JP2947837B2 - Control motion signal generation circuit - Google Patents

Description

Description: TECHNICAL FIELD The present invention relates to a signal processing circuit for a television signal,
In particular, the present invention relates to a control motion signal generation circuit suitable for generating a signal for controlling a motion adaptive circuit.

[Conventional technology]

A signal processing method of a television signal according to the movement of an image is a very effective means for obtaining high image quality. Later, for NTSC composite color television signals, YC separation processing using frame correlation for still images and YC separation processing using line correlation for moving images can cause dot interference and cross color. Can be suppressed. Also,
For scanning line interpolation, inter-field interpolation is performed for still images, and intra-field interpolation is performed for moving images.
Line flicker can be prevented and vertical resolution can be improved.

However, this motion-adaptive signal processing exerts the above-mentioned effects when the motion of the image is accurately detected. However, when the motion is not accurately detected, the processing of the moving image is performed on the still image, and the processing of the still image is performed on the moving image. Processing is performed, which increases interference and deteriorates image quality.

Japanese Patent Application Laid-Open No. 62-111586 is a known example of a motion-adaptive circuit in which an error in motion detection is small even in response to a fast motion of an object.
FIG. 2 is a block diagram showing an example of the motion detection part in the above-mentioned known example. In FIG. 2, 12 is an input terminal, 13 and 14 are first and second field memories, respectively, 3 is a subtraction circuit,
5 is an absolute value device, 6 is a nonlinear conversion circuit, and 10 is a spatiotemporal processing circuit.

First and second field memories 13, 14 connected in series
From the input terminal of the first memory 13 and the output terminal of the second memory 14, two signals for operating the same position on the screen one frame period apart are obtained. After inputting these two signals to the subtraction circuit 3 to obtain an inter-frame difference signal, the absolute value of the inter-frame difference signal extracted via the absolute value device 5 is input to the non-linear conversion circuit 6 to detect the amount of motion. I do. Here, the absolute value of the inter-frame difference signal input to the nonlinear conversion circuit 6 is also 8 bits if the input television signal is quantized by 8 bits. On the other hand, the output motion amount is compressed to, for example, 4 bits to reduce the number of bits. Even in this case, switching of the motion adaptive signal processing circuit
There are 16 steps, and it can be done smoothly. A signal having a small absolute value of the input frame difference signal is regarded as noise,
By performing a non-linear conversion such that the amount of output movement becomes zero, erroneous detection due to noise is prevented.

By inputting the motion amount output from the non-linear conversion circuit 6 to the spatio-temporal processing circuit 10 and expanding the motion amount in the time direction and the space direction, the known example can be applied to a fast-moving image or an image with a fine pattern. An error-free motion determination is performed to control the motion adaptive signal processing circuit.

[Problems to be solved by the invention]

In the above conventional example, noise which is not removed by the non-linear conversion circuit 6 and has a large signal component later will be removed by the spatio-temporal processing circuit 10.
, The spatiotemporal processing circuit 10 considers the noise as a motion amount, extends the noise in the spatiotemporal direction, and outputs it as a control motion signal. For this reason, even if a still image is subjected to a moving image process due to the influence of noise, the image quality is degraded.

SUMMARY OF THE INVENTION An object of the present invention is to provide a control motion signal generation circuit capable of removing erroneous detection of a motion amount due to noise and performing more accurate motion detection.

[Means for solving the problem]

A control motion signal generating circuit according to the present invention for achieving the above object receives a television signal, detects a motion of an image based on a difference between frames of the television signal, and creates a motion information signal. A motion detection unit, comparing the motion information signal at a certain pixel on the image with a pixel adjacent to the pixel in the vertical direction, for example, the motion information signal at a pixel on one line at the same horizontal position as the pixel; A selecting unit that outputs a smaller value as the motion information signal of the certain pixel; and a spatio-temporal processing unit that performs processing in the time direction and the spatial direction on the motion information signal output from the selecting unit. It is characterized by the following.

[Action]

The detected motion amounts of pixels adjacent in the vertical direction have no correlation with each other regarding erroneous motion detection due to noise. For this reason, if the amount of motion of a certain pixel is compared with the pixel adjacent to the pixel in the vertical direction and the smaller one is selected as the amount of motion of the pixel, a large amount is erroneously detected by the noise without being affected by the noise. It is possible to prevent the used motion amount from being used as motion information for motion adaptive television processing. Therefore, an accurate motion information signal that is not affected by noise can be obtained.

〔Example〕

Hereinafter, an embodiment of the present invention will be described with reference to FIG. In FIG. 1, 1 is an input terminal, 2 is a frame memory, 3 is a subtraction circuit, 4 is a low-pass filter circuit (hereinafter abbreviated as LPF), 5 is an absolute value device, 6 is a nonlinear conversion circuit, and 7 is a noise removal circuit. , 8 is a line memory for delaying 1H (1H is one horizontal scanning cycle of a television signal), 9 is a selection circuit, 10 is a spatiotemporal processing circuit, and 11 is a control motion signal output direct terminal.

The digitized television signal is input to the input terminal 1, and the motion of the television signal is detected first. An input signal of the input terminal 1 and a signal obtained by delaying the input signal of the input terminal 1 by one frame in the frame memory 2 are supplied to a subtraction circuit 3 to obtain an inter-frame difference signal. Since the phase of the color signal is inverted between one frame, the inter-frame difference signal includes a color signal component in addition to the motion amount component. Therefore, the band is limited by the LPF 4 to remove the color signal component. Further, the positive / negative polarity for the difference signal is removed by the absolute value device 5.

As described above, even if the signal for controlling the motion adaptive signal processing circuit is, for example, 4 bits, there are 16 steps, and control can be performed sufficiently smoothly. Therefore, the output signal of the absolute value device 5 is input to the nonlinear conversion circuit 6, and bit compression is performed. The noise included in the motion amount is removed by inputting the output of the nonlinear conversion circuit 6 to the noise removal circuit 7. The noise elimination circuit 7 includes a 1H memory 8 and a selection circuit 9, and implements noise elimination by comparing and selecting a signal input to the noise elimination circuit 7 and a signal obtained by delaying the signal by 1H in the 1H memory 8. I have. The output signal of the noise elimination circuit 7 is input to the spatio-temporal processing circuit 10, and is stretched in the spatio-temporal direction so that there is a motion even for a fast motion or a fine motion of a picture that cannot be detected only by one frame difference. Output from 11 as a control motion signal.

Next, a more detailed description of the operation of the noise removal circuit 7
This is performed using FIG. 4 and FIG. FIG. 3 is a diagram showing a noise removing operation in this embodiment, and FIGS. 4 (a) and 4 (b) are block diagrams showing a specific circuit configuration of the noise removing circuit 7.
4 (a) and 4 (b), 8 is a 1H memory, 15 is a minimum value selection circuit, and 16 is an AND circuit.

In FIG. 3, it is assumed that a certain pixel on the n-th line is noise. As described above, using the 1H memory 8, (n-
1) Comparison and selection with the pixel at the same horizontal position on the line are performed. For example, the minimum value selection circuit 15 in FIG. 4A is an example of the selection circuit. Here, taking the minimum value of the noise pixel on the n-th line and the non-noise pixel on the (n-1) -th line, the non-noise pixel on the (n-1) -th line is selected. Since it is output as the amount of movement,
This means that noise has been removed. Further, if a certain pixel on the nth line is a moving image, the pixel at the same horizontal position on the (n-1) th line is generally a moving pixel of substantially the same level, so there is no problem even if the output of the minimum value is obtained. The selection circuit 9
Is not limited to the minimum value selection circuit 15 shown in FIG. 4A, and the AND circuit 16 shown in FIG.

In FIG. 4 (b), 16 is an AND comprising a plurality of AND elements
The circuit 8 is a 1H memory.

For example, a motion information signal (b3, b2, b1, b
0 and b3 are the most significant bits, b2 is the second most significant bit, b1
Is the third bit from the high order, and b0 is the least significant bit). The 1H memory 8 stores the motion information signal of the previous line (a3, a2, a1, a0, a3 are the most significant bit, a2
Is the second bit from the high order, a1 is the third bit from the high order, and a0 is the least significant bit. And AND circuit 1
In step 6, an AND operation is performed for each bit (b3 and a3, b2 and a2, b1 and a1, b0 and a0) of the current motion information signal and the motion information signal one line before, and the current motion information signal (c3, c2, c1, c0).
For example, it is assumed that 15 (1111) is input as an erroneous motion information signal at a certain pixel due to noise. Since the motion information signal of the still pixel one line before is 0 (0000), the AND output for each bit is 0 (0000), and erroneous detection due to noise can be removed. In addition, the minimum value selection circuit 15
The circuit scale can be reduced as compared with.

As described above, according to the present embodiment, the noise component included in the motion amount of the television signal can be removed, and there is an effect that the noise prevents the spatio-temporal processing circuit from erroneously expanding the moving image processing range.

Next, another embodiment of the present invention will be described with reference to FIG.
The present embodiment is an example of a control motion signal generation circuit provided with a more accurate noise elimination circuit, which also supports a special case where the correlation with peripheral pixels such as a boundary portion of motion is small.
In FIG. 5, reference numeral 17 denotes a noise removing circuit which is a feature of the present embodiment, reference numeral 18 denotes a sample memory for delaying by N pixels, reference numeral 19 denotes a selection circuit, and the other components are the same as those in FIG. Noise removal circuit
17 comprises a sample memory 18 and a selection circuit 19. Sample memory for the input signal of the noise elimination circuit 17 and the input signal
The signal delayed by 18 is selected by the selection circuit 19, and the space-time circuit
It outputs the amount of movement to 10.

The configuration of the sample memory 18 will be described with reference to FIGS. For example, as shown in FIG. 6, it is assumed that a horizontal and vertical 3 × 3 9-pixel motion information signal is obtained with pixel E as the center. In this case, the sample memory 18 has two 1H memories 4
1, 42 and six shift registers 43, 44, 45, 46, 47, 48. Each output terminal of the sample memory 18 has a pixel G in FIG. 6 from an output terminal 49, a pixel H from 50, a pixel H from 51, a pixel D from 52, a pixel D from 53, a pixel E from 53, a pixel from 54 to 55, 55 , A pixel B, a pixel B, and a pixel C from 57, respectively.

The detailed operation of the noise elimination circuit 17 in this embodiment will be described with reference to FIGS. 8, 9, and 10. FIG. FIG. 8 is a view for explaining processing of a special moving image portion in the previous embodiment, and FIG.
FIG. 10 is a schematic diagram illustrating an example in which patterns of boundary portions of some moving images are classified into modes in the processing of a special moving image portion according to the present embodiment. FIG. 10 is a specific block diagram of the noise removal circuit 17. . In FIG. 10, reference numeral 20 denotes a minimum value selection circuit, reference numeral 21 denotes a mode selection circuit, and others are the same as those in FIG.

The noise elimination circuit 17 of the present embodiment is characterized in that the noise is eliminated by comprehensively judging the motion information of a certain pixel and peripheral pixels in the horizontal and vertical directions centering on the pixel.

As shown in FIG. 8, when the boundary portion of the motion is on the m-th line, several pixels on the m-th line are moving images, and several pixels at the same horizontal position on the (m-1) th line are still images, In the example noise elimination circuit 17, the pixel on the m-th line and (m-1)
Since the comparison (minimum value) is performed with a pixel having no correlation at the same horizontal position on the line, the moving pixel on the m-th line is regarded as noise and is output as a still image. On the other hand, in the noise elimination circuit 17 of the present embodiment shown in FIG. 10, the input signal of the noise elimination circuit 17 and the output signal of the sample memory 18 determine a certain pixel and its surrounding pixels such as 9 pixels.
Pixels are obtained, and the minimum value selection circuit 20 takes the minimum value for every three pixels in each line. At this point, the mode is divided for the information with or without motion of each minimum value, and the mode selection circuit
At 21, it is selected whether to output a certain center pixel as a moving image or a still image according to the mode. FIG. 9 shows an example of mode division. In FIG. 9, for example, for eight patterns of nine pixels on three lines, the patterns (a) to (d) are in mode 1 and the patterns (e) to (h) are in mode 2
And The mode selection circuit 21 outputs mode 1 as a still image and mode 2 as a moving image. Here, in the case where only one line is a moving image as shown in FIG. 9C or only one line is a still image as shown in FIG.
Pixels are regarded as noise, (c) classifies into mode 1 and outputs as a still image, and (g) classifies into mode 2 and outputs as a moving image to remove noise.

As described above, this embodiment has the same effect as the embodiment of FIG.
Further, in the noise removal process, the operation can be performed without erroneously removing a special motion portion as noise, and a more accurate control motion signal can be generated.

FIG. 11 shows another embodiment of the present invention. In this embodiment, in the control motion signal generation circuit, in addition to the detection of the motion of the luminance signal, the motion of a color signal having the same luminance but different colors is also detected,
After the movements of both are combined, noise removal is performed. This makes it possible to remove the obvious noise that exists even after the luminance motion and the color motion are combined, and to output a control motion signal based on the detected image motion with higher accuracy.

In FIG. 11, reference numeral 22 denotes a band-pass filter (hereinafter referred to as a BPF).
23 is an ACC amplification circuit, 24 is a demodulation circuit, 25 is a two-frame memory, 26 is a subtraction circuit, 27 is an absolute value device, 28 is a non-linear conversion circuit, 29 is a synthesis circuit, and the others are the same as those in FIG. It is. The luminance motion detection in this embodiment follows the same process as in the previous embodiment. An inter-frame difference signal of the digitized composite color television signal input from the input terminal 1 is obtained from the frame memory 2 and the subtraction circuit 3, and is converted into a luminance movement amount via the LPF 4 and the absolute value device 5, and is converted into a non-linear conversion circuit 6.
To perform bit compression.

On the other hand, color signal motion detection starts by extracting a signal in a color signal band from the input signal input to the input terminal 1 by the BPF 22. Due to the function of the ACC operating to keep the level of the burst signal included in the output signal of the BPF 22 constant, the variation of the color signal level due to the frequency characteristics of the transmission line is corrected in the output of the ACC amplification circuit 23. In addition, almost constant signal is obtained. Thereafter, the demodulation circuit 24 performs color demodulation of the color signal, and outputs two color difference signals (RY) and (B-
Y) is obtained by dot-sequentially multiplexing the signals. Although the phase of the color subcarrier is inverted between frames, the demodulation circuit 24 operates to cancel the inversion between frames. The signal in the color signal band input to the demodulation circuit 24 includes a high frequency component of the luminance signal in addition to the color signal. Therefore, the high frequency component of the luminance signal is inverted and output between frames at the output of the demodulation circuit 24. The output signal of the demodulation circuit 24 is input to the two-frame memory 25 and is delayed by two frame periods. Then, the output of the demodulation circuit 24 and 2
The output of the frame memory 25 is supplied to obtain a difference signal between two frames. Between the signals separated by two frame periods, the phases of the high-frequency components of the color signal and the luminance signal are also the same, so that when the difference signal between the two frames is other than 0, it can be determined that there is a motion. Thereafter, in the same manner as in the processing of the movement of the luminance, the absolute value converter 27 removes the positive and negative polarities due to the difference signal, and performs the bit compression in the nonlinear conversion circuit 28.

The motion amount of the luminance signal and the motion amount of the chrominance signal, which have been bit-compressed, are synthesized by the synthesizing circuit 29, and then input to the noise elimination circuit 7 for noise elimination. The spatio-temporal processing circuit 10 extends the output of the noise removal circuit 7 by the amount of motion, and outputs the control signal from the output terminal 11 as a control motion signal.

In the present embodiment, the motion detection of the luminance signal and the color signal is separately performed, and the non-linear conversion processing is performed on each motion amount. Therefore, the detection sensitivities of the non-linear conversion circuits 6 and 28 can be adjusted independently, and the detection can be performed separately. After synthesizing the luminance movement and the color movement
A control motion signal generation circuit based on a motion signal detected with high accuracy, which has a feature of removing apparent noise determined as noise.

Next, another embodiment of the present invention will be described with reference to FIG. In this embodiment, the control motion signal generating circuit performs a non-linear conversion after removing noise, and removes noise at a stage before bit compression of a motion amount detected from an inter-frame difference signal. In addition, since this noise removal is performed before the combination of the luminance movement and the color movement, more accurate noise removal is possible.

In FIG. 12, reference numeral 30 denotes a noise removing circuit, and the other components are the same as those in FIG. The detection of the amount of motion in the present embodiment is the same in both the luminance system and the color system as in the previous embodiment. By inputting the output of the absolute value unit 5 to the noise removing circuit 7 in order to perform highly accurate noise removal on the amount of luminance movement obtained through the LPF 4 and the absolute value unit 5, the interframe difference signal is input. Noise removal can be performed with substantially the same number of bits as the signal input to the terminal 1. Thereafter, the non-linear conversion circuit 6 performs bit compression. On the other hand, the two-frame difference signal of the color signal is
Similarly to the luminance, the output of the absolute value unit 27 is input to the noise removal circuit 30 in the same manner as for the luminance of the color obtained through the process, and after the noise removal, the non-linear conversion circuit 28 performs bit compression.

The motion amount of the bit-compressed luminance signal and the color signal is synthesized by the synthesizing circuit 29, the spatio-temporal processing circuit 10 extends the motion amount, and outputs the output terminal 11 as a control motion signal.

According to this embodiment, noise is removed before the motion amount is bit-compressed, so that more accurate noise removal can be performed in a large amount of information. Since detection and noise removal are performed, noise that is difficult to remove after synthesis can be accurately removed, and there is an effect that a control motion signal is generated by more advanced motion detection.

Next, another embodiment of the present invention will be described with reference to FIG.
The present embodiment includes an adaptive noise elimination circuit that turns on / off noise removal processing that occurs in a motion amount signal in accordance with the state of the motion amount, thereby removing a moving image as noise erroneously. No control motion signal can be generated by more accurate motion detection.

FIG. 13 is a block diagram of a control motion signal generation circuit including an adaptive noise elimination circuit. In FIG. 11, reference numeral 32 denotes an adaptive noise elimination circuit, reference numeral 31 denotes a control circuit, and others are the same as those in FIG.

In the embodiment shown in FIG. 1, noise removal processing is performed on all motion amount signals. In this case, for example, motions scattered on the screen, such as panning grass and sand, have less correlation with surrounding pixels than normal moving images, so they can be removed as noise occurring in still images There is.
This means that the detected moving image area is reduced to a still image. As a result, in the motion adaptive signal processing circuit, still image processing is performed on moving images, which may cause image quality deterioration such as double images and blurring.

Therefore, in the present embodiment, the motion amount signal detected in the same manner as in the first embodiment is output from the nonlinear conversion circuit 6, and then processed by the adaptive noise elimination circuit 32, thereby solving the above problem. The adaptive noise elimination circuit includes a noise elimination circuit 7 and a control circuit 31. The noise removing method in the noise removing circuit 7 is the same as that in the embodiment shown in FIG. 1, but the adaptive noise removing circuit 32 in this embodiment turns ON / OFF the noise removing operation. ON / OFF of the noise removal operation is performed by a control signal from the control circuit 31. The control circuit 31 determines whether the input motion amount signal is a still image area (part) or a moving image area (part). Outputs a control signal such as OFF. As a result, noise removal processing is not performed on moving images such as turf to be panned.
Can be output as an input signal.

Next, an example of a specific circuit configuration of the adaptive noise elimination circuit 32 of FIG. 13 will be described with reference to FIG. FIG. 14 shows an example of an adaptive noise elimination circuit 32 characterized in that ON / OFF of a noise elimination operation is controlled depending on the level of an input motion amount signal.

In FIG. 14, 9 is a noise removal circuit, 8 is a 1H memory, 9 is a selection circuit, 31 is a control circuit, and 33 is a bit selection circuit.

The noise elimination circuit 7 in FIG. 14 includes a 1H memory 8 and a selection circuit 9, and performs the same noise elimination processing as in the embodiment shown in FIG. The control circuit 31 includes a bit selection circuit 33, and the bit selection circuit 31 selects and outputs one bit from N bits (N is a natural number) of the input motion amount signal.

For example, if the input motion amount signal is 4 bits now,
The bit selection circuit 33 is the most significant bit (hereinafter, MSB)
) Is output. When the motion amount signal is 4 bits, the motion amount level has 16 levels of 0 to 15, and among the levels of 0 to 7, the MSB is 0 (low) and 8 to 8 levels.
At the level of 15, the MSB is 1 (high). In general, the noise in the motion amount tends to have a lower level of the motion amount signal than the normal motion. Therefore, when the MSB of the motion amount signal is 0, the noise is considered to be noise, and when the MSB is 1, the motion is considered. Signal
The MSB can be used as it is as an ON / OFF control signal of the noise removal circuit 7. In this case, there is an advantage that the bit selection circuit 33 can be simply configured and the circuit scale can be reduced.

For example, in the bit selection circuit 33, the output 1
If a bit signal is used by decoding N bits of the input motion amount signal, the threshold value of the noise / motion level can be set freely. In this case, it is possible to accurately distinguish between noise and moving images, and there is an advantage that ON / OFF of the noise removal circuit 7 can be controlled with higher accuracy.

Then, an output signal from the control circuit 31 is supplied to the selection circuit 9. In the selection circuit 9, the control signal is turned on for noise removal.
In the case of this command, the input motion amount and the motion amount one line before are selected by the same operation as the embodiment of FIG. When the command is a noise removal OFF command, the input motion amount is forcibly output. Thereby, ON / OFF of noise removal can be realized.

As described above, according to the present embodiment, by turning off the noise removal circuit operation for a moving image, it is possible to remove only noise mixed in the motion amount signal, thereby generating a control motion signal with higher accuracy and without side effects. It is possible to do.

Also, in the case where the noise elimination circuit 17 shown in the embodiment of FIG. 5 is applied instead of the noise elimination circuit 7 in the adaptive noise elimination circuit 32 of the embodiment, similarly to this embodiment,
Noise removal in the motion signal can be adaptively performed,
A more accurate control motion signal generation circuit can be realized.

Also, in the embodiment of FIGS. 5, 11, and 12, the description of this embodiment has been made in place of the noise elimination circuits 17 of FIG. 5, 7, 7 of FIG. 11, and 7, 30 of FIG. Even if the adaptive noise elimination circuit 32 is applied, the noise elimination is turned on similarly to the present embodiment.
By adaptively performing / OFF, it is possible to accurately remove only noise without performing noise removal processing on a moving image, so that a higher-performance control motion signal can be generated.

Next, another embodiment of the adaptive noise elimination circuit 32 will be described with reference to FIG. The present embodiment is characterized in that it comprehensively determines the input motion amount and the motion amount information of surrounding pixels in the horizontal and vertical directions around the pixel, and performs ON / OFF of noise removal. 3 is an example of a pattern noise elimination circuit 32.

FIG. 15 is a block diagram showing the present embodiment. In FIG. 15, reference numeral 34 denotes a sample memory, reference numeral 35 denotes a determination circuit, and others are the same as those in FIG.

The control circuit 31 which is a feature of the present embodiment includes a sample memory.
34 and a decision circuit 35. The sample memory 34
The input motion amount signal is delayed in the horizontal and vertical directions, and the motion amounts of a plurality of pixels in a certain range are output and supplied to the determination circuit 35.

The problematic movements scattered on the screen are more frequent within a certain range than the movements caused by the noise generated in the still image. Therefore, it is possible to sum up the motion amounts within a certain range and determine whether the motion or noise is based on the sum. Therefore, in the determination circuit 35,
The sum of the amounts of motion in a certain range in the horizontal and vertical directions obtained by the sample memory 34 is calculated, and a certain threshold is set.
The control signal of F is output. If the control signal is equal to or less than the threshold value, the control signal of noise removal ON is output as noise. As a result, the control circuit 31 of the present embodiment can output a noise removal ON / OFF control signal created by more accurately discriminating between noise in a still image and a moving image.

Next, another embodiment of the adaptive noise elimination circuit 32 will be described in the sixteenth embodiment.
This will be described with reference to the drawings. The present embodiment is an example of an adaptive noise elimination circuit 32 that is characterized by performing ON / OFF of noise elimination using a signal obtained by integrating an input motion amount signal.

FIG. 16 is a block diagram showing the present embodiment. In FIG. 16, reference numeral 36 denotes an integrating circuit, 37 denotes a resistor R, 38 denotes a capacitor C, and others are the same as those in FIG.

The control circuit 31 featured in the present embodiment includes an RC integration circuit 36 including a resistor R37 and a capacitor C38. Now, the time constant of the integration circuit 36 is set to, for example, about 50 μsec (less than 1H) from the values of the R and C elements, and the MSB of the motion amount signal is input. If the motion amount signal is noise, the MSB is an impulse signal whose time width is much narrower than the set time constant, so that the output of the integration circuit 36 is a 0 (low) signal. On the other hand, when the motion amount signal is a moving image, the MSB is a signal having a sufficiently wider time width than the motion amount signal due to noise, so that the MSB is integrated by the integration circuit 36 and has 1 (high) having almost the same time width as the input. Is output. Then, the output signal of the integration circuit 36 is supplied to the selection circuit 9 to control ON / OFF of the noise removal processing.

As described above, the present embodiment has the same effect as the embodiment of FIG. 15, and the noise elimination circuit 7 constituted by a digital circuit.
Can be realized by simple R and C analog elements.

Next, another embodiment of the adaptive noise elimination circuit will be described with reference to FIG. This embodiment is an example of an adaptive noise elimination circuit 32 characterized in that a noise elimination ON / OFF control signal is supplied to a 1H memory.

FIG. 17 is a block diagram showing the present embodiment. In FIG. 17, reference numeral 15 denotes a minimum value selection circuit, and the others are the same as those in FIG.

The noise elimination circuit 7 in this embodiment is configured to output, for example, the smaller of the input motion amount and the motion amount one line before as in FIG. 4A. In this embodiment, the noise removal ON / OFF signal created by the control circuit 31 is input to the 1H memory 8. When the noise removal is ON, the 1H memory 8 outputs a motion amount signal delayed by 1H, and performs a normal noise removal process. When the noise removal is OFF, the output of the 1H memory 8 is turned OFF, and the maximum level of motion is forcibly supplied to the minimum value selection circuit 15. For example, when a noise removal ON / OFF signal is supplied to the output enable terminal 39 of the 1H memory 8, the noise removal is performed.
When OFF, the output terminal 40 of the 1H memory 8 is in a high impedance state. If the level at which the amount of motion is maximized (for example, 15 in the case of 4 bits) is preset at the output terminal by pulling up to a power supply using a resistor, the minimum value selection circuit 15 , The motion signal of the maximum level is input. Therefore, since the input motion amount signal is output from the minimum value selection circuit 15, this is equivalent to not performing the noise removing operation.

As described above, according to the present embodiment, in the minimum value selection circuit 15,
The adaptive noise elimination circuit 32 similar to that of the above-described embodiment can be realized without the need for a circuit part for turning on / off the selection operation. Has the effect of being able to generate

〔The invention's effect〕

According to the present invention, in the control motion signal generation circuit, generation of an erroneous motion amount signal due to noise can be prevented by removing the motion amount having no correlation with the surrounding pixels by the noise removal circuit, thereby improving the accuracy. High control motion signals can be generated.

In addition, by using the adaptive noise elimination circuit as the noise elimination circuit, it is possible to prevent the moving image area from being reduced more than necessary for the noise elimination processing, and it is possible to generate a control motion signal with higher accuracy.

As a result, the motion adaptive signal processing circuit has an effect of preventing a deterioration in image quality by performing a moving image process on a still image due to noise.

[Brief description of the drawings]

FIG. 1 is a block diagram showing an embodiment of the present invention, FIG. 2 is a block diagram of a conventional motion detecting section, FIG. 3 is a schematic diagram for explaining an example of a noise removing operation, and FIG. FIG. 5 is a block diagram showing an example of a noise removing circuit shown in FIG. 5, FIG. 5 is a block diagram showing another embodiment of the present invention, FIG. 6 is a schematic diagram showing a noise removing function, FIG. 8 is a schematic diagram of a boundary portion of a moving image, FIG. 9 is a diagram for explaining the operation of the selection circuit of the embodiment of FIG. 5, FIG. 10 is a block diagram showing an example of the noise removal circuit of FIG. FIG. 11 is a block diagram showing still another embodiment of the present invention, FIG. 12 is a block diagram showing still another embodiment of the present invention, and FIG. 13 is a block diagram showing still another embodiment of the present invention. FIG. 14 is a block diagram showing another example of the adaptive noise elimination circuit of FIG. 13, and FIG. 15 is an adaptive noise elimination circuit of FIG. FIG. 16 is a block diagram showing another example of the adaptive noise elimination circuit of FIG. 13, and FIG. 17 is a block diagram showing still another example of the adaptive noise elimination circuit of FIG. FIG. 4 is a block diagram showing an example of the above. 1, 12 input terminals, 2 frame memories, 3, 26 subtraction circuits, 4 LPF, 5, 27 absolute values, 6, 28 nonlinear conversion circuits, 10 space-time Processing circuit, 11 output terminals,
13,14… Field memory, 15,20… Minimum value selection circuit, 16… AND circuit, 7,17,30 …… Noise removal circuit, 18,3
4 ... Sample memory, 21 ... Mode selection circuit, 22 ... B
PF, 23… ACC amplifier circuit, 24… Demodulation circuit, 25… 2
Frame memory, 29 ... Synthesis circuit. 31 ... Control circuit, 32
…… Adaptive noise elimination circuit, 33 …… Bit selection circuit, 35
... Judgment circuit, 36 ... Integration circuit, 37 ... Resistance R, 38 ...
Capacitor C, 39 ... 1H memory output enable terminal, 40
…… 1H memory output pin.

──────────────────────────────────────────────────続 き Continuing on the front page (72) Inventor Shigeru Hirahata 292 Yoshida-cho, Totsuka-ku, Yokohama-shi, Kanagawa Prefecture Inside the Home Appliances Research Laboratory, Hitachi, Ltd. (72) Inventor Kazuhiro Kaizaki 292 Yoshida-cho, Totsuka-ku, Yokohama-shi, Kanagawa (56) References JP-A-62-111586 (JP, A) JP-A-62-128288 (JP, A) JP-A-63-151282 (JP, A) JP-A-63-128 135094 (JP, A) JP-A-62-237879 (JP, A) JP-A-63-59188 (JP, A) JP-A-1-268293 (JP, A) JP-A-63-90988 (JP, A) JP-A-63-70680 (JP, A) JP-A-63-121375 (JP, A) JP-B-57-9558 (JP, B2) (58) Fields investigated (Int. Cl. ⁶ , DB name) H04N 9/78 H04N 7/00

Claims (4)

(57) [Claims] 1. A motion detection means for receiving a television signal, detecting a motion of an image based on a difference between frames of the television signal, and generating a motion information signal; Selecting means for comparing the information signal with the motion information signal of a pixel adjacent to the pixel in the vertical direction, and outputting a smaller value as the motion information signal of the certain pixel; A spatio-temporal processing means for processing the motion information signal in a time direction and a spatial direction. 2. A motion detection means for receiving a television signal, detecting a motion of an image based on a difference between frames of the television signal, and generating a motion information signal. Comparing the motion information signal with the motion information signal at a second pixel at the same horizontal position as the first pixel in a line one horizontal scanning cycle before the line including the first pixel; Selecting means for outputting the smaller value as the motion information signal of the first pixel; and spatio-temporal processing means for processing the motion information signal output from the selecting means in time and space directions. A control motion signal generation circuit characterized by the above-mentioned. 3. The motion detecting means includes: a frame memory for delaying a digitized television signal by one or two frames, a subtraction circuit for subtracting an input and an output of the frame memory, and an output of the subtraction circuit. A low-pass filter that removes a color signal component; an absolute value generation unit that generates an absolute value of an output of the low-pass filter; and a non-linear conversion circuit that generates the motion information signal by bit-compressing the output of the absolute value generation unit. The control motion signal generation circuit according to claim 1, further comprising: 4. A motion-adaptive signal processing circuit which receives a television signal, detects a motion of an image, generates a motion information signal, and performs television signal processing according to the motion information signal. A selecting means for comparing the motion information signal at a certain pixel of the signal with the motion information signal at a pixel adjacent to the pixel in the vertical direction, and selecting and outputting a smaller value as the motion information signal; A motion adaptive signal processing circuit for performing the television signal processing by using an output from the selection means as a motion information signal of the certain pixel.