JPH01114190A - Video signal processor - Google Patents

Abstract

PURPOSE:To attain low cost by allowing a spatial filter to act like a pre-filter for luminance signal and chrominance signal so as to reduce the hardware of the system applying offset sampling. CONSTITUTION:An adder 113 adds outputs of 3HDL 108 and 1HDL 109 as to a luminance signal Y in a spatial filter comprising a delay line, a coefficient multiplier and adders 107-125 and the output is the sum of a signal having a time difference of + or -1H relatively to the output of the 2HDL 110. Thus, the output of the DL 114 has a time difference of + or -1H relatively to the output of the DL 120 and a coefficient a1 is multiplied with upper/lower picture elements of the center picture element and a coefficient a2 is multiplied with forward/backward picture elements respectively. A coefficient a0 is multiplied with the center picture element and a coefficient a1 is multiplied with the forward/backward picture element. Thus, a desired filtering processing is attained by deciding the coefficient added to 9 picture elements. As to a color difference signal Cw, the output of 4HDL 107 and switch 106 is added and the process is similar to the signal Y except 2H of time difference. The filter output is sampled by the switch 126 and then outputted.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a video signal processing device, and more particularly to a signal processing device for offset subsampling a color video signal.

[Conventional technology]

Generally, a video signal has a large amount of information per unit time, and when a video signal is digitized, the amount of data per unit time becomes enormous. For example, an NTSC television signal of 4
In the case of an 8-bit digital signal with a sampling frequency of fsc (fsc is color subcarrier frequency), the transmission bit rate of this digital signal is approximately 120 Mbps. Further, when the above-mentioned NTSC television signal is converted into an 8-bit digital signal with three systems of R, G, and B signals each having a sampling frequency of 3 fsc, the transmission bit rate of this digital signal is approximately 260 Mbps.

On the other hand, in recent years, research has been conducted on high-definition television signals with approximately twice the number of horizontal scanning lines as conventional television signals, but when this high-definition television signal is digitized, the transmission bit rate is extremely high. For example, if a high-definition television signal is made into three systems of R, G, and B signals, each of which is an 8-bit digital signal with a sampling frequency of 64 MHz, the transmission bit rate is 1.5.
Gbps.

In recent years, in order to improve transmission efficiency, so-called TCI has been implemented, which compresses the time axis of the chrominance signal relative to the luminance signal and time-division multiplexes the chrominance signal and the luminance signal. Even if a high-definition television is transmitted as an 8-bit digital signal at a sampling frequency of 64 MHz, the transmission bit rate will be approximately 510 Mbps, which is a high bit rate.

When a video signal is digitized in this way, it becomes a signal with a high bit rate, but if you try to transmit this digital signal using a communication system, a large-capacity line is required, resulting in problems such as a lack of line capacity or increased transmission costs. bring about a situation. Furthermore, when attempting to record such a high bit rate digital signal on a recording medium such as a magnetic recording medium, there are problems in that the recording head cannot handle it or the recording time becomes extremely short.

Therefore, various band compression techniques have been proposed in order to generally reduce the data amount of digital signals. As one method of band compression, so-called offset subsampling can be considered. Offset subsampling is a method in which diagonal components of an image that are not very important in terms of visual characteristics are removed in advance using a spatial filter, and high-frequency components are placed in the removed frequency region to lower the sampling frequency. Such offset subsampling includes line offset subsampling (Loss) in which the sampling position is shifted between adjacent scanning lines in the same field, and sampling position shifted between mutually adjacent scanning lines in two adjacent fields. field offset subsampling (FO3S)
), but in any case, two-dimensional band limiting must be performed using the aforementioned spatial filter as a pre-filter before performing offset sampling.

Furthermore, when decoding a digital signal subjected to offset subsampling, spatial interpolation must be performed using a spatial filter as a post filter.

[Problem that the invention seeks to solve]

By the way, when subsampling a color video signal, each component signal, for example, a luminance signal and two types of color difference signals, is subsampled. Therefore, the spatial filter used as the bris filter and post filter is applied to each of these components. It is necessary to provide

This spatial filter generally consists of multiple filters connected in series.
It is composed of a horizontal scanning period (H) delay line, a plurality of multipliers for multiplying coefficients, etc., and has a relatively large amount of hardware. Therefore, when offset subsampling is conventionally performed as a method for band compression, a plurality of spatial filters are required, which increases the amount of hardware and is disadvantageous in terms of cost.

SUMMARY OF THE INVENTION In view of the above problems, it is an object of the present invention to provide a video signal processing device that reduces the amount of hardware needed to perform offset subsampling of a color video signal and enables cost reduction.

[Means for solving problems]

For this purpose, in the video signal processing device of the present invention, a signal obtained by time-axis multiplexing a luminance signal and a chrominance signal in units of horizontal scanning lines is supplied to a filter capable of two-dimensional band limiting, The configuration is such that the signal band-limited by the filter is subjected to offset subsampling.

[Effect]

By configuring as described above, one spatial filter functions as a bristle filter for luminance signals and color signals, and the number of spatial filters can be reduced. This makes it possible to reduce the amount of hardware required for the system that performs offset subsampling.

〔Example〕

An embodiment of the present invention will be described below. FIG. 1 is a diagram showing the configuration of a processing device as an embodiment of the present invention, which handles the above-mentioned high-definition television signal.

In the figure, 101, 102, and 103 are terminals into which the luminance signal (Y) and two color difference signals (CN, CW) of the high-definition television signal are input, and the input Y is a 64 MHz, 8-bit digital signal, C, , Cw are 16 MHz, 8-bit digital signals. 104 is a switch whose connection is changed for each IH, and the timing controller 10
Controlled by 0. The output signal of the switch 104 becomes a line-sequential color difference signal in which CN and CW are present alternately for each IH, and is supplied to a time axis compressor 105, where the time axis is compressed to 1/4. The circuit 105 is composed of a line memory, for example, and its write clock is 16 MHz, and its successive clock is 6 MHz.
It is 4MHz.

This successive output timing is set to a period of 1/4H corresponding to the blanking period of Y.

The switch 106 is a switch that selectively outputs the line-sequential color difference signal and Y whose time axis has been compressed to 1/4.
During the 5H period, Y is output, and during the remaining period, a time-axis compressed line-sequential color difference signal is output. Here, the signal output from this switch 106 is a 64MHz, 8-bit digital TCI.
It becomes a signal.

107 is 4H delay line (DL), 108 is 3HDL, 10
9 is IHDL, 110 is 2HDL, and switch 11
1 and 112 are switches connected to the Y side when the switch 106 is connected to the Y side, and to the C side when the switch 106 is connected to the C side, and the timing controller 100 connects the switch 106 111 and 112 to the IH So the ratio of 4=1 is on the Y side.

Connect it to the C side.

113 is an adder, and for Y, I HD Li2
O and 3 HDL110 output, 4 for CN/CW
The output of WDL'107 and the output of switch 106 are respectively added.

114 is D that delays the output of the adder 113 by one data period.
L, 115 is a DL that further delays the output of DL 114 by one data period, 116, 117 is a coefficient unit that multiplies the coefficient al+a2, 118 is an adder that adds the output of adder 113 and the output of DL 115, and 119 is a coefficient This is an adder that adds the outputs of the coefficient multipliers 116 and 117 and the output of a coefficient multiplier 123, which will be described later. Further, 120 and 121 are data period delay lines, and 122
is an adder, 123 is a coefficient unit that multiplies the output of the adder 122 by a coefficient a1, and 124 is a coefficient a for the output of the DLL 20. A coefficient unit 125 multiplies the output of the adder 119 and the coefficient unit 12
This is an adder that adds up the outputs of 4.

Hereinafter, the operation of the spatial filter composed of the above-described elements 107 to 125 will be explained using FIGS. 2(A) and 2(B).

In this filter, the data of the center pixel is DLIIo,
120 and is delayed by (2H+d). Therefore, from the above configuration, the coefficients assigned to the nine pixels centered on the center pixel are as shown in FIG. 2(A) for Y and as shown in FIG. 2(B) for CW (CN). In the figure, the center pixel is double-circled.

First, Y will be explained. Adder 113 uses 3HDL
108. The output of the IHDL 109 is added, and this output is the sum of signals having a time difference of ±IH relative to the output of the 2HDL 110. Therefore, the output of the DL 114 also has a relative time difference of ±IH with respect to the output of the DL 120, and the pixels on the upper and lower lines of the center pixel are multiplied by the coefficient a, and the pixels before and after the center pixel are multiplied by the adder 118. The output of is multiplied by coefficient a2 in coefficient unit 117. Further, the pixels before and after the center pixel on the same line are multiplied by a coefficient a1 in a coefficient unit 123 as an output of the adder 122. For the center pixel, the coefficient a is obtained by the coefficient unit 124. is applied. As a result, coefficients as shown in FIG. 2(A) are assigned to 9 pixels of Y, and a6. Desired filtering processing can be performed by appropriately determining al+82.

Also, for CW, the adder 113 outputs 4HDL 10
7. The outputs of switch 106 are added and these are 2HD
It is the same as Y except that it has a time difference of ±2H relative to the output of L, and as a result, coefficients are assigned as shown in FIG. 2(B). That is, the limited band in the vertical direction of the pixel is 1/2 of that in the case of Y. This is because Cw and CN exist in every IH.

The digital signal that has passed through the filter is subsampled by a switch 126 to reduce the amount of data and is output from a terminal 127. This subsampling is one line offset subsampling for Y, and C
Since N and Cw are line sequential signals, 2 line offset subsampling is performed. FIG. 3(A) schematically shows sampling points for subsampling for Y, and FIG. 3(B) schematically shows sampling points for subsampling for C, where ◯ indicates a transmission pixel and × indicates a non-transmission pixel. As is clear from the figure, the vertical sampling frequency of CN and CW is 1 compared to that of Y.
/2, and accordingly, the vertical limit band of the above-mentioned spatial filter is also 1/2.

According to the circuit configured as described above, by supplying a TCI-converted digital signal to one spatial filter, this spatial filter can perform two-dimensional band limiting of the luminance signal and chrominance signal. We were able to reduce the amount. In addition, for color signals, line-sequential color difference signals are subjected to two-line offset subsampling, but this is handled by changing the vertical limit band of the spatial filter, and the amount of hardware can be kept to a minimum. .

Although the above embodiment deals with TCI signals in which Y and Cw/CN are time-division multiplexed at a ratio of 4:1, other signal formats are also possible; for example, y, cW for each IH is used. ,
It is also possible to adopt a configuration in which signals time-division multiplexed at a ratio of cN of 4:1:1 are input to one spatial filter.

〔Effect of the invention〕

As described above, according to the present invention, it is possible to obtain a video signal processing device that can reduce the amount of hardware and reduce costs when performing offset subsampling on a color video signal.

[Brief explanation of the drawing]

FIG. 1 is a diagram showing a processing device as an embodiment of the present invention, FIG. 2 is a diagram for explaining the operation of the spatial filter in FIG. 1, and FIG. FIG. 3 is a diagram for explaining sampling. In the figure, 101 is a luminance signal input terminal, 102 and 103 are color difference signal input terminals, 104 is a line selection switch, and 10
5 is ``time axis compression circuit, 106゜111.112 is Y/
C changeover switch, 107°108.109,110,1
14,115°120,121 are delay lines, 116,
117゜123.124 are coefficient multipliers, 113, 118 respectively
119, 122 and 125 are adders, and 126 is a switch for subsampling.

Claims (2)

[Claims] (1) A signal obtained by time-axis multiplexing a luminance signal and a chrominance signal in units of horizontal scanning lines is supplied to a filter capable of two-dimensional band limiting, and the signal band limited by the filter is subjected to offset subsampling. A video signal processing device characterized by: (2) The color signal is a line-sequential color difference signal, and the filter switches the limiting band in the vertical direction of the image depending on whether the input signal is a luminance signal or a line-sequential color difference signal. The video signal processing device according to item (1).