JP2003116109A - Motion detection fr for interlace video signal and progressive scanning converter employing the same - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a progressive scanning converter capable of applying excellent progressive scanning conversion to an interlace video signal subjected to pull-down conversion without causing mis-interpolation even when the interlace video signal partially or temporarily causes a pattern deviated from a regular pattern. SOLUTION: A pattern detection circuit 4 detects a pull-down conversion (state 1) where an original frame is shared to a succeeding field or a pull-down conversion (state 2) where the original frame is shared to a preceding field. A motion detection circuit 5 detects a motion by selecting a motion detection method according to the state 1, or the state 2 or a state 3 not being a pull- down conversion state. A selection circuit 6 selects the succeeding field f0 in the case of the state 1 or selects the preceding field f2 in the case of the state 2. A mixing circuit 7 mixes a moving picture interpolation signal M with a still picture interpolation signal S according to a motion detection signal K.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to motion detection suitable for detecting the motion of an interlaced video signal converted from a signal source having a low frame rate into an interlace by pulldown such as 2-3 pulldown or 2-2 pulldown. The present invention relates to an apparatus and a progressive scan conversion apparatus using the apparatus.

[0002]

2. Description of the Related Art Standard television signals such as NTSC signals and HDTV signals are interlaced (interlaced scanning).
It is a signal. The interlace signal causes interlace interference such as line flicker when there are many high frequency line segments in the vertical direction of the image. Therefore, there is a processing method for removing interlace interference by interpolating a scanning line of a portion thinned out by interlacing with a peripheral scanning line and converting it into a sequential scanning signal (also called a non-interlaced signal or a progressive signal). . Such a processing device is called a progressive scan conversion device.

An outline of scanning line interpolation for progressive scanning conversion will be described below. 5A and 5B are diagrams showing a scanning line structure. FIG. 5A shows an interlaced signal, and FIG. 5B shows a signal obtained by sequentially converting the interlaced signal into a scanning signal by scanning line interpolation. In FIG. 5, a circle indicates a transmitted scan line, and a cross indicates an interpolated scan line. In FIG. 5 and FIG. 6 described later, V is the vertical direction, and t is the time direction.

Scan line interpolation for the most general progressive scan conversion is performed by a motion adaptive process in which the interpolation method is changed according to the result of motion detection of an image. That is, as shown in FIG. 6, when the image is stationary, the pixel F in the rear field is
01, an average value of the pixels F21 in the previous field or one of them is set as a new pixel Y indicated by x, and a new scanning line is generated. This is called inter-field interpolation or still image interpolation. If the same interpolation method as for a still image is used when the image is moving, the image becomes a multiple image. Therefore, when the image is moving, the upper and lower pixels F10 and F1
A new scan line is generated by setting the average value of 2 as a new pixel Y indicated by x. This is called intra-field interpolation, line interpolation, or moving image interpolation.

At this time, when the image is stationary, a good conversion image quality with less aliasing distortion and high resolution can be obtained. However, when the image is moving, there is a lot of aliasing distortion and the resolution is low, resulting in degraded conversion image quality. . In addition, it is generally difficult to design a motion detection circuit that detects whether an image is moving or stationary as a circuit that does not cause false detection.

By the way, when the input signal to be converted into a progressive scanning signal is a signal converted from a projection film having a frame rate of 24 frames / sec into an interlace by, for example, 2-3 pulldown, the above-described motion adaptive processing is performed. By adopting a different method, it is possible to obtain a good converted image quality even when the image moves.

The 2-3 pulldown is a frame rate conversion as shown in FIG. Specifically, FIG.
The film video signals A, B, C, ... Of 24 frames / sec of a movie shown in (A) are converted into interlaced signals a, a ', b, b', b of 60 fields / sec shown in FIG. 7B. ，
It is used as a method for converting into c ′, c ... It should be noted that the a ', b' ... Symbols "'" are attached to the even fields, and those not attached "'" are the odd fields. In addition to 2-3 pulldown, there is 2-3 pulldown used for a signal having a frame rate of 30 frames / sec. In this specification, 2-3 pulldown will be described as a representative.

As shown in FIG. 7, 2-3 pulldown is
An image that was originally one frame is distributed to two or three fields. In FIG. 7B, for example, field a and field a ′ are the same image although their scanning lines are deviated from each other. Therefore, if only the rule pattern for this conversion is known, it is possible to interpolate the scanning lines from the adjacent fields (inter-field interpolation) and convert the scanning lines into sequential scanning. When a 60 field / sec interlaced signal is converted into progressive scan, 60 frame / sec progressive scan signals A, A, B, B, B, C, C shown in FIG.
… Will be.

As described with reference to FIG. 6, in the inter-field interpolation, a new scanning line is generated by setting the pixel F21 of the previous field or the pixel F01 of the subsequent field as a new pixel Y, so that the aliasing distortion is small and the resolution is high. Good conversion image quality can be obtained. The method of obtaining the 2-3 pulldown rule pattern is automatically performed by, for example, transmitting a method of distributing and converting by a code or statistically processing a difference between fields or frames of an image. There is a method of guessing a rule pattern.

[0010]

However, the above method of converting to progressive scanning using the pull-down rule pattern causes a problem in the following cases.

(1) In the case where an additional signal such as a subtitle is superimposed on an image having pulldown regularity by editing and there is an image partially deviating from the pulldown rule pattern. (2) Picture with pulldown regularity
In-picture (PIP) is used to fit an image that is out of the regular interlaced signal or pull-down rule pattern, and partially has an image that is out of the pull-down rule pattern. (3) When the regular pattern of the pulled-down video is temporarily disturbed by the splice editing.

In the cases of (1) and (2) above, even when transmitting the pull-down video regular pattern by a code,
Even if the rule pattern is detected by statistical processing, it becomes a big problem. Further, the above (3) is less likely to cause a problem when the pull-down video rule pattern is transmitted by a code. However, when the rule pattern is detected by statistical processing, the rule pattern detection result is corrected. During that time, it becomes a problem because it becomes a sequential scanning conversion by erroneous interpolation.

The present invention has been made in view of the above problems, and when an interlaced video signal converted into an interlace by pulldown has a partial or temporary deviation of a regular pattern, an appropriate motion detection signal is generated. Providing a motion detection device capable of outputting, even if the interlaced video signal converted to interlace by pulldown has a partial or temporary deviation of a regular pattern, it is possible to perform good progressive scan conversion without erroneous interpolation. An object is to provide a scan conversion device.

[0014]

In order to solve the above-mentioned problems of the prior art, the present invention provides (a) a motion detecting device for detecting the motion of an interlaced video signal, wherein the interlaced video signal has a predetermined frame rate. Pulldown detecting means (4) for detecting whether or not the frame image having the following is an interlaced video signal converted to interlace by pulldown, and an interlaced video in which the interlaced video signal is converted to interlace by pulldown by the pulldown detection means. When it is detected as a signal, it is in the first state in which the field of interest and the rear field which is the field one field after the field of interest are common frame images, or the field of interest and the field of interest. One field before Determination means (4) for determining whether or not the front field, which is a field, is a second state in which a common frame image is formed, and signals of the target field, the rear field, and the front field are input, The motion according to the first state, the second state, and the third state in which the pull-down detection unit detects that the interlaced video signal is not the interlaced video signal converted into the interlace by pull-down. Provided is a motion detection device characterized by comprising a motion detection circuit (5) for switching a detection method and detecting a motion, and (b) progressive scan conversion for converting an interlaced video signal into a progressive scan video signal. In the apparatus, the interlaced video signal is pulled down from a frame image having a predetermined frame rate. Pull-down detection means (4) for detecting whether or not the interlaced video signal is converted into interlaced video signal by the pull-down detection means, and the pull-down detection means detects that the interlaced video signal is an interlaced video signal converted into interlaced video signal by pull-down. In this case, the field of interest and the rear field which is a field one field after the field of interest are in the first state in which they are common frame images, or the field of interest and the field one field before the field of interest. Determining means (4) for determining whether or not the previous field is a second state in which a common frame image is formed, and an in-field interpolating means for generating an in-field interpolating signal using the signal of the field of interest. (3) and in the first state,
A first inter-field interpolation signal is generated using the signal of the subsequent field for the field of interest, and in the second state, a signal of the previous field for the field of interest is used to generate an inter-second field signal. Inter-field interpolating means (6) for generating an interpolating signal, the signals of the target field, the rear field, and the front field are input, and the first state, the second state, and the pull-down detection means. According to the third state in which the interlaced video signal is detected as not the interlaced video signal converted to the interlaced by pulldown, the motion detection method is switched to detect the motion, and the motion detection circuit outputs the motion detection signal. (5)
And an interpolating signal output means (7) for mixing or switching the output of the intra-field interpolating circuit and the output of the inter-field interpolating circuit according to the motion detection signal and outputting the mixed signal. A progressive scan conversion device is provided.

[0015]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An interlaced video signal motion detecting apparatus and a progressive scan converting apparatus using the same according to the present invention will be described below with reference to the accompanying drawings. FIG. 1 is a block diagram showing an embodiment of a motion detecting apparatus for interlaced video signals and a progressive scanning conversion apparatus using the same according to the present invention, and FIG. 2 is a block showing a concrete configuration example of a motion detecting circuit 5 in FIG. FIG. 3 is a block diagram showing another specific configuration example of the motion detection circuit 5 in FIG. 1, and FIG. 4 is a waveform diagram for explaining the operation of FIG.

In FIG. 1, the interlaced video signal input from the input terminal 1 is stored in the field memories 21 and 2.
It is sequentially delayed by 2. It is assumed that the interlaced video signal input from the input terminal 1 is a field f0, the output of the field memory 21 is a field f1, and the output of the field memory 22 is a field f2. In the configuration of FIG. 1, since the field f0 is treated as a reference field by delaying the field f1 by one field, the field f1 will be referred to as a target field, the field f0 as a rear field, and the field f2 as a front field.

The moving image interpolation circuit 3 has a field of interest f1.
Is entered. The moving image interpolation circuit 3 generates a moving image interpolation signal M using the signal of the field of interest f1. The method of generating the moving image interpolation signal M by the moving image interpolation circuit 3 is not particularly limited, but as described with reference to FIG. 6, a method of taking the average value of the scanning lines located above and below is generally used.

The field of interest f1 and the rear field f0 are input to the pattern detection circuit 4. The pattern detection circuit 4 detects a 2-3 pulldown rule pattern and outputs a pattern detection signal P. The pattern detection circuit 4 is
The 2-3 pulldown rule pattern is detected by statistically calculating the integrated value of the difference between the input target field f1 and the subsequent field f0 over several to several tens of fields. Since the circuit configuration itself for detecting the pull-down rule pattern by statistical processing is known, the specific configuration of the pattern detection circuit 4 will be omitted. The method of detecting the rule pattern is not limited to this.

The pattern detection signal P output from the pattern detection circuit 4 has at least the following three states. (State 1) A state in which the input video signal is pull-down converted, the original frame is distributed to the rear field, and it is determined that the target field and the rear field are common frame images. (State 2) A state in which the input video signal is pull-down converted, the original frame is distributed to the previous field, and it is determined that the target field and the previous field are common frame images. (State 3) A state in which it is determined that the input video signal is not a pull-down converted signal.

In the case of 2-3 pulldown, there is a state in which the original frame is distributed to both the rear field and the front field, but here it is treated as (state 2). Of course, there is no problem in handling as (state 1).

The back field f0, the attention field f1 and the front field f2 are input to the motion detection circuit 5. The pattern detection signal P is also input to the motion detection circuit 5. The motion detection circuit 5 switches the motion detection method according to the pattern detection signal P, that is, according to the three states (state 1) to (state 3) described above. The motion detection circuit 5 outputs a motion detection signal K as a result of motion detection. The specific configuration of the motion detection circuit 5 and the operation of switching the motion detection method will be described later.

The rear field f0 and the front field f2 are input to the selection circuit 6. The selection circuit 6 is responsive to the pattern detection signal P for the rear field f0 and the front field f2.
Is selected and the selected signal is output as the still image interpolation signal S. The selection circuit 6 selects the rear field f0 when the pattern detection signal P represents the above (state 1), and selects the front field f2 when it represents the (state 2). To represent (state 3), either the rear field f0 or the front field f2 may be selected, but here, the front field f2 is selected.

As shown by the broken line in FIG. 1, the motion detection circuit 5 and the selection circuit 6 constitute the motion detection device of the present invention.

The moving picture interpolation signal M, the still picture interpolation signal S and the motion detection signal K are input to the mixing circuit 7. Mixing circuit 7
Outputs the interpolation signal H by adaptively mixing the moving image interpolation signal M and the still image interpolation signal S according to the motion detection signal K. If the motion detection circuit 5 determines that the image is completely still, K = 0, if it is determined that the image is completely moving, K = 1, and if it is intermediate, 0 <K <1, then H = S × ( 1−K) + M × K (1)

A selection circuit may be provided instead of the mixing circuit 7, and the moving image interpolation signal M and the still image interpolation signal S may be selectively switched according to the motion detection signal K. However, the mixing circuit 7 is preferable.

The target field f1 which is a non-interpolation signal is input to the double speed processing circuit 81, and the double speed processing circuit 81 doubles and outputs the signal of the target field f1. The interpolation signal H output from the mixing circuit 7 is supplied to the double speed processing circuit 82.
Is input, the double speed processing circuit 82 doubles the speed of the interpolation signal H and outputs it. Then, the selection circuit 9 uses the double speed processing circuit 81.
And the output of the double speed processing circuit 82 are alternately selected,
The progressive scan signal is output. Output terminal 10 for progressive scan signals
Will be output.

One of the features of the present invention is that the motion detection circuit 5
But at least 3 according to the pull-down rule pattern
It has two operation modes (motion detection method). In the past, when a pull-down rule pattern was detected, motion detection was completely prohibited and moving image interpolation was completely disabled.However, in the present invention, when a pull-down rule pattern is detected, motion detection is performed. Instead of completely prohibiting it, partially activate motion detection and partially activate video interpolation.

Here, a specific configuration example of the motion detection circuit 5 will be described with reference to FIG. In FIG. 2, the rear field f0 and the front field f2 are input to the frame difference detection circuit 51. The frame difference detection circuit 51 is
The difference between the rear field f0 and the front field f2 is detected, and the frame difference detection signal D1 is output. The frame difference detection signal D1 includes a rear field f0 and a front field f.
It is the absolute value of the difference from 2, and D1 = | f2-f0 | (2) The frame difference detection circuit 51 uses the rear field f
It is an inter-frame motion detector that detects a motion between two fields (one frame) based on 0 and the previous field f2.

The rear field f0 and the field of interest f1 are input to the field difference detection circuit 50. The field difference detection circuit 50 detects the difference between the rear field f0 and the target field f1 and outputs a field difference detection signal d0. The field difference detection signal d0 is
It is an absolute value of the difference between the rear field f0 and the field of interest f1, and d0 = | f1-f0 | (3) The field difference detection circuit 50 is an inter-field motion detection unit that detects a motion between one field based on the rear field f0 and the target field f1.

The previous field f2 and the attention field f1 are input to the field difference detection circuit 52. The field difference detection circuit 52 detects the difference between the previous field f2 and the field of interest f1 and outputs a field difference detection signal d2. The field difference detection signal d2 is
It is the absolute value of the difference between the previous field f2 and the field of interest f1, and d2 = | f1-f2 | (3). The field difference detection circuit 52 is an inter-field motion detection unit that detects a motion between one field based on the previous field f2 and the attention field f1. In addition,
Instead of providing the field difference detection circuit 52, it is also possible to delay the field difference detection signal d0 by one field and use it as the field difference detection signal d2.

Two field difference detection signals d0 and d2
Is input to the subtraction circuits 53 and 54. The subtraction circuit 53 calculates d0-d2, and the subtraction circuit 54 calculates d2-d0. The outputs of the subtraction circuits 53 and 54 are the limiter 5 respectively.
5,56 are input. The limiters 55 and 56 are for limiting to 0 when the input becomes negative (underflow). That is, when the outputs of the subtraction circuits 53 and 54 become negative, the outputs of the limiters 55 and 56 become 0.

The frame difference detection signal D1 output from the frame difference detection circuit 51 and the output signal D from the limiter 55.
0 and the output signal D2 of the limiter 56 are input to the selection circuit 57. The above-described pattern detection signal P is input to the selection circuit 57, and the selection circuit 57 receives the pattern detection signal P.
Depending on the input signal, the final motion detection signal K is obtained. The selection circuit 57 uses the pattern detection signal P
Represents the above (state 1), D0 is selected to represent the (state 2), and D1 is selected to represent the (state 3). The selection circuit 57 serves as a motion detection signal output unit.

FIG. 2 shows only the basic structure. For the purpose of avoiding erroneous detection at the edge portion of the image, for example, the frame difference detection circuit 51 and the field difference detection circuit 5
A filter may be provided in the subsequent stage of 0 and 52 to perform various filter processes on the outputs of the frame difference detection circuit 51 and the field difference detection circuits 50 and 52, or sensitivity (gain) control may be performed in the subsequent stage. Further, in the case of inter-field difference detection, vertical filtering is applied to the input signals of the field difference detection circuits 50 and 52 for the purpose of matching the scanning line phases and avoiding erroneous detection due to vertical high frequency components. There is also.

Next, the reason why the device is constructed as shown in FIG. 2 and operates as described above will be explained. First, (state 3)
In this case, since the pull-down rule pattern is not detected, by selecting the frame difference detection signal D1 which is the output of the frame difference detection circuit 51, the frame generally used for the motion detection of the progressive scan conversion circuit is selected. Difference detection is used.

In the case of (state 1), since the combination assigned by pull-down is the rear field, the difference d0 between the rear field f0 and the target field f1 should not appear originally. However, as described as a conventional problem, when a caption or the like is added later by editing, it may be locally deviated from the rule pattern. In such a case, the difference d0 between the rear field f0 and the target field f1 is locally output.
Since the subtitles do not move continuously, the difference d2 between the previous field f2 and the attention field f1 does not appear in the subtitle portion.

The present invention makes use of this property, and the subtraction circuit 53 and the limiter 55 cause the field difference detection signal d.
0 and the field difference detection signal d2 are compared with each other, and the motion can be determined only in a portion where the field difference detection signal d0 is larger than the field difference detection signal d2, that is, a specific portion which is out of the pull-down rule pattern. Although it is described as a local or peculiar portion, the rule pattern may be disturbed on the entire screen due to splicing editing or the like. In this case, naturally, the entire screen is determined to be in motion.

On the other hand, in the case of (state 2), (state 1)
By the operation opposite to that described above, the subtraction circuit 54 and the limiter 56 compare the field difference detection signal d0 and the field difference detection signal d2, and make a motion judgment only in the portion where the field difference detection signal d2 is larger than the field difference detection signal d0. You can

Further, the operation in the case where the caption is superimposed on the pulled-down video and partially deviates from the regular pattern will be described with reference to the waveform chart of FIG. FIG. 4 shows a state in which an image having a frame rate of 24 frames / second (24 frame image) is converted into an interlaced video signal by 2-3 pulldown. In addition, in FIG. 4, between the part allocated to 2 fields and the part allocated to 3 fields, it is easy to understand the part where the original frame is allocated to 2 fields and the part allocated to 3 fields. Is attached to the broken line.

In FIG. 4, t is the time direction, and 24
The frame image changes from (A) to (H). As shown in FIGS. 4A to 4D, captions are superimposed on a part of the 24-frame image by editing. Subtitles are shown in Figure 4.
It disappears in the field of (E). Previous field f
2 and the attention field f1 and the rear field f0 are assumed to be in the states of FIGS. 4D to 4F, respectively. In the example of FIG. 4, since the combination of pull-downs is the previous field f2 and the attention field f1, the above-mentioned (state 2)
Equivalent to. At this time, the selection circuit 6 in FIG. 1 selects the previous field f2, and the selection circuit 57 of the motion detection circuit 5 is selected.
Selects the output signal D2 of the limiter 56.

FIG. 4 (I) shows the field difference detection signal d2 (= | f1-f2 |), and FIG. 4 (J) shows the field difference detection signal d0 (= | f1-f0 |).
Is shown. As can be seen by comparing the field difference detection signal d0 and the field difference detection signal d2, the field difference detection signal d0 is larger than the field difference detection signal d2 in the 24-frame image other than the subtitle portion. However, in the subtitle portion, the field difference detection signal d2 is larger than the field difference detection signal d0. At this time, since the output of the subtraction circuit 54 is not limited by the limiter 56, the output signal D2 is output and only the subtitle portion is motion-determined.

Next, another configuration example of the motion detection circuit 5 will be described with reference to FIG. 3, the same parts as those in FIG. 2 are designated by the same reference numerals, and the description thereof will be omitted as appropriate. 3 is different from FIG. 2 in that a minimum value calculation circuit 58 for calculating the minimum value of the output of the selection circuit 57 and the frame difference detection signal D1 is provided. The minimum value calculation circuit 58 is
It is calculated which of the selection result of the selection circuit 57 and the frame difference detection signal D1 is smaller, and the smaller one is output as the final motion detection signal K.

In the example of FIG. 3, first, the selection circuit 57 selects one from the frame difference detection signal D1, the output signal D0 of the limiter 55 and the output signal D2 of the limiter 56, and the minimum value calculation circuit. Although the smaller one of the selection result of the selection circuit 57 and the frame difference detection signal D1 is selected by 58, it may be configured as follows. The output signal D0 of the limiter 55 and the limiter 56
It is also possible to take the minimum value of the frame difference detection signal D1 for each of the output signals D2 of 1, and then select one of the signals according to (state 1) to (state 3).

That is, in the (state 1), the field difference detection signal when the field difference detection signal d0 output from the field difference detection circuit 50 is larger than the field difference detection signal d2 output from the field difference detection circuit 52. The smaller of the difference between d0 and the field difference detection signal d2 and the frame difference detection signal D1 output from the frame difference detection circuit 51 may be output as the motion detection signal.

In the (state 2), the field difference detection signal d2 output from the field difference detection circuit 52 is larger than the field difference detection signal d0 output from the field difference detection circuit 50. The smaller of the difference between d2 and the field difference detection signal d0 and the frame difference detection signal D1 output from the frame difference detection circuit 51 may be output as the motion detection signal. Further, in the (state 3), the frame difference detection signal D output from the frame difference detection circuit 51 is output.
1 may be output as the motion detection signal. In the configuration of FIG. 3, the selection circuit 57 and the minimum value calculation circuit 58 serve as a motion detection signal output unit.

The purpose of constructing as shown in FIG. 3 is to avoid erroneous detection because the field difference detection circuits 50 and 52 are likely to make erroneous detection in a portion containing a large amount of vertical high frequency components. In the field difference detection, even in an originally static image, a motion is detected in a portion including many vertical high frequency components. To prevent this, a vertical low-pass filter (vertical LPF) is often used, but it cannot be completely suppressed. One object of the present invention is to detect the subtitle portion, and if the vertical LPF removes too much vertical high-frequency component, the subtitle portion becomes difficult to be detected and the effect of the present invention is diminished. .

On the other hand, since the frame difference detection circuit 51 does not have a problem like the field difference detection in the frame difference detection, one of the output signals D0 and D2 of the limiters 55 and 56 which is the field difference detection is selected. Even in such a case, if the minimum value calculation circuit 58 obtains the minimum value with the frame difference detection signal D1, the wrong motion detection signal K will not be output. In the configuration of FIG. 3, it is possible to make the characteristics of the vertical LPF for removing the vertical high frequency components gentle. Therefore, the cost can be reduced. From the above points, the configuration of FIG. 3 is preferable to the configuration of FIG. 2 as the motion detection circuit 5.

As described above, in the motion detecting apparatus of the present invention, even when the pull-down rule pattern is detected, the motion detection is operated in the portion outside the rule pattern to generate the motion detection signal. Therefore, an appropriate motion detection signal can be output in such a portion. In the progressive scan conversion device of the present invention using the motion detection device of the present invention, even if there is a portion deviating from the regular pattern, the sequential scan conversion can be favorably performed without incorrect interpolation.

Note that moving images are processed in the portion out of the regular pattern, and interpolation signals are obtained from the pixels F10 and F12 in the upper and lower lines in FIG. 6, which causes flicker interference and lowers vertical resolution, but one field (60 frames). Since only one image per second / second) is disturbed, there is no problem at all.

Based on the above basic structure, the following modifications can be made. In FIG. 1, when the selection circuit 6 may select either the rear field f0 or the front field f2, that is, when it is determined that the pull-down conversion is not performed as in (state 3) or the original frame is selected. In the case where is distributed to both the rear field and the front field, the average value of the rear field f0 and the front field f2 may be added as an option of the selection circuit 6.

Although only the basic part of the structure of the motion detection circuit 5 is shown, a technique for improving the accuracy of motion detection may be incorporated. For example, in the case of (state 3), for the normal 60 field / sec video that is not the pull-down converted signal, the frame difference detection signal D1
However, in addition to the frame difference detection signal D1, the field difference detection signal d0,
The motion detection signal may be generated by using d2 in combination.

The means for detecting the pull-down rule pattern is arbitrary, and there is a method of detecting that the frame difference between the rear field f0 and the front field f2 has a 5-field period, or is coded and transmitted from the transmitting side. It may be a method of decoding things. Also, pull-down is 2-
The pull-down is not limited to 3 pull-down, and may be 2-2 pull-down. According to the present invention, even if the regular pattern is deviated, it does not cause a big problem, so that it is not necessary to increase the precision for detecting the regular pattern. Therefore, the cost can be reduced.

[0052]

As described in detail above, in the motion detecting device for interlaced video signals of the present invention, the interlaced video signal is an interlaced video signal converted from a frame image having a predetermined frame rate into an interlaced video by pulling down. A pull-down detecting means for detecting whether or not the pull-down detecting means detects that the interlaced video signal is an interlaced video signal converted into interlaced by pull-down, and the field of interest and a field one field after the field of interest. Or the second field in which the rear field is a common frame image, or the target field and the front field that is one field before the target field are common frame images. In the state The determination means for determining whether or not, the signals of the target field, the rear field, and the front field are input, and the first state, the second state, and the pull-down detection means convert the interlaced video signal into an interlace by pull-down. According to the third state that the interlaced video signal is not detected,
Since it is configured with a motion detection circuit that switches the motion detection method to detect the motion, if the interlaced video signal converted to the interlace by pulldown has a partial or temporary deviation from the regular pattern, an appropriate motion detection signal Can be output.

Further, the progressive scan conversion apparatus of the present invention comprises pull-down detection means for detecting whether or not the interlaced video signal is an interlaced video signal converted from a frame image having a predetermined frame rate to interlace by pulling down. When the pull-down detection means detects that the interlaced video signal is an interlaced video signal converted into interlaced by pull-down, the field of interest and the rear field which is a field one field after the field of interest have a common frame image. Is it in the first state,
A determination unit that determines whether or not the field of interest and the previous field, which is a field one field before the field of interest, are common frame images, and an intrafield interpolation using a signal of the field of interest. A first inter-field interpolation signal is generated using an intra-field interpolating means for generating a signal and a signal of a rear field for the field of interest in the first state, and a front field for the field of interest in the second state. An inter-field interpolating means for generating a second inter-field interpolated signal using the field signal, and signals of the target field, the rear field and the front field are input, and the first state,
The motion detection method is switched according to the second state and the third state in which the pull-down detection means detects that the interlaced video signal is not the interlaced video signal converted into the interlaced by the pull-down, and the motion is detected, and the motion is detected. Since the motion detection circuit that outputs the detection signal and the interpolation signal output means that mixes or switches the output of the intra-field interpolation circuit and the output of the inter-field interpolation circuit according to the motion detection signal are provided, Even if the interlaced video signal converted into the interlaced image has a partial or temporary deviation from the regular pattern, it is possible to perform the progressive scan conversion satisfactorily without erroneous interpolation.

[Brief description of drawings]

FIG. 1 is a block diagram showing an embodiment of the present invention.

FIG. 2 is a block diagram showing a specific configuration example of a motion detection circuit 5 in FIG.

FIG. 3 is a block diagram showing another specific configuration example of the motion detection circuit 5 in FIG.

FIG. 4 is a waveform diagram for explaining the operation of FIG.

FIG. 5 is a diagram for explaining an outline of scanning line interpolation for progressive scan conversion.

FIG. 6 is a diagram for explaining a specific method of scanning line interpolation.

FIG. 7 is a diagram for explaining 2-3 pulldown and progressive scan conversion in that case.

[Explanation of symbols]

3 video interpolation circuit (in-field interpolation means) 4 pattern detection circuit (pull-down detection means, determination means) 5 motion detection circuit 6 selection circuit (inter-field interpolation means) 7 mixing circuit (interpolation signal output means) 9, 57 selection circuit 21 , 22 field memory 50, 52 field difference detection circuit (inter-field motion detection section) 51 frame difference detection circuit (inter-frame motion detection section) 57 selection circuit (motion detection signal output section) 58 minimum value calculation circuit (motion detection signal output) Part) 81,82 double speed processing circuit

Claims (6)

[Claims] 1. A motion detection device for detecting the motion of an interlaced video signal, wherein it is detected whether the interlaced video signal is an interlaced video signal converted from a frame image having a predetermined frame rate into an interlace by pulling down. Pull-down detection means, and when the pull-down detection means detects that the interlaced video signal is an interlaced video signal converted into interlaced by pull-down, a field of interest and a rear field which is a field one field after the field of interest. Is a common frame image, or the second field is a common frame image of the field of interest and the previous field, which is a field one field before the field of interest. Determining means for determining whether the interlaced video signal is input by the first field, the second state, and the pull-down detection means. Is configured to include a motion detection circuit that switches the motion detection method according to a third state in which it is detected that the signal is not an interlaced video signal converted into an interlace by pulling down and detects the motion. Detection device. 2. The motion detection circuit includes a first inter-field motion detection unit that detects a motion based on a difference between the target field and the rear field, and a difference between the target field and the front field. A second inter-field motion detecting section that detects a motion; an inter-frame motion detecting section that detects a motion based on a difference between the previous field and the subsequent field; and, in the first state, between the first fields. If the output of the motion detector is greater than the output of the second inter-field motion detector, the difference between the output of the first inter-field motion detector and the output of the second inter-field motion detector is calculated. If the output of the second inter-field motion detecting section is larger than the output of the first inter-field motion detecting section in the second state, the The difference between the output of the second inter-field motion detection section and the output of the first inter-field motion detection section is output as a motion detection signal, and in the third state, the output of the inter-frame motion detection section. 2. The motion detection device according to claim 1, further comprising a motion detection signal output unit that outputs as a motion detection signal. 3. The motion detection circuit includes a first inter-field motion detector that detects a motion based on a difference between the target field and the rear field, and a difference between the target field and the front field. A second inter-field motion detecting section that detects a motion; an inter-frame motion detecting section that detects a motion based on a difference between the previous field and the subsequent field; and, in the first state, between the first fields. A difference between the output of the first inter-field motion detection section and the output of the second inter-field motion detection section when the output of the motion detection section is greater than the output of the second inter-field motion detection section; The smaller one of the outputs of the inter-frame motion detector is output as a motion detection signal, and in the second state, the output of the second inter-field motion detector is the first The second when the output is larger than the output of the inter-field motion detector
Of the difference between the output of the inter-field motion detection section and the output of the first inter-field motion detection section and the smaller of the output of the inter-frame motion detection section are output as a motion detection signal, and the third The motion detection apparatus according to claim 1, further comprising a motion detection signal output unit that outputs an output of the inter-frame motion detection unit as a motion detection signal in a state. 4. A progressive scan conversion apparatus for converting an interlaced video signal into a progressively scanned video signal, wherein the interlaced video signal is an interlaced video signal converted from a frame image having a predetermined frame rate into an interlace by pulling down. A pull-down detection means for detecting whether or not there is a pull-down detection means, and when the pull-down detection means detects that the interlaced video signal is an interlaced video signal converted to interlace by pull-down, a field of interest and a field one field after the field of interest. In the first state in which the rear field is a common frame image, or the target field and the front field which is a field one field before the target field are common frame images. Determining means for determining whether or not it is in the second state, in-field interpolating means for generating an in-field interpolating signal using the signal of the field of interest, and in the first state, The field signal is used to generate a first inter-field interpolation signal, and in the second state, the field signal is used to generate a second field interpolation signal.
Inter-field interpolating means for generating inter-field interpolating signal, and signals of the target field, the rear field, and the front field are input, and the first state, the second state, and the pull-down detecting means According to the third state in which the interlaced video signal is detected as not the interlaced video signal converted to the interlaced by pulldown, the motion detection method is switched to detect the motion, and the motion detection circuit outputs the motion detection signal. And the interpolating signal output means for mixing or switching the output of the intra-field interpolating circuit and the output of the inter-field interpolating circuit according to the motion detection signal, and outputting the mixed signal. apparatus. 5. The motion detection circuit includes a first inter-field motion detector that detects a motion based on a difference between the target field and the rear field, and a difference between the target field and the front field. A second inter-field motion detecting section that detects a motion; an inter-frame motion detecting section that detects a motion based on a difference between the previous field and the subsequent field; and, in the first state, between the first fields. If the output of the motion detector is greater than the output of the second inter-field motion detector, the difference between the output of the first inter-field motion detector and the output of the second inter-field motion detector is calculated. If the output of the second inter-field motion detecting section is larger than the output of the first inter-field motion detecting section in the second state, the The difference between the output of the second inter-field motion detection section and the output of the first inter-field motion detection section is output as a motion detection signal, and in the third state, the output of the inter-frame motion detection section. 5. The progressive scan conversion apparatus according to claim 4, further comprising a motion detection signal output section that outputs as a motion detection signal. 6. The motion detection circuit includes a first inter-field motion detector that detects a motion based on a difference between the field of interest and the subsequent field, and a difference between the field of interest and the previous field. A second inter-field motion detecting section that detects a motion; an inter-frame motion detecting section that detects a motion based on a difference between the previous field and the subsequent field; and, in the first state, between the first fields. A difference between the output of the first inter-field motion detection section and the output of the second inter-field motion detection section when the output of the motion detection section is greater than the output of the second inter-field motion detection section; The smaller one of the outputs of the inter-frame motion detector is output as a motion detection signal, and in the second state, the output of the second inter-field motion detector is the first The second when the output is larger than the output of the inter-field motion detector
Of the difference between the output of the inter-field motion detection section and the output of the first inter-field motion detection section and the smaller of the output of the inter-frame motion detection section are output as a motion detection signal, and the third 5. The progressive scan conversion apparatus according to claim 4, further comprising a motion detection signal output unit that outputs an output of the inter-frame motion detection unit as a motion detection signal in a state.