Television – Image signal processing circuitry specific to television – Noise or undesired signal reduction
Patent
1993-06-22
1997-05-13
Peng, John K.
Television
Image signal processing circuitry specific to television
Noise or undesired signal reduction
348609, 348612, H04N 521, H04N 5217
Patent
active
056297424
DESCRIPTION:
BRIEF SUMMARY
BACKGROUND OF THE INVENTION
1. Field of the Invention
When PAL or other similar subcarrier based video signals are transmitted through a satellite or other f.m. channel, the resulting chrominance signal to noise ratio is generally less than the luminance signal to noise ratio. This arises because of the triangular noise spectrum which results in high noise density around the colour subcarrier frequency (see FIG. 1).
This invention relates to a method for applying non linear pre-emphasis and de-emphasis in such a way that both the luminance and chrominance signal to noise ratios can be improved. The sensitivity of the signal to interference is also improved by the application of this technique.
2. Prior Art
Non linear pre/de-emphasis has been developed for MAC packet family signals. See, for example the E7 technique described in EBU Technical Document 3258. The signal is split into two frequency bands using high pass and low pass filters. The high frequency components are subjected to a non linear processing and then re-combined with the un-modified low frequency components. FIG. 2a shows de-emphasis circuit configurations which require only one filter. The complementary filter is produced by subtraction of the low pass or high pass filter output from the input signal. The non linearity, N, is generally of the form shown in FIG. 2b, which may be expressed by the following equation:
When the complementary non-linear pre-emphasis technique (using similar filters F.sub.1 or F.sub.2, and a non-linear function N.sup.-1) is used on a signal, the effect is to boost the level of low level signals occurring at high frequencies, leaving high level signals at these frequencies substantially unaffected. If the pre-emphasised signal passes through an f.m. channel noise will disproportionately affect the high frequencies in the signal, as mentioned above. When the signal is subsequently de-emphasised the noise affecting low level signals at high frequency will be reduced in level along with the low level signals. The noise affecting high level signals at high frequency is not altered.
If such a technique is applied to a signal including a subcarrier, such as a PAL signal, the absence of a noise improvement for high level high frequency signals is a significant problem for the signal modulating the subcarrier, e.g. the chrominance signal. This is because the noise affecting high level high frequency luminance is masked by the high frequency variation of the signal itself, whereas the chrominance signal is at high frequencies because of its modulation on a subcarrier, rather than because of intrinsic high frequency variation occurring in the chrominance, and so when colour saturation is high (i.e. for high level chrominance) there will not be a masking effect to hide the noise.
A second problem is that the effect of the pre-emphasis network would be to produce large amounts of differential gain, i.e. the gain would vary depending on the chrominance subcarrier level. Conventional receivers cannot adjust to gain changes which are level dependent and so a subcarrier based signal pre-emphasised in the known manner would not be compatible with conventional receivers. Receivers not equipped with the de-emphasis circuit would display a seriously impaired picture.
One way of overcoming these problems would be to apply non-linear pre-emphasis to the YUV components before PAL coding. The de-emphasis would be applied to the YUV components after de-coding. This has the problem that three circuits are required in the receiver. Additionally, these signals may not be available in the receiver. It is the object of this invention to provide a non linear de-emphasis which can be implemented using a single circuit applied to a composite waveform, e.g. PAL signal, in such a way that it can provide both luminance and chrominance noise improvement, preferably in a compatible way so that existing receivers not employing the de-emphasis circuitry will produce an essentially unimpaired picture.
SUMMARY OF THE INVENTION
The present invention pr
REFERENCES:
patent: 3286024 (1966-11-01), Melchlor
patent: 3365541 (1968-01-01), Melchlor
patent: 4684983 (1987-08-01), Acampora
patent: 4922331 (1990-05-01), Ezaki
patent: 5027210 (1991-06-01), Beech
patent: 5225910 (1993-07-01), Sugimori et al.
Miller John W.
National Transcommunications, Ltd.
Peng John K.
LandOfFree
Compatible non linear pre-emphasis for composite video signals does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Compatible non linear pre-emphasis for composite video signals, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Compatible non linear pre-emphasis for composite video signals will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-1389006