JPH0429488A - Motion detecting circuit and picture processing device using this circuit - Google Patents

Abstract

PURPOSE:To prevent erroneous discrimination of motion detection by comparing picture element signals having one field period between them to detect the difference signal and comparing picture element signals having one frame period between them to detect the sum signal and checking amplitude relations between sample points in the preceding field and the preceding frame. CONSTITUTION:Output signals of threshold circuits 116 and 120 are inputted to an OR circuit 121. The output of this circuit 121 is inputted to an AND circuit 123 together with the outputs of threshold circuits 110 and 112. If an input signal consists of only color signals for a still picture, the output of the threshold circuit 110 is '1' because of the same picture, and the output signal of the threshold circuit 112 is '1' because color signals have opposite phases. One of outputs of threshold circuits 116 and 120 is '1', and the output of the OR circuit 121 goes to '1'. Consequently, all inputs of the AND circuit 123 are '1', and the output is '1', and the output of an OR circuit 124 is '1', thus indicating the still picture.

Description

DETAILED DESCRIPTION OF THE INVENTION Field of Industrial Application The present invention relates to a motion detection circuit that determines whether a video signal of an NTSC signal is a moving image or a still image, and a method for detecting the luminance of a video signal using this motion detection circuit. The present invention relates to an image processing device that separates signals and color signals.

Conventional technology In recent years, advances in semiconductor technology and transmission technology have expanded the possibilities of economical image communication, and as living standards have improved and social institutions have become more sophisticated, various image communication services have expanded. It is developing. Depending on whether the image signal is a still image or a moving image, the characteristics of the image signal are effectively utilized to simplify the image processing circuit and perform band compression.

Therefore, it is desired to improve the performance of a motion detection circuit, which is a circuit that reliably detects whether an image is a still image or a moving image.

FIG. 6 is a block diagram of a conventional motion detection circuit.

As shown in FIG. 6, the constituent elements 601 are low-pass filters (hereinafter referred to as LPF), 602° 607 60
8 is a frame memory (
(hereinafter referred to as FM), 603°609 is a subtractor, 604,
610 is an absolute value circuit, 605 and 611 are threshold circuits that determine the magnitude relationship between the input signal and the threshold value, 606 is a bypass filter (hereinafter referred to as HPF), and 612 is a motion determination circuit based on the outputs of the two threshold circuits 604 and 610. This is a judgment circuit.

Next, the related operations of the components in the conventional motion detection circuit configured as described above will be explained. terminal 6
NTSC signal is input from 00, LPF601 and H
It is input to PF606.

The LPF 601 passes the low frequency component of the NTSC signal, that is, the low frequency luminance signal, and inputs it to the FM 602 . Subtractor 6
In step 03, the non-delayed low-range luminance signal is subtracted from the low-range luminance signal delayed by one frame, and the absolute value of the output of the subtracter 603 is obtained in the absolute value circuit 604.

Since the two signals separated by one frame period are at the same position on the screen, when the image is a still image, these two signals are exactly the same, and the output of the absolute value circuit 604 is zero. Conversely, if there is movement in the image, the signals separated by one frame period will be different, and the absolute value circuit 604 will obtain a non-zero signal.

The threshold circuit 605 compares the output of the absolute value circuit 604 with a certain threshold value, and if the threshold value is smaller, it determines that there is movement and outputs "1°".On the other hand, if the threshold value is larger, it outputs "1°". Assuming that it is a still image, '°0'' is output.

One of the NTSC signals input from terminal 600 is HPF
606. The HPF 606 outputs a high frequency video signal in which a high frequency/luminance signal and a carrier color signal are mixed. This high frequency video signal is input to two FM607608Ls in sequence.

In the subtracter 609, the undelayed high frequency video signal is subtracted from the delayed high frequency video signal. Since the phase of the carrier color signal is inverted for each frame, they are in phase when two frames apart. The output from the subtracter 609 is input to the absolute value circuit 610 and the threshold circuit 611 as before, and the output from the threshold circuit 611 is "O" for a still image and "B" for a moving image. 0 judgment circuit 612 is threshold circuit 6
05 or when the output of the threshold circuit 611 is "Bi", it outputs "Bi". That is, if motion is detected in either the low-frequency luminance signal or the high-frequency video signal, it is determined that there is motion, and "Vi" is output.

FIG. 7 shows an embodiment of a second conventional motion detection circuit. As shown in FIG. 7, the components include a 263H delay 701 for delaying 263 lines, 702 and 705 a subtracter, 704 an FM, 703706 an absolute value circuit, and 707 a threshold circuit.

Next, the related operations of the constituent elements in the motion detection circuit configured as described above will be explained. NTSC from terminal 700
When a signal is input, the subtracter 702 subtracts the undelayed signal from the 262-line delayed signal.

Since the phase of the carrier color signal of the signal 262 lines before is the same, the output from the subtracter 702 includes a two-dimensional vertical signal difference and a signal difference in the case of a moving image.

This difference signal is input to an absolute value circuit 703, and its absolute value is obtained. Also, the output of the absolute value circuit 703 is FM705
The subtracter 705 obtains the difference between the signal delayed by one frame period and the signal not delayed. The difference signal that is the output of the subtracter 705 is input to an absolute value circuit 706 to obtain its absolute value, and is input to a threshold circuit 707.

FIG. 4 shows the relationship between the sample points of each field and the phase of its subcarrier. In FIG. 4, SO is a sample point of field n, and Sl and S2 are sample points of field n+1. Sl is one line above 52 on the screen. Further, S3 is a sample point of the n+2 field, and S4 is a sample point of the n+3 field.

The direction of the arrow at each sample point indicates the phase of the subtransport, So, Sl, S5. S6 is in phase, S2. S3. S4
is also in phase, but in opposite phase to SO.

If the signal input from terminal 700 is a still image,
The difference signal output from the subtracter 702 is SO in FIG.
and 31, and indicates only the difference in signals in the vertical direction within two dimensions. In the case of a still image, this two-dimensional difference signal corresponds to the difference between S3 and S34 in the next frame as well. ), the output from the subtracter 705 is zero.

Furthermore, if the signal input from the terminal 700 is dynamic,
The difference signal from subtractor 702 includes vertical difference and motion information in two dimensions. In the next frame, the output from the subtracter 705 will not be zero since it will be a difference signal of a different picture.

FIG. 8 shows an image processing apparatus that performs conventional luminance signal and color signal separation using the conventional motion detection circuit described above.

As shown in FIG. 8, as a component, 801 is a frame filter that separates color signals using a pixel that is one frame period apart from the pixel of interest, and 803 is a frame filter that uses neighboring pixels in the same field as the pixel of interest. Separate color signals 2
Dimensional filter (hereinafter referred to as 2D filter) L 806 is a conventional motion detection circuit, 802 is a switching circuit, 804 is H
PF, 805 is a subtracter.

Next, the related operations of the constituent elements of the conventional image processing apparatus for separating luminance and color signals configured as described above will be explained. When a video signal is input from a terminal 800, a frame filter 801 separates a color signal CI, and a 2D filter 803 separates a color signal C2. C1 is completely separated in still images, but in videos it includes new deteriorations such as false contours, dot interference, and cross color that are not present in C2. Although it is not a signal that has been degraded, it does not include any new deterioration.

Therefore, the switching circuit 802 is activated by the output of the conventional motion detection circuit 806 to select C1 for a still image and C2 for a moving image, thereby selecting the color signal with the least deterioration. Selected color signal is HPF with switching Kaimuro 802
The signal is band-limited to only the 804 high frequency range, and subtracted from the original signal by a subtracter 805 to obtain a luminance signal.

Problem to be Solved by the Invention However, when a vertical striped pattern consisting of high-frequency components as shown in FIG. The shading of the striped pattern in the field is just reversed. FIG. 5(B) shows this state in detail.

FIG. 5(B) shows n, n+1. n+2. The horizontal sample points of the n+3 field are shown.

As shown in the n field, the vertical stripe pattern repeats every four white and black sample points. If the sample points move horizontally two by two in one field, the pattern will be such that the white and black of the n field are reversed in the n+2 field.
It has the same phase relationship as the color signal.

Furthermore, the four sample points at the same position in the horizontal direction of each field are given a code of 33.4 from the SOI as shown in the figure.

When this signal is input to the motion detection circuit shown in FIG. 6, since the vertical stripe signal has only high frequency components, the output of the LPF 601 becomes zero, and motion detection must be performed using only the output of the HPF 606. As shown in FIG. 5(B), this signal is 1
Since the signals are inverted in each frame, signals two frames apart will be in phase.

Therefore, the output from the subtracter 609 becomes zero, and this signal is erroneously determined to be a still image.

Furthermore, when this signal is input to the motion detection circuit shown in FIG.
When field 522 is input, the output of the subtracter becomes zero since these two sample points are in phase. FM
The output from 704 is the absolute value of the difference between S12 of the n+1 field and 502 of the n field, which is also zero.

Therefore, the output of subtractor 705 also becomes zero.

When 333 of the n+3 field and 523 of the n+2 field are input to the subtracter 702, the output of the subtracter 702 becomes a certain value because these two sample points are in opposite phases.

However, the output of the FM 704 is also the absolute value of the difference between S13 and 303, and is at the same level as the absolute value of the difference between S33 and 323, so the output of the subtracter 705 is zero.

In this way, even the motion detection circuit shown in FIG. 7 incorrectly determines that the signal shown in FIG. 5 is a still image.

As described above, the conventional motion detection circuits shown in FIGS. 6 and 7 have a problem in that when a signal as shown in FIG. 5 is input, it is incorrectly determined to be a still image.

Furthermore, in the conventional image processing apparatus that performs luminance signal and color signal separation as shown in FIG. 8 using this conventional motion detection circuit, the fifth
When the signal shown in the figure is input, the frame filter 801 is used even though it is a moving image, resulting in problems such as dot interference and false contours. The present invention has been made with the above-mentioned problems in mind, and an object thereof is to provide a motion detection circuit that does not cause erroneous motion detection judgments and an image processing circuit that does not cause deterioration such as dot interference or false contours.

Means for Solving the Problems In order to achieve the above object of the present invention, the relationship between the amplitudes of the pixel signal of interest, the pixel signal 262 lines before, the pixel signal 263 lines before, and the pixel signal one frame before is determined. It is equipped with a plurality of threshold circuits to be examined and a logic circuit that detects movement from the outputs of these plurality of wj value circuits, and is a discretized NTSC
This is a motion detection circuit that detects the motion of a signal image. The present invention also provides an image processing device that performs luminance signal color signal separation and is characterized by being equipped with this motion detection circuit.

The configuration includes a first detection means that receives a discretized so-called digitized NTSC signal as an input signal and obtains a first neighboring pixel signal that is one field period apart from the pixel signal of interest, and a pixel signal of interest. a second detecting means for obtaining a second neighboring pixel signal separated by one frame period from the pixel signal; a first comparing means for obtaining a correlation between the pixel signal of interest and the first neighboring pixel signal; A second comparing means is provided to obtain a correlation between the signal and a second neighboring pixel signal, and a movement of an image of the input signal is detected from the output of the first comparing means and the output of the second comparing means. This is a motion detection circuit having determination means.

In particular, as another configuration, the first detection means is a motion detection circuit including detection means for obtaining a pixel signal 262 lines before and a pixel signal 263 lines before the pixel of interest.

In addition, a frame filter that takes a discretized NTSC signal as an input signal and separates a luminance signal and a color signal using a pixel signal that is one frame period apart from the pixel signal of interest, and a frame filter that uses a pixel signal of interest and a pixel signal that is one frame period apart, and a two-dimensional filter that separates a luminance signal and a color signal using a pixel signal; and a switching circuit that selects the output of either the frame filter or the two-dimensional filter using the motion detection circuit of the present invention. This is an image processing device that performs color signal separation.

Operation The motion detection circuit of the present invention having the above configuration compares pixel signals separated by one field period and detects, for example, a difference signal therebetween, and compares pixel signals separated by one frame period and detects, for example, their sum signal. do.

Since the difference signal between l fields is the same image if it is a still image, the difference signal of pixel signals whose color signals are matched in phase becomes zero.

The sum signal between one frame is the sum of signals whether it is a still image or a moving image, and therefore generally does not become zero. However, in the case of an image containing only color signals, the phase is reversed every l frames, so this sum signal becomes zero. Furthermore, if the movement of the image is such that the shading is reversed every frame, this sum signal will be zero. The second comparing means is used to detect such a special image, and the output signal can control whether to operate the first comparing means or another motion detecting means.

Further, the first detection means detects the pixel signal of interest and the second detection means.
By detecting the pixel signal 62 lines before and the pixel signal 263 lines before, the first comparing means compares the pixel signal of interest with the pixel signal 262 lines before, and 263 lines before.
By comparing the pixel signal of the previous line and the pixel signal of one frame before, the pixel signals whose color signals match are matched, so that accurate image movement can be detected.

Furthermore, if the output of the motion detection circuit of the present invention is used as a signal for circuit switching between video processing and still image processing in an image processing device, motion can be reliably detected even in the case of special images, resulting in more accurate image processing. will be held.

Example FIG. 1 is a block diagram of a first embodiment of the present invention.

As shown in FIG. 1, the component 102 is a 262H delay that delays 262 lines, so-called 1 field.
103 is an IH delay that delays the I line, 104 is 2
262H delay that delays 62 lines, 105 is an adder, 106°114.118 is f(PF, 107 is FM
, 108113.117 is a subtracter, 109. III.

115.119 is an absolute value circuit, 110, 112116.
120 is a threshold circuit, 122.123 is an AND circuit, 12
1, 124 is an OR circuit, and 125 is a smoothing circuit.

Next, the related operations of the components in the motion detection circuit configured as described above will be explained.

When an NTSC signal is input from terminal 101, one of the signals is 262H delay 102, IH delay 103, 2
The signals are sequentially input to the 62H delay 104 and operate as a means for being delayed by exactly one frame period.

Here, if the input signal is 33 in Fig. 4, then 262
The output of the H delay 102 is S2, the output of the IH delay 103 is 51.262, and the output of the H delay 104 is SO. The adder 105 obtains a sum value of the l-frame delayed signal and the undelayed signal. This additional value is HPF
106 and extracts only the high-frequency component, the absolute value is determined in the absolute value circuit 111, and a comparison corresponding to the first comparison means is performed in the threshold value circuit 112, and the input signal is compared with the threshold value. When it is small, "1°" is output.The threshold circuit 112 outputs "1" when the signals separated by one frame period have the same amplitude but opposite phases.
This applies when there is only a color signal in a still image, or when a luminance signal moves at the same speed as a pattern in which the shading is reversed in exactly one frame, as shown in FIG.

The output of the 262H delay 104 is input to the FM 107 and delayed by one frame period. The difference value between the input signal of the terminal 101 and the output of the FM\107 delayed by two frame periods is obtained in the subtraction circuit 108, the absolute value of the difference value is obtained in the absolute value circuit 109, and compared with a threshold value in the threshold value circuit 110. This subtracter 10 outputs “B” when the value is smaller than the threshold value.
8, the absolute value circuit 109, and the threshold value circuit 110 have the same configuration as a normal motion detection circuit, and output "1" when the image is a still image.

Input signal from terminal 101 (referred to as 41st DS3) and 2
Output of 262H delay 102 separated by 62H period (4th
32) is obtained by the subtraction circuit 113, and
At 114, the high-frequency component is extracted, and at an absolute value circuit 115, the absolute value is determined.A comparison corresponding to the second comparison means is performed at a threshold circuit 116, where it is compared with a threshold value, and when it is smaller than the threshold value, 1° is set. Output.

The difference value between the output signal of the IH delay 103 (represented as 31 in FIG. 4) and the output of the 262H delay 104 separated by 262H period (represented as SO in FIG. 2) is determined by the subtraction circuit 117, and the HP
FLLB extracts the high frequency component and sends it to the absolute value circuit 119.
Find the absolute value and compare it with the threshold value of the threshold circuit 120,
Similar to this comparison in which l'' is output when the value is smaller than the threshold value, this is a comparison between one field.

The output signals of the threshold circuit 116 and the threshold circuit 120 are input to the OR circuit 121.

The output of the OR circuit 121 is input to the AND circuit 123 along with the outputs of the threshold circuit 110 and the threshold circuit 112. If the input signal is a still image with only color signals, the output of the threshold circuit 110 will be "1" because the screen is the same, and the output signal of the threshold circuit 112 will also be "1" because the color signals are in reverse phase. Become. Also, 53 and 32 in Figure 4 or Sl
Generally, there is a strong correlation between SO and SO, so the output of either subtractor circuit 113 or 117 will be almost zero, and therefore either of the outputs of threshold circuits 116 and 120 will be 'BI', and the OR circuit The output of 121 becomes 1'.

Therefore, all the inputs of the AND circuit 123 become "B", the output becomes "B", and the output of the OR circuit 124 also becomes "1".
, indicating that it is a still image. As described above, the OR circuit 121124 and the AND circuits 122 and 123 operate as determination means for motion detection.

When the input signal from the terminal 101 is a special moving image as shown in FIG. 5, the outputs of the threshold circuit 110 and the threshold circuit 112 become "1". However, when the input signal to the terminal 101 is 321 or 22 in FIG. circuits 116 and 120
Both outputs are "0".

Therefore, the output of the OR circuit 121 becomes "0",
The output of the AND circuit 123 also becomes "0°", indicating that it is a moving image.

The output of the threshold circuit 110 and the inversion of the output of the threshold circuit 112 are input to the AND circuit 122 . This circuit provides the output of a conventional motion detection circuit when signals separated by one frame period (corresponding to SO and 33 in FIG. 2) have different amplitudes or different phases.

The outputs of the AND circuits 122 and 123 are input to the OR circuit 124. In the case of a special moving image as shown in FIG. 5 where the conventional motion detection circuit shown in FIG. 3 would make a mistake, the output of the OR circuit 124 is blocked by the AND circuit 122 and the output of the conventional motion detection circuit is The correct judgment result is the output of the circuit 123, and otherwise the conventional motion detection method is used, that is, the output of the threshold circuit 112 operates as a switching signal.

The output of the OR circuit 124 is input to a smoothing circuit 125. If the input signal to the terminal 101 is the fifth signal, it is determined that it is a moving image at half of the sample points, so the smoothing circuit 125 correctly determines that it is a moving image.

Furthermore, even if the vertical striped signal in FIG. 5 contains a DC component, the first motion detection circuit in FIG. 1 operates in the same manner as described above and still correctly determines that it is a moving image.

FIG. 2 is an embodiment of a second motion detection circuit showing another configuration of the motion detection circuit.

The difference from the motion detection circuit in FIG. 1 is that the IH delay 103 and 262H delay 104 in FIG.
The difference is that the H delay 202 and the subtraction circuit 117, HPF 118, absolute value circuit 119, threshold circuit 120, and OR circuit 121 in FIG. 1 are omitted.

In the second motion detection circuit, by replacing the OR circuit 121 with a subtraction circuit 206, HPF 114, absolute value circuit 115, and threshold circuit 116, even if the moving image shown in FIG. Since the signal difference can be compared under the condition that the phases of the color signals are in the same phase, it is correctly determined that the video is a moving image.

Further, even if the fir stripe signal in FIG. 5 contains a DC component, the second motion detection circuit in FIG. 2 operates in the same manner as described above and correctly determines that it is a moving image.

FIG. 3 shows an image processing apparatus for separating luminance signals and color signals, which is a third embodiment of the present invention.

As shown in FIG. 3, the constituent elements 301 are a frame filter that separates color signals using a pixel that is one frame period apart from the pixel of interest; 303 is a 2D filter that separates color signals; and 306 is a frame filter that separates color signals using The motion detection circuit of the first embodiment or the second embodiment, 302 is a switching circuit, 304
is an HPF, and 305 is a subtraction circuit.

Next, the related operations of the constituent elements in the image processing apparatus that performs luminance signal and color signal separation configured as described above will be explained. When a video signal is input from the terminal 300, the frame
A filter 301 separates the color signal C1, and a 2D filter 303 separates the color signal C2. C1 is completely separated in still images, but in videos it includes new deteriorations such as false contours, dot interference, and cross color that are not present in C2.
On the other hand, C2 is not a completely separated signal regardless of whether it is a still image or a moving image, but it does not include any new deterioration.

Based on the output of the motion detection circuit 306, the switching circuit 302 selects C1 for still images and C1 for moving images.
By selecting 2, the color signal with the least deterioration is selected. The color signal selected by the switching circuit 302 is HPF3.
04, the signal is band-limited to only the high frequency range, and is subtracted from the original signal in a subtracter 305 to obtain a luminance signal.

In this way, since motion detection is performed using the motion detection circuit 306 of the first embodiment or the second embodiment, there is no error.
Therefore, even though it is a moving image, it does not suffer from deterioration such as false contours or dot interference due to the use of a frame filter.

Effects of the Invention As is clear from the above explanation, according to the present invention, even if a moving image is determined to be a still image by conventional motion detection methods, the sample point of interest, the previous field, and the previous frame can be detected. By examining the relationship between the amplitudes of the sample points, false judgments in motion detection can be prevented, and this motion detection circuit can be used to switch between a frame filter and a 2D filter in an image processing device that separates luminance and color signals. Therefore, it is possible to realize an image processing apparatus that can use an optimal filter and is free from deterioration such as dot interference and false contours, which has great practical effects.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing the configuration of a motion detection circuit according to a first embodiment of the present invention, FIG. 2 is a block diagram showing the configuration of a motion detection circuit according to a second embodiment of the present invention, and FIG. A block diagram showing the configuration of an image processing apparatus that performs luminance signal and color signal separation according to the third embodiment of the present invention, FIG. 4 is an explanatory diagram showing the phase of sub-carrying of pixels of each field, and FIG. 5 (A) , (B) is an explanatory diagram showing the characteristics of a certain moving image, Figures 6 and 7 are block diagrams showing the configuration of a conventional motion detection circuit, and Part 8 is a conventional image processing device that separates luminance and color signals. FIG. 2 is a block diagram showing the configuration of FIG. 101... Input terminal, 102 104...
...262H delay, 103...IH delay, 105...addition circuit, 107...
Frame memory, 108.113.117...
Subtraction circuit, 109111.115.119...-
Absolute value circuit, 110°112.116.120...
...Threshold circuit, 122°123...AND circuit, 121.124...OR circuit, 105...
...Addition circuit, 108.113117...
Subtraction circuit. Name of agent: Patent attorney Shigetaka Awano

Claims (8)

[Claims] (1) A first detection means that uses a discretized NTSC signal as an input signal and obtains a first neighboring pixel signal that is one field period apart from the pixel signal of interest;
a second detection means for obtaining a pixel signal in a second neighborhood separated by a frame period; a first comparison means for obtaining a correlation between the pixel signal of interest and the pixel signal of the first neighborhood; a second comparing means for obtaining a correlation between the pixel signal and the second neighboring pixel signal, and determining the movement of the image of the input signal from the outputs of the first comparing means and the second comparing means. A motion detection circuit having a determination means for detecting. (2) The first detection means includes a detection means for obtaining a pixel signal 262 lines before the pixel of interest, and the first comparison means has a signal difference between the pixel signal of interest and the pixel signal 262 lines before the pixel of interest. 2. The motion detection circuit according to claim 1, further comprising a comparison means for comparing the value and the predetermined value. (3) The motion detection circuit according to claim 1, wherein the second comparing means comprises means for comparing a signal sum of a pixel signal of interest and a pixel signal separated by one frame period with a predetermined value. (4) a detection means that uses a discretized NTSC signal as an input signal and obtains a pixel signal of interest and a pixel signal of 262 lines and one frame before from the pixel of interest; A first threshold circuit that compares a signal sum of pixel signals with a predetermined value; and a second threshold circuit that compares a signal difference between the pixel signal of interest and the pixel signal 262 lines before the predetermined value. and determining means for determining that there is movement in the image of the input signal when the signal sum in the first threshold circuit is smaller than a predetermined value and the signal difference in the second threshold circuit is larger than a predetermined value. motion detection circuit. (5) detection means that uses a discretized NTSC signal as an input signal and obtains a pixel signal of interest and a pixel signal of 263 lines and one frame before from the pixel of interest; a first threshold circuit that compares the signal sum of the pixel signals with a predetermined value; and a second threshold circuit that compares the signal difference between the pixel signal 263 lines before and the pixel signal one frame before the predetermined value. and determining that there is movement in the image of the input signal when the signal sum of the first threshold circuit is smaller than a predetermined value and the signal difference of the second threshold circuit is larger than a predetermined value. A motion detection circuit having means. (6) detection means that uses a discretized NTSC signal as an input signal and obtains a pixel signal of interest and a pixel signal of 262 and 263 lines and one frame before the pixel of interest; the pixel signal of interest and the one frame; a first threshold circuit that compares the signal sum of the previous pixel signal with a predetermined value; and a second threshold circuit that compares the signal difference between the pixel signal of interest and the pixel signal 262 lines before the predetermined value. , a third threshold circuit that compares a signal difference between the pixel signal of interest and the pixel signal 263 lines before the pixel signal with a predetermined value, and the signal sum in the first threshold circuit is smaller than the predetermined value; A motion detection circuit comprising determining means for determining that there is movement in the image of the input signal when either the second or third threshold circuit has a signal difference greater than a predetermined value. (7) A frame filter that uses a discretized NTSC signal as an input signal and separates a luminance signal and a color signal using a pixel signal that is one frame period apart from the pixel signal of interest; a two-dimensional filter that separates a luminance signal and a color signal using neighboring pixel signals; a motion detection circuit that detects motion of the image of the input signal; and an output of the motion detection circuit that separates the frame filter and the two-dimensional filter. a switching circuit that switches and outputs one of the outputs, the motion detection circuit according to claim (1), (4), (5), (
An image processing device that performs luminance signal color signal separation, which is the motion detection circuit according to any one of item 6). (8) A detection means which uses a discretized NTSC signal as an input signal and is the detection means according to claim (6) with an additional function of obtaining a pixel signal two frames before from a pixel of interest; A threshold circuit that compares a difference signal between a pixel signal and a pixel signal of interest with a predetermined value; the first, second, and third threshold circuits according to claim (6); the output of the threshold circuit; a first AND circuit whose input is the inverted output of the first threshold circuit according to claim 6); and a first logical sum circuit whose inputs are the outputs of the second and third threshold circuits according to claim (6). a second AND circuit that receives three outputs: the output of the first OR circuit, the output of the threshold circuit, and the output of the first threshold circuit according to claim (6); A motion detection circuit, comprising: a second AND circuit that receives the outputs of the first and second AND circuits as inputs, and uses the output of the second OR circuit as a motion detection signal.