-
8
)
×
8
(
QP
=
3
or
QP
=
4
)
QS
=
(
[
QP
×
2
(
N
-
8
)
]
/
2
)
+
6
(
3
<
QP
<
4
or
QP
>
4
)
,
and
wherein QS is the quantization coefficient, QP is a quantization variable defined by a user, and N is the number of bits per pixel.
37. An image data decoding method comprising:
decoding, using at least one processing device, a bitstream of pixel values of an image;
calculating a quantization coefficient proportional to a number of bits per pixel of the image from the bistream;
dequantizing the decoded pixel values based on the calculated quantization coefficient; and
detransforming the dequantized pixel values having the number of bits per pixel,
wherein, in the dequantization, when non-uniform dequantization of a high frequency domain is performed, the quantized pixel values are dequantized using the following equation
c′=q×QS +{sign( c )× QP× 2 (N−8) }, and
wherein c′ is a dequantized pixel value, q is a quantized pixel value, QS is the quantization coefficient, and {sign(c)×QP×2 (N−8) } is an offset for the non-uniform dequantization.
38. An image data decoding method comprising:
decoding, using at least one processing device, a bitstream of pixel values of an image;
determining a quantization coefficient proportional to a number of bits per pixel of the image, with the quantization coefficient being selectively different in a low frequency domain than in a high frequency domain based on a same quantization variable according to the determining;
dequantizing the decoded pixel values based on the determined quantization coefficient; and
detransforming the dequantized pixel values,
wherein the bitstream is an encoder generated bitstream, the encoder transforming to-be-encoded pixel values of a to-be-encoded image in a time domain into to-be encoded pixel values in a frequency domain, quantizing the transformed pixel values with a quantization coefficient equal to the determined quantization coefficient, and generating the encoder generated bitstream of the quantized pixel values.
39. A non-transitory computer readable recording medium storing a computer readable program to control at least one processing device to implement the method of claim 22 .
40. A non-transitory computer readable recording medium storing a computer readable program to control at least one processing device to implement the method of claim 30 .