WO1999045645A1 - Cascading of up conversion and down conversion - Google Patents
Cascading of up conversion and down conversion Download PDFInfo
- Publication number
- WO1999045645A1 WO1999045645A1 PCT/GB1999/000682 GB9900682W WO9945645A1 WO 1999045645 A1 WO1999045645 A1 WO 1999045645A1 GB 9900682 W GB9900682 W GB 9900682W WO 9945645 A1 WO9945645 A1 WO 9945645A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- conversion
- filter
- down conversion
- γêæ
- chosen
- Prior art date
Links
- 238000006243 chemical reaction Methods 0.000 title claims abstract description 84
- 238000001914 filtration Methods 0.000 claims description 14
- 238000000034 method Methods 0.000 claims description 8
- 238000005070 sampling Methods 0.000 description 21
- 238000012545 processing Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 238000013459 approach Methods 0.000 description 1
- 239000012141 concentrate Substances 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03H—IMPEDANCE NETWORKS, e.g. RESONANT CIRCUITS; RESONATORS
- H03H17/00—Networks using digital techniques
- H03H17/02—Frequency selective networks
- H03H17/06—Non-recursive filters
Definitions
- This invention relates to digital filtering of sampled signals and in an important example to filtering of digital video.
- the sampling rate of a digital video has to be changed for converting between different formats.
- a video signal may be stored and transmitted in a low resolution format but displayed in an up converted high resolution format.
- the sampling rate of the low resolution is related to the sampling rate of the high resolution by a ratio 1 :M or by a ratio N:M where N and M are integers.
- N and M are integers.
- several cascades of up and down conversion filtering can occur. For example in a broadcast chain this is due to repeated editing, preview and storage operations.
- up conversion from the low resolution to the high resolution and subsequent down conversion should result in a transparent cascade.
- Transparent cascading of N:M up and M:N down conversion filtering imposes additional problems on the filter design when compared with a cascade of up and down conversion filters for ratios 1 :M and M:1.
- Figure 1 is a diagram showing the low resolution and high resolution sampling grid for1:2 up conversion and subsequent 2:1 down conversion
- Figure 2 is a diagram showing the low resolution and high resolution sampling grid for 3:4 up conversion and subsequent 4:3 down conversion and showing, additionally, the sampling grid for the intermediate resolution;
- Figure 3 is a block diagram illustrating the three stages of 3:4 up conversion, that is to say 1 :4 up sampling, interpolation filtering, and 3:1 down sampling;
- Figure 4 is a block diagram illustrating the three stages of 4:3 down conversion, i.e.that is to say 1 :3 up sampling, interpolation filtering, 4:1 down sampling; and
- Figure 5 is a graph showing the frequency response of 3:4 up conversion filter, bilinear and proposed filter.
- the sampling grid of the low resolution signal forms a subset of the sampling grid of the high resolution signal as shown in Figure Ifor the example ratio 1 :2.
- the samples of the low resolution signal are copied onto the corresponding positions of the high resolution grid on 1 :M up conversion filtering, one subset that contains every M-th sample of the up converted signal coincides with the set of samples of the low resolution signal.
- An up conversion filter that has this property is called a l:M Nyquist filter. -3 -
- the low resolution signal s is up sampled by the factor 1:4 resulting in the signal t which is defined on the intermediate sampling grid as follows,
- the interpolated signal u is down sampled by the factor
- Equations (1) to (3) define the functional relationship between the low resolution signal s and the up converted signal s U p,
- equation (6) a 3:4 up conversion filter g that fulfils equation (6) is called a 3:4 Nyquist filter.
- equation (5) is called the 3:4 Nyquist condition. - 5 -
- h denotes the impulse response of the 4:3 down conversion filter.
- a 4:3 down conversion filter h that fulfils equation (9) is called a 4:3 Nyquist filter.
- Equation (12) is the condition for transparent cascading that can be derived straightforwardly from equations (4) and (7). It shows that the filter coefficients of g and h cannot be chosen independently. For given coefficients of g equation (12) becomes a linear equation system for the coefficients of h and vice versa.
- the up converted signal s up should give a good picture quality on displaying. Therefore, firstly the coefficients of the up conversion filter g can be chosen to fulfil this requirement and then a corresponding down conversion filter h can be designed to comply with equation (12), resulting in a transparent cascade.
- the length of the impulse response is an obvious limitation in practical applications. Bilinear interpolation results in a short impulse response.
- the coefficients of the symmetric impulse response gui are shown in Table 1.
- Table 1 shows also the symmetric impulse response of the filter gf det that is proposed for 3:4 up conversion.
- the filter coefficients are derived from a windowed sin(x)/x waveform to give a better low pass characteristic than the bilinear interpolation filter.
- the frequency responses of gbn and gt ⁇ etare compared in Figure 5.
- the improved low pass characteristic comes at a price as 15 taps are needed for g-jei but only 7 taps for gus.
- the coefficients of gnet are rounded to six decimal digits as shown in Table 1. It is not difficult to verify that both up conversion filters of Table 1 are 3:4 Nyquist filters.
- the filter coefficients are also listed in Table 1. Similar to gfdet the coefficients of hun d o- ef are rounded to six decimal digits. In contrast to the up conversion filters, the corresponding down conversion filters of Table 1 do not have a symmetric impulse response. Again, the impulse response length of the proposed filter is larger, i.e. 29 taps are needed for hundo-def but only 9 taps for h un do-bn.
- the ratios 3:4 and 4:3 have been chosen by way of illustration only.
- the invention is more broadly applicable to cascaded N:M up conversion and subsequent M:N down conversion, where the up and down conversion ratios N:M and M:N respectively are rational numbers and the integers N and M satisfy the condition 1 ⁇ N ⁇ M.
- the invention is applicable not only to horizontal processing, but also to vertical processing and temporal processing. If processing is required in two or more of these dimensions, it will generally be possible to identify pairs of one dimensional filters in accordance with the present invention and then to cascade those filters to achieve the desired two or three dimensional processing.
- Table 1 Nyquist filter for 3:4 up conversion, bilinear (bil) and proposed filter (def) and Nyquist filter for subsequent 4:3 down conversion, undo bilinear (undo-bil), undo proposed (undo-def) 3:4 up conversion
Abstract
In digital video, 3:4 up conversion and 4:3 down conversion are transparently cascaded. The up conversion filter is: Sup(n)=ΣSinput(k) . g(3n - 4k) and the down conversion filter is: Sdown(n)=ΣSup(k) . h(4n - 3k). The pair of up and down conversion filters are designed so that Σh(4n - 3k) . g(3k - 4m) is equal to unity if n = m and is otherwise equal to zero.
Description
-1-
CASCADING OF UP CONVERSION AND DOWN CONVERSION
This invention relates to digital filtering of sampled signals and in an important example to filtering of digital video.
The sampling rate of a digital video has to be changed for converting between different formats. For example, in order to save bandwidth a video signal may be stored and transmitted in a low resolution format but displayed in an up converted high resolution format. Typically, the sampling rate of the low resolution is related to the sampling rate of the high resolution by a ratio 1 :M or by a ratio N:M where N and M are integers. Depending on the application, several cascades of up and down conversion filtering can occur. For example in a broadcast chain this is due to repeated editing, preview and storage operations. In order to achieve a perfect reconstruction of those parts of the pictures that have not been changed by editing operations up conversion from the low resolution to the high resolution and subsequent down conversion should result in a transparent cascade. Transparent cascading of N:M up and M:N down conversion filtering imposes additional problems on the filter design when compared with a cascade of up and down conversion filters for ratios 1 :M and M:1.
It is an object of one aspect of the present invention to provide an improved digital filtering process capable of transparent up conversion and cascaded down conversion. Accordingly, the present invention consists in one aspect in a digital filtering process for achieving a transparent cascade on N:M up conversion and subsequent M:N down conversion, where the up and down conversion ratios N:M and M:N respectively are rational numbers and the integers N and M satisfy the condition 1<N<M, wherein the up conversion filter operates on a sampled signal sinput and is chosen to take the form su (n) =∑ Sinput(k) . g(Nn - Mk) and wherein a corresponding down conversion filter operates on the up converted signal sup and is chosen to take the form Sdown(n) =∑ sup(k) . h(Mn - Nk); the pair (g,h) of up and
-2 - down conversion filters being chosen so that ∑ h(Mn - Nk).g(Nk - Mm) is equal to unity if n = m and is otherwise equal to zero.
The invention will now be described by way of example with reference to the accompanying drawings, in which :-
Figure 1 is a diagram showing the low resolution and high resolution sampling grid for1:2 up conversion and subsequent 2:1 down conversion;
Figure 2 is a diagram showing the low resolution and high resolution sampling grid for 3:4 up conversion and subsequent 4:3 down conversion and showing, additionally, the sampling grid for the intermediate resolution;
Figure 3 is a block diagram illustrating the three stages of 3:4 up conversion, that is to say 1 :4 up sampling, interpolation filtering, and 3:1 down sampling;
Figure 4 is a block diagram illustrating the three stages of 4:3 down conversion, i.e.that is to say 1 :3 up sampling, interpolation filtering, 4:1 down sampling; and
Figure 5 is a graph showing the frequency response of 3:4 up conversion filter, bilinear and proposed filter.
For a ratio 1 :M and isochronous sampling, the sampling grid of the low resolution signal forms a subset of the sampling grid of the high resolution signal as shown in Figure Ifor the example ratio 1 :2. Thus, if the samples of the low resolution signal are copied onto the corresponding positions of the high resolution grid on 1 :M up conversion filtering, one subset that contains every M-th sample of the up converted signal coincides with the set of samples of the low resolution signal. An up conversion filter that has this property is called a l:M Nyquist filter.
-3 -
Therefore, if a 1 :M Nyquist filter is followed by a M:1 down sampler one obtains mathematically the identity operator which makes the cascade transparent.
However, this approach does not work for a ratio N:M such as 3:4. The reason is that the sampling grid of the low resolution signal is not an entire subset of the sampling grid of the high resolution signal. For example in the case of 3:4 up conversion only every third grid position of the low resolution signal coincides with a corresponding grid position of the high resolution signal as shown in Figure 2. Thus, only one third of the samples can be copied from the low to the high resolution. Consequently, only one third of the samples of the low resolution signal can be recovered simply by sub-sampling the up converted signal.
Additional filtering is needed on down conversion to recover the remaining two- thirds.
An example of a filtering process according to the present invention will now be described. This will for illustration concentrate on the example ratios 3:4 on up conversion and 4:3 on down conversion.
First, some principles of up and down conversion filtering are briefly reviewed. The sampling grids of the low and the high resolution are shown in Figure 2 for 3:4 up conversion and subsequent 4:3 down conversion. Additionally, Figure 2 shows the intermediate sampling grid on which all filter operations are performed. It is assumed throughout that all filter operations are linear and shift invariant. Thus, each filter can be described by its impulse response. Let g denote the impulse response of the 3:4 up conversion filter. The 3:4 up conversion process can be considered in three stages, see Figure 3.
In the first stage the low resolution signal s is up sampled by the factor 1:4 resulting in the signal t which is defined on the intermediate sampling grid as follows,
(s(k) if = 4k 0 otherwise
-4 - ln the second stage the signal f is interpolated into the signal u,
u(n) = ∑t(k) - g(n-k) (2) k
In the third stage the interpolated signal u is down sampled by the factor
3:1 , resulting in the up converted signal
sup (n) = u(3n) (3)
Equations (1) to (3) define the functional relationship between the low resolution signal s and the up converted signal sUp,
sup(n) = ∑s(k) - gQn -Ak) (4) k
As every third position of the low resolution sampling grid coincides with a position of the high resolution sampling grid, the samples of these positions can be copied on 3:4 up conversion, sup(^n) = s(3n)
(5)
One can derive equation (5) from equation (4) if the filter g fulfils the
Nyquist condition for 1 :4 up conversion, f l if n = 0 0 otherwise
In this description a 3:4 up conversion filter g that fulfils equation (6) is called a 3:4 Nyquist filter. Correspondingly, equation (5) is called the 3:4 Nyquist condition.
- 5 -
Similar to 3:4 up conversion, the 4:3 down conversion process can be considered in three stages, see Figure 4. The relationship between the up converted signal sup, and the down converted signal Sdown becomes
sdown (n) = ∑^P(k) - h(4n - 3k) (7) k
where h denotes the impulse response of the 4:3 down conversion filter.
Similar to equation (5), the Nyquist condition for 4:3 down conversion can be specified, sdown (3n) = sup(4n) (8)
One can derive equation (8) from equation (7) if the filter h fulfils the Nyquist condition for 1 :3 up conversion,
f l if = 0 h(3n) = . (9)
0 otherwise
In this description a 4:3 down conversion filter h that fulfils equation (9) is called a 4:3 Nyquist filter.
One concludes from equations (5) and (8) that the cascade of a 3:4 Nyquist filter which is followed by a 4:3 Nyquist filter gives a perfect reconstruction for every third sample of the low resolution signal, i.e.
sdown(3ri) = s(3ή) (10)
However, in order to recover the remaining two-thirds, resulting in
Sdθm, (n) = s n) (11)
the pair (g, h) of up and down conversion filters has to fulfil the condition:
_, fl if n = m
∑h(4n -3k) - g(3k - 4m) = \ J . (12)
£ 10 otherwise
Equation (12) is the condition for transparent cascading that can be derived straightforwardly from equations (4) and (7). It shows that the filter coefficients of g and h cannot be chosen independently. For given coefficients of g equation (12) becomes a linear equation system for the coefficients of h and vice versa.
From equation (12) one can derive the corollary, that if equation (12) is fulfilled by the pair of up and down conversion filters (g,h) and, additionally, the impulse response of g is symmetric, i.e. g(n) =g(-n) then:
i) equation (12) is also fulfilled by the pair of filters (g, h_) with h_(n) = h(-n)
ii) equation (12) is also fulfilled by the pair of filters (g, h~) with h~ = (1- λ)-h + λ-h_ and -oo < λ < oo , in particular the impulse response of h~ = (h+h_)/2 is symmetric.
The up converted signal sup should give a good picture quality on displaying. Therefore, firstly the coefficients of the up conversion filter g can be chosen to fulfil this requirement and then a corresponding down conversion filter h can be designed to comply with equation (12), resulting in a transparent cascade.
The length of the impulse response is an obvious limitation in practical applications. Bilinear interpolation results in a short impulse response. The coefficients of the symmetric impulse response gui are shown in Table 1. However, bilinear interpolation filtering does not result in a sophisticated low pass
- 7 - characteristic. Therefore, Table 1 shows also the symmetric impulse response of the filter gfdetthat is proposed for 3:4 up conversion. The filter coefficients are derived from a windowed sin(x)/x waveform to give a better low pass characteristic than the bilinear interpolation filter. The frequency responses of gbn and gtøetare compared in Figure 5. The improved low pass characteristic comes at a price as 15 taps are needed for g-jei but only 7 taps for gus. The coefficients of gnet are rounded to six decimal digits as shown in Table 1. It is not difficult to verify that both up conversion filters of Table 1 are 3:4 Nyquist filters.
For the given 3:4 Nyquist filters gw and cfaet corresponding 4:3 Nyquist filters hυndo-bii and hun o-def respectively, are designed in order to comply with equation
(12). The filter coefficients are also listed in Table 1. Similar to gfdet the coefficients of hundo- ef are rounded to six decimal digits. In contrast to the up conversion filters, the corresponding down conversion filters of Table 1 do not have a symmetric impulse response. Again, the impulse response length of the proposed filter is larger, i.e. 29 taps are needed for hundo-def but only 9 taps for hundo-bn.
It follows from the corollary to equation (12) that the non-symmetric impulse responses of the down conversion filters of Table 1 can be converted into symmetric impulse responses by mirroring. However, this would increase the number of filter taps. It will be understood that this invention has been described by way of example only and that a wide variety of modifications are possible without departing from the scope of the invention.
It has already been explained that the ratios 3:4 and 4:3 have been chosen by way of illustration only. The invention is more broadly applicable to cascaded N:M up conversion and subsequent M:N down conversion, where the up and down conversion ratios N:M and M:N respectively are rational numbers and the integers N and M satisfy the condition 1<N<M. In the field of video, the invention is applicable not only to horizontal processing, but also to vertical processing and temporal processing. If processing is required in two or more of these dimensions, it will generally be possible to identify pairs of one dimensional filters in accordance
with the present invention and then to cascade those filters to achieve the desired two or three dimensional processing.
n 8bi,(n) n hundo-bil(n n W> n hundo-d f(n )
0 1 -8 1/3 0 1 -8 0.003449
-1, 1 3/4 -7 0 -1, 1 0.854892 -7 0
-2, 2 1 /2 -6 0 -2, 2 0.530332 -6 0
-3, 3 1 /4 -5 -4/3 -3, 3 0.204804 -5 0 otherwise 0 -A -1 /3 -4, 4 0 -4 0.053745
-3 0 -5, 5 -0.054860 -3 0
-2 2 -6, 6 -0.030332 -2 0.113710
-1 4/3 -7, 7 -0.004832 -1 0
0 1 otherwise 0 0 1 otherwise
0 1 1.099232
2 1.771901
3 0
4 -0.222855
5 -1.099232
6 0
7 0.005838
8 0.280050
9 0
10 0.000604
1 1 -0.005838
12 0
13 0
14 -0.000604
15 0
16 0.000018
17 0
18 0
19 0
20 -0.000018 otherwise
0
Table 1: Nyquist filter for 3:4 up conversion, bilinear (bil) and proposed filter (def) and Nyquist filter for subsequent 4:3 down conversion, undo bilinear (undo-bil), undo proposed (undo-def) 3:4 up conversion
Claims
1. A filtering process for achieving a transparent cascade on N:M up conversion and subsequent M:N down conversion, where the up and down conversion ratios N:M and M.N respectively are rational numbers and the integers N and M satisfy the condition 1<N<M, wherein the up conversion filter operates on a sampled signal smput and is chosen to take the form Sup(n) =Γêæ S╬╣nput(k) . g(Nn - Mk) and wherein a corresponding down conversion filter operates on the up converted signal sup and is chosen to take the form Sdown(n) =Γêæ Sup(k) . h(Mn - Nk); the pair (g,h) of up and down conversion filters being chosen so that Γêæ h(Mn -Nk) . g(Nk - Mm) is equal to unity if n = m and is otherwise equal to zero.
A process according to Claim 1 , wherein the up conversion filter is a N:M Nyquist filter with a symmetric impulse response and the down conversion filter is a M:N Nyquist filter with a non-symmetric impulse response.
A process according to Claim 1 , wherein the up conversion filter is a N.M Nyquist filter with a symmetric impulse response and the down conversion filter is a M:N Nyquist filter with a symmetric impulse response. - 10 -
A pair of respectively N.M up conversion and subsequent M:N down conversion filters, where the up and down conversion ratios N.M and M:N respectively are rational numbers and the integers N and M satisfy the condition 1<N<M, wherein the up conversion filter is adapted to operate on a sampled signal sinput and is chosen to take the form sup(n) =Γêæ Sjnput(k) . g(Nn - Mk) and wherein the down conversion filter is adapted to operate on the up converted signal sup and is chosen to take the form Sdo n(n) =Γêæ sup(k) . h(Mn - Nk); the pair (g,h) of up and down conversion filters being chosen so that Γêæ h(Mn -Nk) . g(Nk - Mm) is equal to unity if n = m and is otherwise equal to zero.
A M:N down conversion filter adapted to operate on the output Sup(n) of a N.M up conversion digital filter of the form Sup(n) =Γêæ Sinput(k) . g(Nn - Mk), where the up and down conversion ratios N:M and M:N respectively are rational numbers and the integers N and M satisfy the condition 1<N<M, wherein the down conversion filter is chosen to take the form Sdown(n) =Γêæ sup(k) . h(Mn - Nk); where Γêæ h(Mn - Nk) . g(Nk - Mm) is equal to unity if n = m and is otherwise equal to zero.
Priority Applications (4)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| AT99938023T ATE247339T1 (en) | 1998-03-06 | 1999-03-08 | CASCADE OF UPWARD CONVERSION AND DOWNWARD CONVERSION |
| US09/622,383 US6760379B1 (en) | 1998-03-06 | 1999-03-08 | Cascading of up conversion and down conversion |
| DE69910370T DE69910370T2 (en) | 1998-03-06 | 1999-03-08 | CASCADING UPWARDING AND DOWNWARDING |
| EP99938023A EP1074089B1 (en) | 1998-03-06 | 1999-03-08 | Cascading of up conversion and down conversion |
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| GB9804936.4 | 1998-03-06 | ||
| GB9804936A GB2335104B (en) | 1998-03-06 | 1998-03-06 | Cascading of up conversion and down conversion |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO1999045645A1 true WO1999045645A1 (en) | 1999-09-10 |
Family
ID=10828202
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/GB1999/000682 WO1999045645A1 (en) | 1998-03-06 | 1999-03-08 | Cascading of up conversion and down conversion |
Country Status (6)
| Country | Link |
|---|---|
| US (1) | US6760379B1 (en) |
| EP (1) | EP1074089B1 (en) |
| AT (1) | ATE247339T1 (en) |
| DE (1) | DE69910370T2 (en) |
| GB (1) | GB2335104B (en) |
| WO (1) | WO1999045645A1 (en) |
Families Citing this family (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB2460069A (en) * | 2008-05-15 | 2009-11-18 | Snell & Wilcox Ltd | Sampling conversion between formats in digital image processing |
| US10754813B1 (en) | 2011-06-30 | 2020-08-25 | Amazon Technologies, Inc. | Methods and apparatus for block storage I/O operations in a storage gateway |
| US8832039B1 (en) | 2011-06-30 | 2014-09-09 | Amazon Technologies, Inc. | Methods and apparatus for data restore and recovery from a remote data store |
| US9294564B2 (en) | 2011-06-30 | 2016-03-22 | Amazon Technologies, Inc. | Shadowing storage gateway |
| US8806588B2 (en) | 2011-06-30 | 2014-08-12 | Amazon Technologies, Inc. | Storage gateway activation process |
| US9258579B1 (en) | 2011-06-30 | 2016-02-09 | Sprint Communications Company L.P. | Temporal shift of object resolution and optimization |
| US8793343B1 (en) | 2011-08-18 | 2014-07-29 | Amazon Technologies, Inc. | Redundant storage gateways |
| US8789208B1 (en) | 2011-10-04 | 2014-07-22 | Amazon Technologies, Inc. | Methods and apparatus for controlling snapshot exports |
| US9635132B1 (en) | 2011-12-15 | 2017-04-25 | Amazon Technologies, Inc. | Service and APIs for remote volume-based block storage |
Family Cites Families (29)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5086488A (en) | 1989-08-19 | 1992-02-04 | Mitsubishi Denki Kabushiki Kaisha | Transform coding apparatus |
| US5142380A (en) | 1989-10-23 | 1992-08-25 | Ricoh Company, Ltd. | Image data processing apparatus |
| GB2242800B (en) * | 1990-04-03 | 1993-11-24 | Sony Corp | Digital phase detector arrangements |
| US5500739A (en) * | 1990-05-31 | 1996-03-19 | Samsung Electronics Co., Ltd. | Frequency-multiplexing FM luma signal with color and 2nd under signals having overlapping frequency spectra |
| NL9001441A (en) * | 1990-06-22 | 1992-01-16 | Philips Nv | SIGMA-DELTA MODULATOR. |
| US5532820A (en) * | 1990-08-17 | 1996-07-02 | Samsung Electronics Co., Ltd. | Digital modulators for use with sub-nyquist sampling of raster-scanned samples of image intensity |
| US5249053A (en) | 1991-02-05 | 1993-09-28 | Dycam Inc. | Filmless digital camera with selective image compression |
| DE4111884A1 (en) | 1991-04-09 | 1992-10-15 | Klippel Wolfgang | CIRCUIT ARRANGEMENT FOR CORRECTING THE LINEAR AND NON-LINEAR TRANSMISSION BEHAVIOR OF ELECTROACOUSTIC TRANSDUCERS |
| TW256010B (en) | 1991-04-18 | 1995-09-01 | Ampex | |
| JPH05283978A (en) * | 1992-03-31 | 1993-10-29 | Sony Corp | Sampling rate conversion device |
| JP3395311B2 (en) * | 1993-02-03 | 2003-04-14 | ソニー株式会社 | Bidirectional rate converter and imaging device |
| US5748245A (en) | 1993-03-29 | 1998-05-05 | Canon Kabushiki Kaisha | Encoding apparatus for encoding input information data while controlling a code quantity of encoded information data |
| US5479527A (en) | 1993-12-08 | 1995-12-26 | Industrial Technology Research Inst. | Variable length coding system |
| US5532755A (en) * | 1994-01-05 | 1996-07-02 | Samsung Electronics Co., Ltd. | Apparatus for suppressing ghosts in signals modulating a carrier in quadrature phasing with a video carrier |
| KR0183688B1 (en) | 1994-01-12 | 1999-05-01 | 김광호 | Image encoding method and device |
| US5512956A (en) | 1994-02-04 | 1996-04-30 | At&T Corp. | Adaptive spatial-temporal postprocessing for low bit-rate coded image sequences |
| US5671298A (en) | 1994-08-30 | 1997-09-23 | Texas Instruments Incorporated | Image scaling using cubic filters |
| JP3058028B2 (en) | 1994-10-31 | 2000-07-04 | 三菱電機株式会社 | Image encoded data re-encoding device |
| US5802218A (en) | 1994-11-04 | 1998-09-01 | Motorola, Inc. | Method, post-processing filter, and video compression system for suppressing mosquito and blocking atrifacts |
| US6005952A (en) | 1995-04-05 | 1999-12-21 | Klippel; Wolfgang | Active attenuation of nonlinear sound |
| US5991456A (en) | 1996-05-29 | 1999-11-23 | Science And Technology Corporation | Method of improving a digital image |
| WO1998026602A1 (en) | 1996-12-12 | 1998-06-18 | Sony Corporation | Equipment and method for compressing picture data |
| US6539120B1 (en) | 1997-03-12 | 2003-03-25 | Matsushita Electric Industrial Co., Ltd. | MPEG decoder providing multiple standard output signals |
| US6278735B1 (en) | 1998-03-19 | 2001-08-21 | International Business Machines Corporation | Real-time single pass variable bit rate control strategy and encoder |
| US6269120B1 (en) | 1998-03-23 | 2001-07-31 | International Business Machines Corporation | Method of precise buffer management for MPEG video splicing |
| GB9807201D0 (en) | 1998-04-03 | 1998-06-03 | Nds Ltd | Filtering video signals containing chrominance information |
| US6249549B1 (en) * | 1998-10-09 | 2001-06-19 | Matsushita Electric Industrial Co., Ltd. | Down conversion system using a pre-decimation filter |
| US6151362A (en) | 1998-10-30 | 2000-11-21 | Motorola, Inc. | Joint rate control for stereoscopic video coding |
| US6570922B1 (en) | 1998-11-24 | 2003-05-27 | General Instrument Corporation | Rate control for an MPEG transcoder without a priori knowledge of picture type |
-
1998
- 1998-03-06 GB GB9804936A patent/GB2335104B/en not_active Revoked
-
1999
- 1999-03-08 DE DE69910370T patent/DE69910370T2/en not_active Expired - Lifetime
- 1999-03-08 WO PCT/GB1999/000682 patent/WO1999045645A1/en active IP Right Grant
- 1999-03-08 US US09/622,383 patent/US6760379B1/en not_active Expired - Fee Related
- 1999-03-08 AT AT99938023T patent/ATE247339T1/en not_active IP Right Cessation
- 1999-03-08 EP EP99938023A patent/EP1074089B1/en not_active Expired - Lifetime
Non-Patent Citations (1)
| Title |
|---|
| BOCK A M: "DESIGN CRITERIA FOR VIDEO SAMPLING RATE CONVERSION FILTERS", ELECTRONICS LETTERS, vol. 26, no. 16, 2 August 1990 (1990-08-02), pages 1259/1260, XP000108280 * |
Also Published As
| Publication number | Publication date |
|---|---|
| DE69910370T2 (en) | 2004-07-01 |
| GB9804936D0 (en) | 1998-04-29 |
| GB2335104A (en) | 1999-09-08 |
| ATE247339T1 (en) | 2003-08-15 |
| DE69910370D1 (en) | 2003-09-18 |
| EP1074089A1 (en) | 2001-02-07 |
| EP1074089B1 (en) | 2003-08-13 |
| US6760379B1 (en) | 2004-07-06 |
| GB2335104B (en) | 2002-01-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US4674125A (en) | Real-time hierarchal pyramid signal processing apparatus | |
| US6356315B1 (en) | Sampling rate conversion using digital differential analyzers | |
| JP3160675B2 (en) | Sample data interpolator | |
| US5089893A (en) | Picture element number converter | |
| EP1611679B1 (en) | Up and down sample rate converter | |
| JPS61236210A (en) | Variable pass band filter apparatus | |
| WO1999045645A1 (en) | Cascading of up conversion and down conversion | |
| US8417066B2 (en) | Method and system for polyphase filtering by combining IIR and FIR filters and its applications in video scaling | |
| US7330865B2 (en) | Digital filter and its designing method | |
| US6310566B1 (en) | Digital data sample rate conversion system with delayed interpolation | |
| US6437827B1 (en) | Filtering video signals containing chrominance information | |
| JPH0693596B2 (en) | Filter and data transmission system using the same | |
| US7728743B2 (en) | Device and method for polyphase resampling | |
| KR0178003B1 (en) | Negative band digital filter bank design method | |
| KR100641741B1 (en) | A digital filter and a method for filtering sample data | |
| JPS63245085A (en) | Subsample interpolating method | |
| GB2143046A (en) | Real-time hierarchal signal processing apparatus | |
| US7123652B1 (en) | Sampled data digital filtering system | |
| KR20030043665A (en) | Digital sample frequency converter | |
| JPH0783235B2 (en) | Signal processor | |
| US5440593A (en) | Combined aligner blender | |
| KR970008103B1 (en) | 2-dimensional digital filter | |
| KR930009147B1 (en) | Quadruture mirror filter | |
| EP1078513B1 (en) | Video signal processing | |
| KR930006705B1 (en) | Quadrature mirror filter of linear phase |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| AK | Designated states |
Kind code of ref document: A1 Designated state(s): US |
|
| AL | Designated countries for regional patents |
Kind code of ref document: A1 Designated state(s): AT BE CH CY DE DK ES FI FR GB GR IE IT LU MC NL PT SE |
|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application | ||
| DFPE | Request for preliminary examination filed prior to expiration of 19th month from priority date (pct application filed before 20040101) | ||
| WWE | Wipo information: entry into national phase |
Ref document number: 1999938023 Country of ref document: EP |
|
| WWE | Wipo information: entry into national phase |
Ref document number: 09622383 Country of ref document: US |
|
| WWP | Wipo information: published in national office |
Ref document number: 1999938023 Country of ref document: EP |
|
| WWG | Wipo information: grant in national office |
Ref document number: 1999938023 Country of ref document: EP |