JPH0132715B2 - - Google Patents

Description

[Detailed description of the invention]

Industrial Application Fields A composite video signal is converted into a luminance signal (hereinafter referred to as Y signal) and a color signal (hereinafter referred to as C signal) within a device that converts a color television signal into a high-quality and high-definition image through digital processing. (referred to as signal). Conventional technology YC separation of color television signals has conventionally been performed using frequency separation methods, line delay lines, etc.
Comb separation method (hereinafter referred to as inter-line YC separation)
was used. Furthermore, as real-time digital processing of images became possible, YC separation using frame memory and interframe correlation of signals (hereinafter referred to as interframe YC separation) began to be performed. However, the interframe YC separation is 1
If there is movement in an image after a period of time corresponding to a frame has elapsed, inter-frame correlation becomes small, resulting in significant deterioration of image quality. Therefore, by detecting the movement of the image,
A method has been proposed in which the inter-line YC separation and the inter-frame YC separation are weighted and combined based on this motion information. Figure 3 shows an example of YC separation according to the prior art.The blocks numbered 110 are circuit blocks with a function used for YC separation between frames, and the blocks numbered 120 are used for YC separation between lines.
Circuit block with function used for separation, 13
The number 0 is a circuit block having a function used for motion detection. First, the inter-frame YC separation system will be explained. The composite video signal (NTSC signal) input from the input terminal of 111 has a line-to-line YC separation system.
Since it has a 1H delay, it passes through a line memory 112 to compensate for the delay in YC separation between frames, and a frame memory 113 obtains two signals 113a and 113b with a time difference of one frame. Since the NTSC signal differs in color subcarrier position by 180 degrees between frames, the subtracter 114
Then, the difference between the two signals 113a and 113b is obtained, and the signal level is reduced to 1/2 by the subtracter 115.
Components other than the color subcarrier frequency component are removed through a bandpass filter 116 (hereinafter referred to as BPF) having characteristics as shown in FIG.
Get 7C. The luminance signal 117Y is the original signal 11
3a by a subtracter 117. Next, the line-to-line YC separation system will be explained. Similar to the interframe YC separation, the video signal input from the input terminal 111 is taken out as information spanning three lines by the line memory 121 arranged in series. The information on these three lines is reduced to -1/4 by the attenuator 122 at the first and third taps, and then
The tap is halved by the attenuator 123. These three signals are added by an adder 124, passed through a BPF 125 having the same characteristics as the BPF 116, and extracted as a color signal 126C. Further, this color signal 126C is subtracted from the original signal by a subtracter 126, and is extracted as a luminance signal 126Y. The YC signals separated as described above are classified as shown in the table below.

[Table] Each YC signal is weighted and combined using the motion information of the image described above. motion detector 1
31 detects the movement of the image based on the interframe difference signal, and generates a coefficient k and a 1-k generator 132.
, generates a coefficient according to the movement. That is, when there is no movement, k=1 is input to the multipliers 118 and 127. When k=1, 1-k=0, so 117Y and 117C are directly output as outputs Y and C. Also, when the movement is large, set k = 0,
Using the same concept as above, outputs Y and C have 126
Y, 126C is directly output. When the motion is not very large, an appropriate value is determined in the range of 0<k<1 depending on the degree of motion, and the adder 11 weights the separation signal between lines and the separation signal between frames.
9,128 are added and combined. As mentioned above, there are multiple types of images depending on the movement of the image.
Conventionally, a method of weighted synthesis of YC separation has been commonly used. However, this method is only used for the inter-frame YC separation function, even though the frame memory has a relatively large capacity. Problems to be solved by the invention When digitally processing video signals in this way,
The amount of information is much larger than that of alphanumeric characters, symbols, etc., and when the video is accompanied by movement, the amount of information must be several times larger. In particular, this type of device uses a semiconductor memory, and the cost of this semiconductor memory is determined by the unit of information storage of one bit. Therefore, one of the important issues is how to reduce the memory capacity of such digital circuits. Means for Solving the Problems The present invention utilizes the conventional inefficient frame memory to provide a noise reduction function in order to improve the S/N ratio, and in particular, to provide independent YC separation between lines. The apparatus is simplified by using a color subcarrier phase inverter (hereinafter referred to as a chroma inverter), which is indispensable as a noise reducer. Embodiment FIG. 1 shows the best mode for carrying out the present invention. Conventional 1 frame memory (=525H memory) consists of 524H memory 311 and two 1H memories 3
Replaced by 15. Subtracters 312, 314 and coefficient unit 313 are parts added for noise reducer, 1H memory 315, attenuators 316, 317, and adder 3.
18, BPF319, amplifier 320 and subtractor 32
Step 1 forms a chromine inverter using YC separation between lines. From the input NTSC signal, a subtracter 312 subtracts the initial state value of a coefficient multiplier 313 having the characteristics shown in FIG. The initial state may have any value. This is because when the input completes one cycle (that is, when it is transmitted to 321a), the next data is entered. Now, assuming that the output of the coefficient multiplier is 0, the input NTSC signal is output as is from the output of the subtracter 312. This signal is sent to the 524H memory 311,
524H delay, 1H memory 3 for chroma inverter
15, obtain a 3-tap signal and apply YC between the lines.
Similar to separation, separate color signals. This color signal is output 321a from the subtracter 321, and in order to obtain a signal with only the color signal phase inverted,
Multiplyed and subtracted from the original signal. Both inputs of the subtracter 314 are signals separated by one frame and having the same color signal phase. At this time, the output of the subtracter 314 is
Corresponding to the inter-frame difference signal 314a, noise generated only in a certain frame on a still screen, that is, noise without frame correlation, becomes a difference signal and appears in 314a. In general, noise is considered to have a somewhat low level, and therefore corresponds to the region of the characteristics shown in FIG.
is canceled out. Furthermore, on the video screen, the interframe difference signal 314a has a considerably high level,
This is considered to correspond to the region of the characteristic shown in FIG. 4, and at this time, the output of the coefficient multiplier 313 becomes 0, and the information from one frame before is completely erased and will not affect the subsequent frame. The above noise reducer is effective in improving the S/N, and the luminance that has been subjected to the subsequent YC separation processing,
A color signal with less noise can be obtained. By the way, since the present invention uses YC separation between lines as a chromine inverter, this is
It can be used for YC separation between lines when performing weighted synthesis by adapting the inter-frame YC separation and inter-line YC separation to motion, as described in the prior art.
In other words, the output of the BPF 319 can be considered as a color signal of line-to-line YC separation, and can be subtracted from the original composite signal by the subtracter 322 to obtain a luminance signal. Also, the interframe YC separation is
524H memory 311 input and line memory 31
By tapping for 1H in the middle of step 5, a signal delayed by one frame time is obtained, so the subtracter 32
3. Using the BPF 324 and another subtractor 325, the luminance due to the inter-frame YC separation,
Color signals can be obtained. Regarding image motion detection, there are two methods: the conventional method of detecting by removing the color signal from the difference signal Y between frames, and the method of simply detecting from the difference signal X because the color phase is inverted by a chroma inverter. , it is also possible to use both as appropriate. Here, the motion detector 330 coefficient generation 331
It has the same function as the one shown. Also, the coefficients generated by motion detection are sent to the multiplier 3 as in FIG.
k and 1-k are multiplied by 26 and 327 and combined by adders 328 and 329. Effects of the Invention As described above, according to the present invention, it is possible to perform YC separation between lines using the subtracters 322 and 323 from the signal generated by the 1H memory 315 forming the chroma inverter. The number of memory is reduced compared to the previous one. Also,
524H memory 311 is for NTSC composite signal.
Since it is connected before the 1H memory 315 forming the chroma inverter, it is possible to perform interframe YC separation more completely.

[Brief explanation of drawings]

FIG. 1 is a main circuit diagram showing an embodiment of the present invention;
Fig. 2 is a frequency vs. throughput characteristic diagram showing an example of the characteristics of a bandpass filter, Fig. 3 is a main circuit diagram showing an example of a conventional circuit of this type, and Fig. 4 is an input signal in a composite video signal receiving system. It is an output characteristic diagram. 311...524 line memory, 315...2 line memory, 318...adder, 319...first
band wave generator, 320... amplifier, 321... first subtractor, 322... second subtractor, 323...
. . . third subtractor, 324 . . . second band wave generator.

Claims (1)

[Scope of Claims] 1. A memory having a capacity for (n-1) lines for extracting the composite video signal equivalent to one frame before the current composite video signal, which consists of one frame with n lines; After multiplying each of the signals extracted from the 2-line memory to which the previous composite video signal is given, the input point of this 2-line memory, the 1-line delay point, and the 2-line delay point by a predetermined coefficient, a first bandpass waver that extracts a color signal having a line correlation from the output signal of the adder; and an amplifier that doubles the signal drawn from the first bandpass waveform generator. By subtracting the output signal of this amplifier and the signal extracted from the 1-line delay point of the 2-line memory, a composite video signal in which the phase of the color signal is inverted is obtained using line correlation. a first subtracter that generates a subtracter, and a line correlation is utilized by subtracting a signal extracted from the first bandpass filter and a signal extracted from a 1-line delay point of the 2-line memory. a second subtracter that generates a luminance signal; a signal extracted from the one-line delay point;
a third subtractor that generates a color signal using frame correlation by subtracting the current composite video signal provided to the memory having a capacity for the (n-1) lines; a second bandpass filter that extracts a color signal using strict frame correlation from the output signal of the third subtracter; A color subcarrier phase inverter is formed by a subtracter, and the composite video signal of the previous frame generated by the first subtractor is subtracted from the current composite video signal, and the subtracted signal is An output signal obtained by subtracting the signal obtained by multiplying an appropriate coefficient and the current composite video signal again is the composite video signal to be applied to the memory having a capacity for the (n-1) lines. A YC separation circuit for a television signal, characterized in that it forms a cyclic filter having a noise reduction function on a frame-by-frame basis.