JPS6142910B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention provides a method for detecting video signals such as television signals.
The object of the present invention is to provide a noise removal circuit that can reliably remove noise without damaging the video signal itself.

Conventionally, various noise removal circuits have been proposed that process video signals based on frame correlation to remove noise. Such a noise removal circuit eliminates noise by processing a delayed signal supplied to a delay means having a delay time equal to an integral multiple (usually 1 time) of the frame period of a video signal and an original signal that is not delayed. This is to obtain a video signal from which .

An example of this type of noise removal circuit will be described below with reference to the drawings. FIG. 1 shows the entirety of this noise removal circuit, which will be explained below.

31 is an input terminal from which a video signal from which noise is to be removed is supplied. The video signal is quantized, and its sampling period is T
shall be. 36 is an output terminal from which a video signal from which noise has been removed is obtained. Reference numeral 48 indicates a prediction filter, which is supplied with an input video signal for N frames (N-1, 2, 3, . . . , N-1 in this example).
It is. ) minutes is provided. The input video signal supplied to the input terminal 31 is supplied to this delay means 32 and also to a synthesizer 33 as a subtracter, and in this synthesizer 33, the video signal from the input terminal 31 is converted to the output of the delay means 32. The prediction filter 48 is configured in such a way that . 3
4 is a nonlinear circuit to which the output of the prediction filter 48 is supplied. Here, a nonlinear circuit is one in which the input level vs. output level characteristic exhibits linearity when the absolute value of the input signal level is below a predetermined value or above a predetermined value.
The circuit exhibits nonlinearity in other parts.
As shown in FIG. 2, this circuit has a limiter characteristic in which the input level vs. output level characteristic is linear when the absolute value of the input level is below a predetermined value, and becomes a constant value when it exceeds a predetermined value. In this case, it is also possible to use a stripping circuit in which the output level becomes zero when the absolute value of the input level exceeds a predetermined value. 35 is the output of the nonlinear circuit 34 and the input terminal 3
In this case, in this synthesizer 35, the input terminal 3
The output of the nonlinear circuit 34 is subtracted from the video signal from 1. Reference numeral 37 is a feedback circuit in which the output of the synthesizer 35 is fed back to the input side of the delay means 32, and here it is an attenuator having an attenuation rate K (K≦1). 39 is the sum of the absolute values of the level differences of signals at multiple points in the vicinity of a certain point in the input video signal, and signals at corresponding points in time that are delayed by N frames, here, one frame, from the multiple points. This is a detection circuit for detecting.

The video signal from the input terminal 31 is supplied to an attenuator 43 having an attenuation ratio of 1-K/2, and its output is supplied to a synthesizer 38. Then, in the synthesizer 38, the output of the attenuator 43 and the output of the above-mentioned attenuator 37 are added, and the output is supplied to the delay means 32 through an amplifier 58 with an amplification ratio of 2/1+K. Furthermore,
The video signal from the input terminal 31 is supplied to a synthesizer 33, and the output of the delay means 32 is subtracted from the video signal from the input terminal 31 in the synthesizer 33. Thus, the prediction filter 48 here is constructed by adding an attenuator 43 and an amplifier 58 in addition to the delay means 32 and the synthesizer 33. Synthesizer 33
The output of the prediction filter 48 is supplied to the nonlinear circuit 34 through an amplifier 44 with an amplification ratio of 1+K/2, and its output is supplied to the combiner 35. This synthesizer 35 converts the input video signal from the input terminal 1 into the output of the nonlinear circuit 34, that is, the amplifier 44.
The output of is subtracted. The output of the synthesizer 35 is obtained at an output terminal 36 and is also supplied to a feedback circuit 37.

The detection circuit 39 includes a combiner 55, two-dimensional low-pass filters 40 and 42, and a double-wave rectifier circuit 41. That is, in the synthesizer 55, the input terminal 3
The output of the delay means 32 is subtracted from the input video signal from the input video signal 1, the output of the synthesizer 55 is supplied to the first two-dimensional low-pass filter 40, the output is supplied to the double-wave rectifier circuit 41, Its output is the second 2
dimensional low pass filter 42 . The configurations of the first and second two-dimensional low-pass filters 40 and 42 in this detection circuit 39 may both be configured as shown in FIG. 3, for example.

That is, in FIG. 3, 58 is the input terminal of a two-dimensional low-pass filter, and 69 is its output terminal. The output from the input terminal 58 is supplied to a delay circuit 59 with a delay amount of 1 horizontal period, and the output is further delayed with a delay amount of 1 horizontal period.
horizontal period delay circuit 60 and synthesizer 64
The outputs of the delay circuits 59 and 60 are added together, and the output of the combiner 64 and the input signal from the input terminal 58 are added together in the combiner 65. Furthermore,
The output of the synthesizer 65 is supplied to a delay circuit 61 with a delay amount of T (sampling period), the output of the synthesizer 65 and the output of the delay circuit 61 are added in the synthesizer 66, and the output of the synthesizer 66 is further added. amount of delay
The output of the synthesizer 66 is added to the output of the delay circuit 62 in a synthesizer 67, and the output of the synthesizer 67 is supplied to a delay circuit 63 with a delay amount of 3T. , the output of the synthesizer 67 and the output of the delay circuit 63 are added, and the added output is obtained at the output terminal 69.

Thus, the two-dimensional low-pass filter 4 of FIG.
0,42 is constructed, and its transfer function is as follows.

G(Z) − (1+Z ^-h +Z ^-2h・(1+Z ⁻¹ )・(1+Z ⁻² )・(1+Z ⁻³ )) However, Z is expressed as Z−e ^j 〓 ^T , where ω is the angle This is the frequency. Also, h is the value obtained by dividing the horizontal period by T.

Returning to FIG. 1, the output of the detection circuit 39 is supplied to control circuits 49 and 50, respectively. The output of the control circuit 49 causes an attenuator 37 as a feedback circuit.
K of the amplifiers 43, 44 and the attenuator 58 are controlled together with K of the amplifiers 43, 44 and the attenuator 58. The output of the control circuit 50 controls the input level value at which the characteristic of the nonlinear circuit 34, here the input level vs. output level characteristic, changes from linear to nonlinear, ie, the input threshold level.

Next, the operation of the detection circuit 39 will be explained. The output of the prediction filter 48 is a signal representing the difference between the video signal and a signal delayed by one frame.The signal at a certain point in the video signal is designated as X, and the signal delayed by one frame is designated as X'. If so, Figure 5A
As shown in , X-X' is a curve as time t changes.
If it changes like S ₁ , then this curve
It can be seen that the point where S ₁ crosses the horizontal axis t means that the image is stationary for one frame, and that the image moves between one frame as it moves away from the horizontal axis up and down. In Figure 5A, S ₂
indicates irregular noise superimposed on this video signal. Figure 5B shows the signal X- shown in Figure 5A.
The output when X' is supplied to the first two-dimensional low-pass filter 40 is shown, and this is defined as L ₁ (X−X'). According to this figure, it can be seen that a noise curve S' ₂ from which the high-frequency components of the above-mentioned noise S ₂ of the video signal have been removed is shown. FIG. 5C shows a curve when the output of the first two-dimensional low-pass filter 40 is supplied to the double-wave rectifier circuit 41 for double-wave rectification, and this curve is expressed as |L ₁ (X−X′) Indicated as |. According to this,
It can be seen that the noise S'' ₂ in FIG. 5C is the noise S _{2 '} in FIG. This fifth
The signal in Figure C is further filtered through a second two-dimensional low-pass filter 4.
2 and the output obtained by supplying it to L ₂ | L ₁
Let (X−X′)| According to this, it can be seen that even if pulse-like noise is mixed in the output side of the double-wave rectifier circuit 41, the noise _S2 is removed. Thus, on the output side of the second two-dimensional low-pass filter 42, a detection output indicating whether the screen is stationary or moving during one frame of the input video signal is obtained. In other words, the output of the prediction filter 48 was the result of combining curves S ₁ and S ₂ in FIG. 5A, but the output of the detection circuit 39 was the result of combining these curves S ₁ and S ₂ as shown in FIG. However, it can be seen that the curve is approximately similar to the video signal _S1 .

That is, the detection circuit 39 detects the sum of the absolute values of the level differences of signals at a plurality of points in the vicinity of a certain point in the input video signal and corresponding signals at a plurality of points delayed by N frames each from the plural points. The reason for this is that in the case of a still image, the absolute value of each level difference is very small, and the sum of the absolute values is also small, indicating that this is a still image. This is because even if the absolute value of the level difference is small to some extent, most of the level differences are large and can be regarded as a moving image.

The constant K (or the characteristics of the nonlinear circuit 34) is changed based on the detection output of the detection circuit 39. First, the case where K is changed will be explained. The frequency characteristics of the output video signal of the synthesizer 35 are as shown in FIG. This is similar to the characteristics of a comb filter (C type) (the amount of delay is different)
It is. . Note that ₀ is the frame frequency (30Hz). In this case, the solid line shows the case of K-0, and the broken line shows the case of K-0.5. Then, this video signal becomes a periodic function curve whose frequency becomes zero at ₀ , ₃₀ , ₅₀ , . . . . And K goes from 0 to 1
As it approaches , the shape of the curve changes to become thinner. In this noise removal circuit, the closer K is to 1, the better the S/N becomes. Therefore, the detection circuit 3
9, when the screen movement between one frame is small, this K is set to a value as large as possible, that is, close to 1, and when the image movement is large between one frame of the input video signal by the output of the detection circuit 39. It can be seen that it is sufficient to control K so that it is small. If K is made small, a decrease in resolution on a moving screen can be avoided.

Also, the threshold level of the nonlinear circuit 34 is set to the maximum when the output level of the detection circuit 39 is zero (still image), and as the output level increases (as the movement of the screen increases), the threshold level is increased. Make it smaller. In addition, in the case of a still image, the slope of the straight line part of the nonlinear circuit 34 is increased,
In the case of a moving screen, the slope may be reduced.

According to the noise removal circuit shown in FIG. 1, noise can be reliably removed from both static and moving parts of the video signal reproduction screen, and there is little reduction in resolution.

Although the case where N is 1 has been described above, cases where N is 2, 3, . . . are also possible. However, if the movement of the screen is rapid, it is better for N to be small, and if the movement of the entire screen is relatively slow, N may be somewhat large. However, in the case of a completely stationary surface, N may be set to a fairly large value, but in the case of a normal moving image, N cannot be set very large.

Next, another noise removal circuit of this type will be explained with reference to FIG. 6. Parts corresponding to those in FIG. 1 are given the same reference numerals, and only the parts different from those in FIG. 1 will be explained. There is no amplifier 44 with an amplification ratio of 1+K/2, and an attenuator 43' with an attenuation ratio of 1-K is provided in place of the attenuator 43 with an attenuation ratio of 1-K/2. Also, output terminal 3
6 is provided not on the output side of the synthesizer 35 but on the output side of the synthesizer 38.

By the way, the above-mentioned noise removal circuit inputs the difference signal (output of the synthesizer 33) between the input video signal and a signal delayed by one frame of the input video signal, regardless of the nature of the input video signal. Since the video signal is uniformly supplied to the nonlinear circuit 34 regardless of the nature of the video signal, not only noise but also effective video signal components are removed, resulting in an output video signal with reduced fidelity to the input video signal. There is a possibility that it will be obtained.

In view of this point, the present invention provides a delay means to which an input video signal is supplied and is delayed by a time period of N (N-1, 2, 3, ...) frames, and an input video signal and an output of the delay means. , a nonlinear circuit to which this difference signal is supplied, and a synthesizing means for synthesizing the input video signal and the output of the nonlinear circuit. It is an object of the present invention to propose a noise removal circuit for obtaining an output video signal that can reliably remove noise without impairing the fidelity of the output video signal to the input video signal.

Below, with reference to FIG. 7, an embodiment in which the present invention is applied to the noise removal circuit of FIG. 1 will be described in detail. Omitted.

In the present invention, an input video signal is supplied to N
(However, N-1, 2, 3, ..., in this example N-
1) A delay means 32 that is delayed by the time equivalent to a frame, a difference signal detection means (synthesizer) 33 that detects a difference signal between the input video signal and the output of the delay means 32, and inputs the detected difference signal. A signal dividing means 71 that divides the video signal into a plurality of difference signals according to the properties thereof, and a plurality of nonlinear circuits 72-1, 72- to which the divided plurality of difference signals are respectively supplied.
2, . .

To explain further, the synthesizing means 74 combines the synthesizer 73 which adds the outputs of the nonlinear circuits 72-1, 72-2, . . . , 72-n, and the output of the synthesizer 73 from the input video signal from the input terminal 31. Combiner 3 that subtracts
It consists of 5.

The signal dividing means 71 includes, for example, a determining means for determining the video pattern of the input video signal, and divides the detected signal into a plurality of difference signals according to the video pattern of the input video signal, or detects The obtained difference signal is divided into a plurality of difference signals according to the frequency distribution of the input video signal.

The output of the combiner 33 is supplied to the signal dividing means 71 through the amplifier 44. By the output of the control circuit 50, the nonlinear circuit (limiter circuit) 72-1,
The value of the input level at which the characteristic of input level versus output level of 72-2, .

Next, a specific example of the configuration consisting of the signal dividing means 71, the nonlinear circuits 72-1, 72-2, . . . , 72-n, and the combiner 73 shown in FIG. .

Here, prior to describing a specific example of this configuration, orthogonal transformation and inverse transformation will first be described. Now, the block of the video signal sequence that is the input signal is X, the block of the output signal sequence is Y, the orthogonal transformation matrix is A, and the inverse transformation matrix is similarly B.
If expressed as, Therefore, when the input signal is orthogonally transformed, Therefore, the inverse transformation output is Therefore, the transform coefficient, ie, the orthogonal transform output Y, is a linear combination of the row vector and the input signal.

By the way, orthogonal transforms such as Warschu, Hadamard, and Haar can be used as this orthogonal transform, but they have the characteristics of being able to extract the characteristics of the video signal well and that the inverse transform can be performed in the same procedure as the transform. A Hadamard transform with .

In FIG. 8, 101 is an input terminal to which a difference signal (in this case, an AD-converted difference signal) which is the output of the amplifier 44 in FIG. 7 is supplied. This noise removal circuit includes a serial/parallel conversion circuit 102 - orthogonal conversion circuit 103 - nonlinear circuit 104 - inverse conversion circuit 105
- Consists of a cascade circuit 107 of a parallel/serial conversion circuit 106. Incidentally, the nonlinear circuit 104 is controlled by the output of the control circuit 50 shown in FIG. The video signal from the input terminal 101 is transmitted to this cascade circuit 1.
07, and the output from the output terminal 8 is supplied to the synthesizer 35 in FIG. 7, where it is subtracted from the input video signal from the input terminal 31.

In this case, the serial/parallel conversion circuit 102 and the orthogonal conversion circuit 103 constitute the signal dividing means 71, and the latter circuit 103 constitutes the discrimination means for discriminating the video pattern. Also, nonlinear circuit 1
04 corresponds to the nonlinear circuits 72-1 to 72-n. Inverse conversion circuit 105 and parallel/serial conversion circuit 1
A synthesizer 73 is configured in step 06.

Next, the circuits of each part of the circuit of FIG. 8 will be explained with reference to FIG. 9 and subsequent figures. First, the specific configuration of the serial/parallel conversion circuit 102 will be explained with reference to FIG. The input signal S _i from the input terminal 110 is delayed through delay circuits 111 to 112 whose delay amounts are one horizontal period and one sampling period (this sampling period is the sampling period of the A-D converted video signal), respectively. output terminal 11
4, an output signal S _i4 is obtained from the delay circuit 111, an output signal S i3 is obtained from the output terminal 115, and an output signal S _i3 is obtained from the input terminal 110.
The input signal S _i is delayed through the delay circuit 113 whose delay amount is one sampling period to obtain the output signal S _i2 at the output terminal 116, and the input signal S _i is supplied as it is to the output terminal 117 to obtain the output signal S _i1 . I'm trying to get it.

Next, the parallel/serial conversion circuit 106 will be explained with reference to FIG. 10 when the serial/parallel conversion circuit 102 of FIG. 8 is the same as that described with reference to FIG. 9. Input signal from input terminal 119
S′ _i1 is supplied to the synthesizer 124, and the input signal S′ _i2 from the input terminal 120 is delayed through the delay circuit 123 whose delay amount is one sampling period.
4, and the summed output of both signals from the combiner 124 is supplied to the combiner 128. Input terminal 12
1 to the combiner ₁₂₆ ;
The input signal S′ _i4 from the input terminal 122 is delayed by 1
The signal is delayed through a sampling period delay circuit 125 and supplied to a synthesizer 126 . This synthesizer 12
The summed output of both signals from 6 is supplied to a synthesizer 128 through a delay circuit 127 whose delay amount is one horizontal period. Then, the summed output of both signals from the combiner 128 is obtained at the output terminal 129 as the output signal S' _i .

Next, specific configurations of the orthogonal transform circuit 103 and the inverse transform circuit 105 shown in FIG. 8 will be explained with reference to FIG. 11. In this case, since the orthogonal transform circuit 103 uses a Hadamard transform circuit, the inverse transform circuit 105
has the same configuration. In this example, the conversion circuit 103,1
As 05, a fourth-order Hadamard transform circuit is used. The fourth-order Hadamard transformation matrix H ₄
is as shown in the following equation.

Note that the 2nd to 4th rows of the matrix in equation (8) can be interchanged.

In Fig. 11, 138 to 141 are input terminals;
142 to 149 are synthesizers, of which 142 and 14
4, 146, 149 are addition combiners, 143, 14
5,147,148 are subtractive synthesizers, 150-15
3 is an output terminal. And orthogonal transformation circuit 10
3, the input signal S is input to the input terminals 138 to 141.
The configuration is such that when signals _i1 to S _i4 are supplied, output signals S'' _i1 to S'' _i4 satisfying the following equations are obtained from the output terminals 150 to 153.

As shown in FIG. 12A, a pixel point a ₁ at a certain time point on an adjacent scanning line of a certain field, and a pixel point a at a time point one sampling period, one horizontal period, and one horizontal period + one sampling period before, respectively. ₂ ,
Consider a ₃ and a ₄ . The input signals S _i1 to _{S i4} are difference signals of signals of pixel points a ₁ to _{a 4} that differ by one frame, respectively, and the portions of the screen have a planar pattern as shown in FIG. 12B, respectively, and FIG. 12C. In the case of a horizontal pattern as shown in FIG. 12D, a diagonal pattern as shown in FIG. 12D, and a vertical pattern as shown _{in FIG} .
The absolute value of the level of _S''i4 becomes small.In addition, in FIGS. 12 B to E, hatched areas indicate areas with low brightness, and areas without hatching indicate areas with high brightness. They may be interchanged.

In the inverse conversion circuit 105, the input terminals 138 to
By supplying input signals S _p1 to S _p4 to 141, output signals S' _i1 to S' _i4 satisfying the following equations are obtained at output terminals 150 to 153.

Next, regarding the nonlinear circuit 104 in FIG.
This will be explained with reference to the figures. Figure 13 shows nonlinear circuit 1.
04, input terminals 155 to 15 are shown.
The output signals _S''i1 to _S''i4 from the orthogonal transform circuit 103 in FIG. 11 are supplied to the output terminals 162 to 8, respectively.
165, output signals S _p1 to S _p4 to be supplied to the inverse conversion circuit 105 are obtained. In this case, between input terminal 155 and output terminal 162, input terminal 1
56 and the output terminal 163, between the input terminal 157 and the output terminal 164, and between the input terminal 158 and the output terminal 165, respectively, as a nonlinear circuit whose characteristics are shown in FIG. A limiter circuit (or a stripping circuit is also possible) 159-1, 159 in which the output level is a constant value when the absolute value exceeds a predetermined value, and the relationship between the input level and the output level is linear when the absolute value is below the predetermined value.
-2, 159-3, 159-4 are provided. The threshold level of these nonlinear circuits is controlled by the control output of the control circuit 50 shown in FIG. 7, as will be described later.

Next, a modification of the circuit shown in FIG. 8 will be described with reference to FIG. 14, and parts corresponding to those in FIG. The cascade circuit 107 has approximately the same configuration as that in FIG. 8, but
Circuit 15 in FIG. 13 of nonlinear circuit 104
When considering analog signals whose characteristics are shown in FIG. 15, the circuits corresponding to 9, 160, and 161 have a linear relationship between the input level and the output level when the absolute value of the input level exceeds a predetermined value, and This is a coring circuit whose output level is zero when the value is below the value. Then, the difference signal from the input terminal 101 is supplied to the synthesizer 171 through the delay amount compensation delay circuit 170, and the signal from the cascade circuit 107 is subtracted from this signal, and the signal obtained at the output terminal 108 is as shown in FIG. It is supplied to the synthesizer 35.

The operation of the noise removal circuit shown in FIG. 7 when the circuit having the configuration shown in FIG. 8 described above is adopted will be described, but the explanation of the parts that overlap with those in FIG. 1 will be omitted.

Output terminals 150 to 153 of orthogonal transformation circuit 103
Since each output signal is a difference signal between video signals that differ for one frame period for each different video pattern as shown in FIGS. 12B to 12E, it contains almost no video signal component for each different video pattern. It's noise.

Therefore, depending on the video pattern of the input video signal, the nonlinear circuits 72-1 to 72-4 in FIG. 7, that is, the nonlinear circuit (limiter circuit) 159 in FIG.
The threshold level from -1 to 159-4 may be selected as an appropriate value.

For example, in telecine equipment, high-frequency noise is a problem, but in order to eliminate it, limiter circuit 159-1 is used.
The threshold hold level of the limiter circuits 159-2 to 159-4 is set low to output only small level noise, and the threshold hold levels of the limiter circuits 159-2 to 159-4 are set high to output high level noise. It would be nice if you could also output it.

In addition, in telecine equipment, scraping errors or flicker noises are a problem, but in order to eliminate this problem, the threshold level of the limiter circuit 159-1 is increased to output even a high level of noise. and limiter circuit 1
Each of the threshold levels 59-2 to 159-4 may be set arbitrarily depending on the noise level.

Further, in order to remove noise corresponding to horizontal stripes, diagonal stripes (moiré stripes in the diagonal direction), or vertical stripes, limiter circuits 159-2, 159-3 or 15 are used, respectively.
Increase the threshold hold level of 9-4,
It is only necessary to output even a large level of noise.

In this way, noise corresponding to the video pattern of the input video signal is obtained from the synthesizer 73, and the noise is obtained from the synthesizer 35.
, and is subtracted from the input video signal from the input terminal 31 to obtain an output video signal at the output terminal 36 which has high fidelity to the input video signal and from which noise has been well removed.

Next, the configuration consisting of the signal dividing means 71, the nonlinear circuits 72-1, 72-2, . . . , 72-n, and the combiner 73 in FIG. Two examples will be explained, but the explanation of their specific configuration will be omitted. One example will be explained with reference to FIG. Orthogonal transform circuit (Hadamard transform circuit)
is the following equation 8 for 8 input signals S _i1 to _{S i8}
The configuration is such that two output signals S″ _i1 to S″ _i8 are obtained.

As shown in FIG. 16A, on adjacent scans of a certain field, pixel point b ₁ at a certain time, pixel points b ₂ , b ₃ and b ₄ at times 1, 2 and 3 sampling periods earlier, respectively. Also, consider pixel points b ₄ , b ₅ , b ₆ and b ₇ at times one horizontal period, one horizontal period + 1, 2 and 3 sampling periods before pixel point b ₁ , respectively.
Then, the input signals S _i1 to _{S i8} correspond to pixel points b ₁ to _{b 8} , respectively.
When that part of the screen has the pattern shown in FIGS. 16B to I, the absolute value of the level of the output signals S″ _i1 to S″ _i8 is small correspondingly. becomes. Note that in FIGS. 16B to 16, hatched areas indicate low brightness areas, and non-hatched areas indicate high brightness areas, but these may be interchanged.

Therefore, by appropriately selecting the threshold hold level of each nonlinear circuit (limiter circuit) to which each of the output signals _S''i1 to _S''i8 is supplied according to the video pattern of the input video signal, the video pattern can be adjusted. Noise can be effectively removed depending on the situation.

Next, another example will be explained with reference to FIG. In this case, the input video signal is a color video signal, and the sampling frequency is selected to be four times the color subcarrier frequency. The orthogonal transform circuit (Hadamard transform circuit) receives four input signals S _i1
~S _i4 , the four output signals of equation 10 S″ _i1 ~S″ _i4
Configure it so that it can be obtained.

As shown in FIG. 17A, consider pixel points at a certain time point c ₁ on an adjacent scan of a certain field, two sampling periods, one horizontal period, and one horizontal period+2 sampling periods before. and,
The input signals S _i1 to S _i4 are difference signals of signals that differ by one frame at pixel points c ₁ to c ₄ , respectively, and that part of the screen is the first
In the case of the patterns shown in FIGS. 7B to 7E, the absolute values of the levels of the output signals _S''i1 to _S''i4 become small correspondingly. In FIGS. 17B to 17E, hatched areas indicate low brightness areas, and unshaded areas indicate high brightness areas, but these may be interchanged.

Therefore, by appropriately selecting the threshold hold level of each nonlinear circuit (limiter circuit) to which each of the output signals S″ _i1 to S″ ₄ is supplied, depending on the video pattern of the input video signal, noise can be effectively removed.
In this case, color noise can be effectively removed by setting the threshold level of the nonlinear circuit (limiter circuit) to which the output signal _S''i3 corresponding to the image pattern shown in FIG. 17D is made higher than the others. can do.

Furthermore, the number of the above-mentioned video patterns is 4, 8 as mentioned above.
It is not limited to just one piece, but can be made larger.

If only color noise is to be removed, the difference signal is supplied to a cascade circuit consisting of a filter that attenuates the frequency band of the color signal component, a nonlinear circuit (limiter circuit), and a filter that emphasizes the frequency band of the color signal component. The output may be supplied to the synthesizer 35 and subtracted from the input video signal.

Next, a specific example of another type of the configuration consisting of the signal dividing means 71, the nonlinear circuits 72-1 to 72-n, and the combiner 73 shown in FIG. 7 will be described with reference to FIG. In FIG. 18, the difference signal which is the output of the amplifier 44 in FIG. 7 is supplied to the signal dividing means 71,
Its four outputs are nonlinear circuits (limiter circuits) 7
2-1 to 72-4, their respective outputs are supplied to the combiner 73 and added, and the output obtained at the output terminal 87 is supplied to the combiner 35 in FIG. is subtracted from the input video signal.

Next, the configuration of this signal dividing means 71 will be explained. The difference signal from the input terminal 80 is supplied to a surface scanning direction (vertical direction) low-pass filter 81, the difference signal from the input terminal 80 is subtracted from the output by a synthesizer 82, and the output from the synthesizer 82 is The signal is supplied to a low-pass filter 84 in the scanning direction (lateral direction). The output of low pass filter 84 is supplied to nonlinear circuit 72-3. The output of the combiner 82 is subtracted from the low-pass filter 84 in the combiner 86, and the output of the combiner 86 is applied to the nonlinear circuit 72-4.
supplied to The output of the low pass filter 81 is supplied to a line scanning direction (horizontal direction) low pass filter 83, and its output is supplied to the nonlinear circuit 72-1. In the combiner 85, the output of the low-pass filter 81 is subtracted from the output of the low-pass filter 83, and the output of the combiner 85 is applied to the nonlinear circuit 72-2.
supplied to

Here, the line scanning direction (horizontal direction) low-pass filters 83 and 84 are ordinary low-pass filters such as CR type or LB type. The surface scanning direction (vertical direction) low-pass filter 81 connects one or more delay circuits with a delay time of one horizontal period in cascade, supplies an input signal to the cascade circuit, and outputs the input signal and each delay. Each output of the circuit is configured to be arithmetic averaged using a combiner and an attenuator.

Thus, the signals supplied to the nonlinear circuits 72-1 to 72-4 are a vertical low frequency and horizontal low frequency signal, a vertical low frequency and horizontal high frequency signal, and a vertical high frequency signal of the difference signal, respectively. In addition, they are a horizontal low frequency signal, a vertical high frequency signal, and a horizontal high frequency signal.

And nonlinear circuit (limiter circuit) 72-1
By arbitrarily selecting the threshold levels from 72-4 to 72-4, noise can be effectively removed according to the two-dimensional frequency distribution of the input video signal.

Note that the input video signal is a color video signal (for example,
NTSC system), noise removal may be performed without separating the luminance signal and carrier color signal as in each of the above embodiments, but if they are separated, the fidelity of the output video signal with respect to the input video signal will be reduced. The noise can be effectively removed by making it even higher.

In this case, the difference signal from the amplifier 14 shown in FIG. The outputs of each nonlinear circuit are supplied to a plurality of nonlinear circuits (limiter circuits), each output of each nonlinear circuit is combined, and the synthesizer 35 in FIG. Make it.

In this case, from the carrier color signal component in the difference signal,
The orthogonal transform circuit is used to generate an output signal _S''i3 corresponding to the video pattern shown in FIG. The noise is extracted by the means described in connection with FIG. 13, or FIGS. 14 to 15, or FIG. 16, or FIG. 17, or FIG. 18, and subtracted from the input color video signal. .

In this way, color noise dispersed in the luminance signal band of the color video signal can also be effectively removed.

According to the present invention described above, there are the following advantages. A delay means capable of delaying by a time corresponding to N (N=1, 2, 3, ...) frames is provided, and a difference signal between the input video signal and a signal delayed by supplying this to the delay means is detected. The detection signal is then fed to a nonlinear circuit and the resulting signal is combined with the input video signal, so it is possible to effectively remove noise from video signals with little movement on the screen without reducing fidelity. can do. In the case of the example,
Noise can be effectively removed from video signals with movement on the screen without significantly reducing fidelity.

In the present invention, the detected difference signal is divided into a plurality of difference signals according to the characteristics of the input video signal, and each of the divided signals is supplied to a separate nonlinear circuit. Since the output is combined with the input video signal, noise can be removed even more effectively without reducing the fidelity to the input video signal. For example, input video signals of images such as facial avatars, sand, and low-luminance areas are not removed as noise, and even if the positions of the images are slightly shifted between frames ( For example, in the case of a telecine device) there is little risk of it being removed as noise.

Incidentally, the present invention can also be applied to the noise removal circuit shown in FIG.

[Brief explanation of the drawing]

FIG. 1 is a system diagram showing a noise removal circuit used for explaining the present invention, FIG. 2 is a characteristic curve diagram, and FIG.
FIG. 4 is a characteristic curve diagram, FIG. 5 is a waveform diagram, FIG. 6 is a system diagram showing a noise removal circuit used to explain the present invention, and FIG. 7 is a diagram of the present invention. System diagram showing one embodiment of the invention noise removal circuit, No. 8
The diagram is a system diagram showing the specific configuration of a part of Figure 7.
Figures 10 and 11 are system diagrams showing a specific configuration of a part of Figure 8, respectively, Figure 12 is an explanatory diagram, and Figure 1
3 is a system diagram showing a specific configuration of a part of FIG. 8, FIG. 14 is a system diagram showing another example corresponding to FIG. 8,
FIG. 15 is a characteristic curve diagram, FIGS. 16 and 17 are explanatory diagrams of other embodiments of the present invention, and FIG. 18 is a system diagram showing another example corresponding to FIG. 8. 32 is a delay means, 33 is a combiner as a difference signal detection means, 71 is a signal division means, 72-1, 72
-2, . . . 72-n is a nonlinear circuit, and 74 is a synthesis means.

Claims (1)

[Claims] 1. An input video signal is supplied to N (however, N-1,
2, 3, ...) a delay means for delaying by a time corresponding to a frame; a difference signal detection means for detecting a difference signal between the input video signal and the output of the delay means; a signal dividing means for dividing the input video signal into a plurality of difference signals according to the nature of the input video signal; a plurality of nonlinear circuits to which the plurality of divided difference signals are respectively supplied; What is claimed is: 1. A noise removal circuit comprising a synthesis means for synthesizing each output of a nonlinear circuit, and an output video signal from which noise has been removed is obtained from the synthesis means. 2. The signal dividing means includes a determining means for determining a video pattern of the input video signal, and divides the detected difference signal into a plurality of difference signals according to the video pattern of the input video signal. A noise removal circuit according to claim 1, characterized in that: 3. The noise removal according to claim 1, wherein the signal dividing means divides the detected difference signal into a plurality of difference signals according to the frequency distribution of the input video signal. circuit.