SU1646074A1 - Device for digital time-and-spatial filtering of tv picture signal - Google Patents

Abstract

The invention relates to television. The purpose of the invention is to improve the accuracy of the reconstructed television (TV) image signal. The device comprises a block-by-line filtering unit 1, a control unit 2 and a block-by-frame filtering unit 3. Block 2 controls the operation of blocks 1 and 3 according to an algorithm that takes into account the spectral components of the used TV sampling sampling structure. In blocks 1 and 3, multidimensional filtering of the TV image signal is performed, and its accuracy is increased, thereby increasing the clarity of the reproduced TV image. Device on PP. 2 and 3 f-ly is characterized by the execution of blocks 1 and 3. 2 h. n. f-ly, 3 ill.

Description

The invention relates to the field of television, and can be used, for example, as a correction device for a television (TV) image signal.

The purpose of the invention is to improve the accuracy of the reconstructed television image signal.

FIG. 1 shows a block diagram of the device; in fig. 2 - pulse characteristics of the base cell (unit) of the device; in fig. 3 illustrates the formation of a spatially filtering aperture and an example filtering procedure.

The device contains (Fig. 1) block 1 of line filtering, block 2 of control, block 3 of frame filtering, while block of line filtering contains adder 4, first and second blocks 5 and 6 of delay and subtractor 7, and block of frame filtering contains first and second multipliers 8 and 9, first and second blocks 10 and 11 delays, first, second and third keys

12-14, the first and second subtractors 15 and 16 and the adder 17.

The device works as follows

The input of the adder 4 (Fig. 1) receives a TV-signal of video information, sampled according to the applicable TV-scan standard, which in block 1 is subjected to a convolution operation with the impulse response (IH) of the block (line aperture), the formation of which when fed to the input device of a single pulse (IT), is shown graphically in FIG. 2a (where X (pT) is the single data input. Y (nT) is the impulse response, K is the switching of signals. The two-dimensional impulse responses (apertures) are shown in Fig. 26. The formation of the row aperture (packs IT) is due to the summation in the adder 4 input IT and delayed in block 5 for one clock cycle input IT, and the limitation of this pack by the number of pulses

Yo

C N O O

Vj

four

takes place in subtractor 7, where subtraction is made from M-ro IT, arriving at the first input of subtractor 7, IT, which is delayed in block 6 for M cycles.

Block 2 provides the constant coefficients (t and M) preprogrammed in the RPZU synchronously with the TV scan, which phase the address counters and select the write-read RAM modes in blocks b, 5 and thereby set the delay times, respectively (t Te + to Tc), and M (t.Te + k Tc), forming two-dimensional line-apertures of various types (Fig. 3), adaptive to the processing of the input TV signal. At the same time, the spectra of the vertical-temporal lattice of samples at the input and output of the filter Si, S2 (o, w and y) are related by the relation

S2-Si Hi.H2,

where Hi, H2 is the filter transfer characteristics are determined

()one,

(2 -),

2n 2n at the same time with, SU at spatial frequencies.

The convolution operation of the input TV signal thus formed by THEIR of Block 1 is equivalent to the interval integration operation (or the filtering procedure — a sliding average, or a comb filtering).

The operation of such a filter is described by the corresponding difference equation:

Yi (nf) aX (pT) - bX (pT MT) + cYi (nT-T) (1) where T is the horizontal TV sampling period;

but. B, c, n, and M are constant coefficients;

X (nT) input data sequence;

Y (nT) is the output data sequence.

With different values of constant coefficients (a, b, c), it is possible to obtain various forms and durations of FIR in a given filtration direction, and the phase response (PC) remains linear.

Formation of a symmetric triangular EM for the low-pass filter (Fig. 2a, b) from the IT package is performed in a subsequent block 3 (frame recyclers). From the sixth output of block 2, the first, second and third keys 2, 13 and 14 receive a control signal, the zero value of which closes the second key 13, the first key 12 sets

in the upper position (the second input is connected to the output), and the third key 14 is in the lower position (the first input is connected to the output).

With such switching the block structure

3 is similar to the structure of a block 1 connected in series with it. The right shoulder of a triangular symmetric EI is formed using an adder 17, to the second input

which the bundle EI comes in, formed in block 1, and on the first input this same bundle, but delayed by one clock in block 10. As a result, a pulse signal with stepwise amplitude (A) equal to is formed at the output of adder 17:

An + 1 2 (An + 1),

where ao 0, n 0, 1,2, ...

The left shoulder of the triangular symmetric THEM is formed in the subtractor 15, the first input of which is fed from the output of the adder 17, formed into M cycles of THEIR right shoulder, and the second input is the same THEIR, but

delayed by M cycles in block 11 (in this particular case, we assume N M). As a result of the subtraction, at the output of the subtractor 15 a triangular symmetric signal is generated in a given spatial direction, which is fed to the first input of the third key 14 and at the switching mode specified above, the third key 14 to the output of the device. The resulting IC can be described by the corresponding difference equation characteristic of interval integration with a linear phase response.

Y2 (nT) X (pT) - 2X (pT - MT) + X (pT - 2MT) + + 2Y2 (nT - T) - Y2 (nT - 2T) (2).

The convolution of the input TV signal with the IC (2) is equivalent to the procedure of quasi-optimal filtering of low frequencies in a given spatial direction with a linear phase characteristic.

If from the sixth output of block 2 the signal of the logical unit arrives at the corresponding inputs of the first, second and third keys 12, 13,14, then the next switch is made in block 3 of the second key 13 - the key is open, the first key 12 - the key is in the lower position. 16 - a key in the upper position. As a result of such switching, in block 3 of frame filtering a structure is formed,

allowing to create antisymmetric bipolar THEIR (Fig. 2 b).

The generated signal THEIR, taken from the output of the adder 17 in the form of the right shoulder of the triangular IT, similar to the previous switching mode, comes

to the first input of the subtractor 15, to the second input of which enters multiplied in the second multiplier 9 by a constant number M (again, in this particular case, we consider N - M) UI, which is delayed in block 6 of M clocks. As a result of the subtraction, the number of pulses in the bundle of the right arm of the triangular EM on the Mth cycle is limited. From the output of the first subtractor 15, the received signal is fed to the first input of the subtractor 16, the second input of which is multiplied in the multiplier 8 by

/ M + 1 hr a constant number g - “-) signal of a pack

EI from the output of block 1. As a result, in the second subtractor 16, a signal is generated at its output with an antisymmetric bipolar wave (FIG. 26) in a given spatial direction, which goes to the second output of key 14 and at the switching mode specified above to output devices.

The resulting antisymmetric bipolar IH can be described by the corresponding difference equation characteristic of interval differentiation:

Ouse (PT) - 1 M X (PT) - X (PT - MT - T) +

+ (PT-T) -2X (PT-MT) +

+ 2Y3 (nT-T) -Y2 (nT-2T)

(3)

The convolution of the input TV signal of the image with this type of IC is equivalent to the filtering procedure of the RF in a given spatial direction with a linear phase response.

Thus, the control unit 2 is programmed in a certain way, taking into account the analysis of the spectral components of the used TV-raster discretization structure, controlling the operation of blocks 1 and 3 by line-by-line and frame-by-frame filtering (or their linear combination), allows for multi-dimensional filtering of the TV signal image, aimed at improving its accuracy, thereby increasing the clarity of the reproduced image.

Claims (3)

1. A digital space-time filtering device for a television image signal comprising a block-by-line filtering unit, the first input of which is an input for a digital image television signal, and a control unit whose input is 5 10 15 0 five 0 five 0 50 five the input signal for the reference frequency, wherein the first and second inputs of the control unit are connected respectively to the second and third inputs of the line filtering unit, characterized in that, in order to improve the accuracy of the reconstructed television image signal, a frame-by-frame filtering unit is inserted into it, the output of the device, the first, second, third, fourth, fifth and sixth inputs of the block frame filtering are connected respectively to the first, second outputs of the block line filtering, third, fourth m, fifth and sixth outputs of the control unit. 2. The device of claim 1, wherein the line filtering unit comprises a sequentially connected adder and a subtractor, the output of which is the first output of the line filtering unit, the first, second and third inputs of which are, respectively, the first input of the adder and the inputs set the delay time of the first and second delay blocks, the output of the adder is connected to the combined signal inputs of the first and second delay blocks, the outputs of which are connected respectively to the second inputs of the adder and the subtractor, When the output of the subtractor is the second input progressive infiltration block 3. Device pop. 1, which indicates that the block-by-frame filtering unit contains the first and second multipliers, the first inputs of which are respectively the first and second inputs of the frame-based filtering unit, the third, fourth, fifth and sixth inputs of which are respectively the second inputs of the first and second multipliers , the input of setting the delay time of the first delay unit and the control input of the first key, which is combined with the control inputs of the second and third keys, as well as the series-connected adder, the first and second subtractors The outputs of the first and second calibrators are connected respectively to the signal inputs of the third key, the output of which is the output of the block-by-frame filtering unit, while the second inputs of the first and second subtractors are connected respectively to the outputs of the first and second keys, the outputs of the first, second smart and output The second delay unit is connected respectively to the signal input of the second key, the first and second signal inputs of the first key, the output of the adder is connected to the combined signal inputs of the first and second blocks delays, output of the first block the delay is connected to the first input; the sum input of the second multiplier is combined with the input, the second input of which is combined with the setting time for the second input of the second multiplier, the second delay unit. Them. S. 1LLLJ-ILLLJJ iilib vy.bl. l 8nl.l.15 ъ .1т T Exercise H2 IIJ , КЬ1 5 H-5: ()} M, ( -o - oo oo o -Ltd c-o-o-oo-oFie .1 -o - oo oo o -Ltd ts-o-oo oo Phage.Z