JP3181431B2 - Adaptive motion interpolation signal generator using motion vector - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention generates a true interpolation signal by adaptively switching between an interpolation method using a motion vector and an interpolation method not using a motion vector based on the image quality of both. The present invention relates to an adaptive motion interpolation signal generation device using a changing motion vector.

[0002]

2. Description of the Related Art In recent years, in the field of television broadcasting, research and development on conversion technology of a television standard system have been vigorously conducted in accordance with an increase in demand for exchange of programs between different systems.

[0003] One of the conversion items of this television standard system
One is the conversion of the number of fields. This conversion is achieved in principle by thinning out one field or repeating one field every predetermined number of fields.

However, this method has a problem that the motion of a moving image becomes discontinuous at every thinning cycle or repetition cycle. For this reason, in a device requiring high image quality, a field interpolation method is often used.

At present, the field interpolation method includes a field interpolation method using two-field linear interpolation (hereinafter, referred to as “linear interpolation method”) and a field interpolation method using four fields (hereinafter, referred to as “linear interpolation method”). And a field interpolation method using a motion vector.

[0006] The linear interpolation method is based on a field interpolation ratio between a television signal of a previous field (hereinafter referred to as "previous field signal") and a television signal of the current field (hereinafter referred to as "current field signal"). By interpolation, a television signal of an interpolation field (hereinafter, referred to as “field interpolation signal”) is generated.

In the four-field interpolation method, a television signal is regarded as a signal sampled at a field frequency, and the conversion of the number of fields is processed as the conversion of the sampling frequency to obtain a field interpolation signal. .

The field interpolation method using a motion vector detects the direction and magnitude of motion of a moving image, that is, detects a motion vector, and performs motion correction (correcting the position of the moving image by using the motion vector). )
This is to obtain a field interpolation signal.

Although the linear interpolation method has a simple circuit configuration, it has a problem that a so-called jerkiness occurs in which a discontinuous portion occurs in a moving image. The four-field interpolation method can reduce jerkiness compared to the linear interpolation method, but has a problem that the resolution of a moving image is reduced.

On the other hand, the field interpolation method using a motion vector captures a moving image itself and corrects the position of the moving image, thereby eliminating jerkiness and lowering the resolution. Therefore, when the field interpolation method is used as the field number conversion method, the field interpolation method using a motion vector is effective.

However, in the case of the field interpolation method using the motion vector, there is a problem that the detection accuracy and the motion correction accuracy of the motion vector have a great influence on the image quality. Therefore, when the field interpolation method using the motion vector is used, it is necessary to employ a highly accurate motion vector detection technique and a motion correction technique.

However, at present, a highly accurate motion vector detecting technique and a motion correcting technique have not been established. Therefore, when a field interpolation method using a motion vector is used, this is combined with, for example, a linear interpolation method, and these are adaptively switched based on the image quality of both to obtain a true interpolation signal. Is obtained. Hereinafter, such a system is referred to as an adaptive switching interpolation system.

FIG. 2 is a block diagram showing a conventional configuration of a field interpolation signal generating apparatus of such an image quality adaptive switching system.

In the figure, a motion vector detecting circuit 13
, The motion vector correction circuit 14, and the motion correction memory 1
5, 16, multiplication circuits 17, 18 and addition circuit 19
A field interpolation signal generation circuit of a field interpolation system using a motion vector is configured.

This circuit performs a field interpolation process using a motion vector from a current field signal PS supplied to an input terminal 11 and a previous field signal FS supplied to an input terminal 12 to generate a field interpolation signal (hereinafter, referred to as a field interpolation signal). S1 (referred to as "motion compensation field interpolation signal").

The multiplication circuits 20 and 21 and the addition circuit 22
A field interpolation signal generation circuit of a linear interpolation system is configured. This circuit performs a field interpolation signal (hereinafter, referred to as “linear field”) from the current field signal PS supplied to the input terminal 11 and the previous field signal FS supplied to the input terminal 12 by an interpolation process using two-field linear interpolation. S2) (referred to as "interpolated signal").

The multiplication circuits 23 and 24 and the addition circuit 2
5, subtraction circuits 27 and 29, and absolute value conversion / accumulation circuit 2
8, 30 and the adaptive motion interpolation switching control circuit 31 constitute an adaptive switching circuit.

This circuit includes an image based on the motion-compensated field interpolation signal S1 (hereinafter referred to as "motion-compensated field interpolation image") and an image based on the linear field interpolation signal S2 (hereinafter "linear field interpolation image"). ) Based on the image quality, a true field interpolation signal S3 is output to the output terminal 26.

As a parameter indicating the image quality of the motion-compensated field interpolated image, a difference value (hereinafter, referred to as a “motion-compensation inter-field difference value”) D1 indicating a level difference between the motion-compensated field signals PS and FS. Is used.
The parameters indicating the image quality of the linear field interpolated image include field signals PS,
A difference value (hereinafter, “motion 0 field difference value”) D2 indicating a level difference between FSs is used.

According to such a configuration, the two are adaptively switched in accordance with the image quality of the motion-compensated field interpolation image and the linear field interpolation image. Thus, it is possible to suppress a decrease in the image quality of the true field interpolation image.

[0021]

However, in the above-described configuration, when the magnitude of the motion vector V changes abruptly, there is a problem that a true field interpolation image is distorted. Hereinafter, this problem will be described with reference to FIG.

Now, as shown in FIG. 3A, consider a case where the camera is panning by capturing a moving object (a circle in this figure). In this case, in the television image, the moving object image is a still image, and the background image is a moving image.

In such a television image, when the background is flat (when there is no change in the level of the image signal) or when the movement is small, distortion is hardly generated in the true field interpolation image.

However, when the background is not flat or when the motion is large, the magnitude of the motion vector V changes abruptly behind the moving object, and the detection accuracy also decreases. For this reason, distortion occurs in the motion-compensated field interpolation image, and when this is selected, distortion occurs in the true field interpolation image. Therefore, in this case, the image distortion is reduced when the linear field interpolation image is selected as the true field interpolation image.

However, in this case, since the background which has been shadowed by the stationary object appears on the screen of the current field, the difference value D2 between the motion 0 fields becomes large. This makes it easy to select a motion compensation field interpolation image, and as shown in FIG. 3B, image distortion occurs behind the stationary object.

The present invention has been made in order to cope with the above situation, and has an adaptive motion interpolation signal generation apparatus using a motion vector capable of suppressing image distortion caused by a sudden change in the magnitude of the motion vector. The purpose is to provide.

[0027]

In order to achieve the above object, the present invention provides (1) adaptive motion estimation using a motion vector.
The interpolation signal generator uses interpolation processing using motion vectors.
A first interpolated signal generation for generating a first interpolated signal S1
Means, and (2) interpolation using no motion vector,
Second interpolation signal generation means for generating a second interpolation signal S2
And (3) a first evaluation giving an image quality of the first interpolation signal.
Values (eg, inter-field difference values) and the second interpolation
A second evaluation value (eg, between fields)
After calculating the difference ΔD from the difference value), it corresponds to the difference ΔD.
The interpolation coefficient β (0 ≦ β ≦ 1) to a characteristic curve (for example, C
Read out by applying 1) to the read out interpolation coefficient β
The first interpolation signal S1 and the second interpolation signal
Interpolation processing with S2 is performed to obtain a third interpolation signal S3,
Third interpolation signal generation for outputting this as a true interpolation signal
Means and (4) the magnitude | V |
Where the upper limit of the torque detectable area is set based on | V | max
When the threshold value THV is exceeded, the motion vector and the predetermined
As the difference ΔV from the threshold value of the third
Change the characteristic curve of the forming means to one with a small increase rate
(For example, switching from C1 to C2),
The interpolation ratio of the second interpolation signal to the interpolation signal S3 (1−
Characteristic curve switching system that controls so that β) is easily increased
Control means.

[0028]

According to the above arrangement, when the magnitude of the motion vector exceeds a predetermined threshold value, the second interpolation signal is more easily selected than when the magnitude does not exceed the predetermined threshold value. Moreover, the tendency increases as the difference between the magnitude of the motion vector and its threshold increases. Therefore, even if the magnitude of the motion vector changes abruptly, it is possible to suppress the occurrence of image distortion due to this.

[0029]

Embodiments of the present invention will be described below in detail with reference to the drawings. FIG. 1 is a block diagram showing the configuration of one embodiment of the present invention.

The apparatus shown in FIG. 1 uses an inter-field difference value D1,
2 in that a true field interpolation signal S3 is obtained by adaptively switching between the motion compensation field interpolation signal S1 and the linear field interpolation signal S2 based on D2.
It is the same as the device of.

However, in the device of FIG. 2, the adaptive switching operation is fixed, whereas in the device of FIG. 1, the adaptive switching operation is controlled based on the magnitude of the motion vector V. Are different in that

Hereinafter, the configuration of the apparatus shown in FIG. 1 having such features will be described in detail. First, the configuration of a field interpolation signal generation circuit of a field interpolation system using a motion vector will be described.

As described above, this circuit comprises the motion vector detecting circuit 13, the motion vector correcting circuit 14, the motion correcting memories 15 and 16, the multiplying circuits 17 and 18, and the adding circuit 19.

Here, the motion vector detection circuit 13 detects the motion vector V of the moving image based on the current field signal PS supplied to the input terminal 11 and the previous field signal FS supplied to the input terminal 12. Having.

Various methods can be considered as a method for detecting such a motion vector V. For example, for example, a television image is subdivided into blocks of m pixels × n lines (m and n are integers). There is a method of detecting each block.

This method is described in, for example,
No. 162683, a pattern matching method described in JP-A-55-162684 and JP-A-60-158786.
Is well known.

The motion vector correction circuit 14 calculates the motion vector V detected by the motion vector detection circuit 13
By multiplying by the field interpolation ratio α, (1−α) (0 ≦ α ≦ 1), the motion vector V is corrected based on the field interpolation ratio α, (1−α).

The motion compensation memory 15 converts the coordinates of the current field signal PS supplied to the input terminal 11 into the motion vector (1-α) V output from the motion vector correction circuit 14.
It has the function to shift only. Similarly, the motion correction memory 16 stores the previous field signal F supplied to the input terminal 12.
It has a function to shift the coordinates of S by the motion vector αV output from the motion vector correction circuit 14.

The multiplication circuits 17 and 18 and the addition circuit 19
The current field signal PS shifted by (1−α) V and α
It has a function of generating a motion-compensated field interpolation signal S1 by weight-adding the previous field signal FS shifted by V according to the field interpolation ratio α, (1−α).

The above is the configuration of the field interpolation signal generation circuit of the field interpolation system using the motion vector.

In this circuit, the motion compensation field interpolation is performed by weight addition of the signal obtained by shifting the current field signal PS by (1−α) V and the signal obtained by shifting the previous field signal FS by αV. The reason why the signal S1 is generated is to reduce the fluctuation of the image caused by the fact that the motion vector V cannot be corrected by the number of pixels or less, and the fluctuation of the image caused when the motion vector cannot be accurately detected.

Next, the configuration of a linear interpolation type field interpolation signal generation circuit will be described.

As described above, this circuit includes the multiplication circuit 20,
21 and an adder circuit 22. The current field signal PS supplied to the input terminal 11 and the previous field signal FS supplied to the input terminal 12 are represented by a field interpolation ratio α,
The linear field interpolation signal S2 is generated by adding weights based on (1-α).

Next, the configuration of the adaptive switching circuit will be described.

This circuit, as described above, has a multiplication circuit 23,
24, an addition circuit 25, subtraction circuits 27 and 29, absolute value conversion / accumulation circuits 28 and 30, and an adaptive motion interpolation switching control circuit 31.

Here, the multiplication circuits 23 and 24 and the addition circuit 2
5, an adaptive interpolation switching coefficient β, (1-β) (0 ≦ β) output from the adaptive motion interpolation switching control circuit 31 by converting the motion compensation field signal S1 and the linear field interpolation signal S2.
≤ 1), and has a function of outputting a true field interpolation signal S3 by adding weights.

Subtraction circuit 27 and absolute value conversion / accumulation circuit 28
Has a function of detecting a difference value D1 between motion correction fields. Similarly, the subtraction circuit 29 and the absolute value conversion / accumulation circuit 30 have a function of detecting a difference value D2 between motion 0 fields.

In this case, the inter-field difference values D1, D2
Is detected in a predetermined interpolation block unit. The size of the interpolation block is set to, for example, 4 pixels × 2 lines. The method of obtaining the inter-field difference values D1 and D2 not in a pixel unit but in a predetermined block unit is a method of obtaining the pixel difference unit in a pixel unit. Because it becomes.

By obtaining the inter-field difference values D1 and D2 in a predetermined block unit, a signal can be read from the motion compensation memories 15 and 16 by one pixel in the block unit. .

The adaptive motion interpolation switching control circuit 31 is based on the inter-field difference values D1 and D2 output from the absolute value conversion / cumulative addition circuits 28 and 30, and the adaptive motion interpolation coefficients β and (1- β).

In this case, the adaptive motion interpolation switching control circuit 3
1 is the difference ΔD between the inter-field difference values D1 and D2 (= D2
−D1), and the adaptive motion interpolation coefficient β is controlled based on the difference ΔD. FIG. 4 shows the control characteristics. In the figure, the characteristic curve indicated by the solid line C1 is the adaptive motion interpolation coefficient β
3 shows the control characteristics.

As shown, the adaptive motion interpolation coefficient β is the difference Δ
It is controlled to increase as D increases. Accordingly, when the difference ΔD increases, the selection amount of the motion compensation field interpolation signal S1 increases, and the selection amount of the linear field interpolation signal S2 decreases. Conversely, when the difference ΔD becomes smaller, the selection amount of the motion compensation field interpolation signal S1 decreases, and the selection amount of the linear field interpolation signal S2 decreases.

The adaptive motion interpolation coefficient β is set to 1 when the difference ΔD exceeds the upper threshold value THH. Thereby, in this case, as shown in FIG. 5, only the motion compensation field interpolation signal S1 is selected. Conversely, the lower threshold T
When it becomes smaller than HL, it is set to 0. This allows
In this case, as shown in FIG. 5, only the linear field interpolation signal S2 is selected.

Next, the configuration of a switching operation control circuit for controlling the switching operation of the adaptive switching circuit will be described.

The switching operation control circuit includes a threshold generation circuit 41, a determination circuit 42, a difference detection circuit 43, and a threshold control circuit 44.

Here, the threshold value generation circuit 41 has a function of generating a predetermined threshold value THV. This threshold value THV may be fixed or variable, but if it is variable, versatility can be given.

The judgment circuit 42 is a motion vector detection circuit 1
3 is compared with the threshold value THV output from the threshold value generation circuit 41, and | V |
Has a function of determining whether or not | exceeds THV. As a comparison method in this case, the two may be directly compared, or the magnitude | V | of the motion vector V is decomposed into a horizontal component | V | x and a vertical component | V | y, You may make it compare with threshold value THVx and THVy of a direction.

The difference detection circuit 43 calculates a difference ΔV (= V) between the threshold value THV output from the threshold value generation circuit 41 and the magnitude | V | of the motion vector V detected by the motion vector detection circuit 13.
| V | -THV).

When the determination circuit 42 determines that the magnitude | V | of the motion vector V exceeds the threshold value THX, the threshold control circuit 43 performs adaptive motion interpolation switching according to the difference ΔV output from the difference detection circuit 43. It has a function of controlling the upper threshold value THH of the control circuit 31.

FIG. 6 is a characteristic diagram showing this control characteristic. As shown in the figure, the threshold value THH is controlled so as to gradually increase as the difference ΔV increases. As a result, the increase rate of the adaptive motion interpolation coefficient β gradually decreases as the difference ΔV increases, as shown by the broken line in FIG.

Thus, the magnitude | V of the motion vector V
Exceeds | threshold THV, the amount of selection of linear field interpolation signal S2 at a certain difference [Delta] V is greater than when | V | does not exceed THV. In other words, | V | is TH
When V is exceeded, the linear field interpolation signal S2 is more easily selected than when V is not exceeded. This tendency increases as the difference ΔV increases.

Note that the threshold value THV is set to be smaller than the magnitude | V | of the motion vector V at the upper limit of the motion vector detectable area, where | V | max. That is, it is set so that THV <| V | max. Here, the motion vector detectable area is an area where the motion vector detection circuit 13 can detect the motion vector V with a certain accuracy.

The reason why THV <| V | max is satisfied is as follows. That is, the magnitude of the motion vector V |
When V | does not exceed | V | max, the image distortion as described above does not occur. Therefore, THV <| V | ma
The setting of x means that the control is performed even in a portion where the control of the threshold value THH does not need to be performed.

On the other hand, | V
It is also possible to detect a motion vector V having a magnitude | V | Therefore, it is preferable that THV = | V | max, and that the control of the threshold value THH is executed only in a portion that requires the control.

However, near | V | max, the detection accuracy of the motion vector V decreases. Therefore, for example, as shown in FIG.
Despite the occurrence of the motion vector V having | 1, the magnitude | V | of the detected motion vector V is | V |
| V | 2 that is smaller than max.

In such a case, although the threshold value THH needs to be controlled, the control is not performed. As a result, image distortion occurs.

Therefore, in this embodiment, THV <| V |
max. That is, the control of the threshold value THH is started in a region where the detection accuracy of the motion vector V is stable.

According to such a configuration, the motion vector V
Even if the original magnitude | V | 1 of is incorrectly detected as | V | 2, the threshold THH can be controlled as long as | V | 2 is not smaller than the threshold THV.

The operation of the above configuration will be described. First, the operation of generating the true field interpolation signal S3 will be described.

The field signals PS and FS supplied to the input terminals 11 and 12 are supplied to a motion vector detection circuit 13. Thereby, the motion vector V of the moving image is detected. This motion vector V is calculated by the motion vector correction circuit 1
4 is supplied. As a result, motion vectors αV and (1−α) V corrected based on the field interpolation ratio α are obtained.

The motion vectors (1-α) V and αV are supplied to the motion correction memories 15 and 16. Thus, the field signals PS, F supplied to the input terminals 11, 12
The coordinates of S are shifted by (1−α) V and αV, respectively.

The field signal P subjected to the motion correction
S and FS are multiplied by the field interpolation ratios (1−α) and α by the multiplication circuits 17 and 18 respectively,
Is added. As a result, a motion-compensated field interpolation signal S1 obtained by weight-adding the motion-compensated field signals FS and PS at the field interpolation ratio α is obtained.

Similarly, the field signals PS and FS supplied to the input terminals 11 and 12 are supplied to
After being multiplied by the field interpolation ratio (1−α), α, the sum is added by the adding circuit 19. As a result, a linear field interpolation signal S2 obtained by weight-adding the field signals FS, PS to which no motion correction is performed at the field interpolation ratio α is obtained.

The field interpolation signals S1 and S2 obtained in this manner are multiplied by adaptive motion interpolation coefficients β and (1−β) in multiplication circuits 23 and 24, respectively, and then added in an addition circuit 25. Is done. Thus, a true field interpolation signal S3 is obtained.

Next, a threshold value control operation which is a feature of the present invention will be described.

If the magnitude | V | of the motion vector V detected by the motion vector detection circuit 13 is smaller than the threshold value THV, the threshold value control circuit 44 does not control the threshold value THH. Thereby, in this case, as shown in FIG.
The threshold value THH is set to THH0. As a result, the adaptive motion interpolation coefficient β is controlled according to the characteristic curve C1 indicated by the solid line in FIG.

In this case, the inter-field difference values D1, D2
Is ΔD1, for example, the adaptive motion interpolation coefficient β is β1. Thus, in this case, the field interpolation signals S1 and S2 are weighted and added based on the adaptive motion interpolation coefficient β1.

On the other hand, the magnitude | V of the motion vector V
Exceeds the threshold THV, the threshold control circuit 44 starts controlling the threshold THH. Thus, the threshold T
HH gradually increases as the difference ΔV detected by the difference detection circuit 42 increases, as shown in FIG. as a result,
The slope of the control characteristic of the adaptive motion interpolation coefficient β gradually decreases, and the linear field interpolation signal S2 is easily selected. As a result, image distortion due to a sudden change in the magnitude | V | of the motion vector V is less likely to occur.

For example, if the difference ΔV is ΔV1 (> 0), the threshold value THH becomes THH1 larger than THH0 as shown in FIG. Thus, the adaptive motion interpolation coefficient β
Is controlled by a characteristic curve C2 having a smaller slope than the characteristic curve C1, as shown in FIG. As a result, when the difference ΔD is ΔD1, the adaptive motion interpolation coefficient β becomes β2 smaller than β1. Thus, the magnitude | V of the motion vector V
Does not exceed the threshold value THV, (1−β) increases, and the selection amount of the linear field interpolation signal S2 increases.

If the difference ΔV is ΔV2 larger than ΔV1, the threshold value THH becomes THH2 larger than THH1, as shown in FIG. Thereby, the adaptive motion interpolation coefficient β is controlled by the characteristic curve C3 having a smaller slope than the characteristic curve C2, as shown in FIG. As a result, when the difference ΔD is ΔD1, the adaptive motion interpolation coefficient β is β
Β3 which is even smaller than 2. Thereby, the selection amount of the linear field interpolation signal S2 is further increased as compared with the case where the difference ΔV is ΔV1.

According to this embodiment described in detail above, the following effects can be obtained.

(1) First, the magnitude | V of the motion vector V
When | exceeds the threshold value THV, the upper threshold value THH for controlling the adaptive motion interpolation coefficient β is increased.
The linear field interpolation signal S2 can be more easily selected than when | V | does not exceed THV. As a result, it is possible to suppress the occurrence of image distortion due to a sudden change in | V |.

(2) The magnitude | V of the motion vector V
As the difference ΔV between | and the threshold value THV increases, the selection amount of the linear field interpolation signal S2 is increased. Therefore, the selection amount of the signal S2 can be controlled according to the degree of image distortion. . Thus, it is possible to prevent the linear field interpolation signal S2 from being selected more than necessary for suppressing image distortion.

(3) Further, since the threshold value THV is set to a value smaller than the magnitude | V | max of the motion vector V in the upper limit of the motion vector detectable area, the threshold value THV is set in a portion where the detection accuracy of the motion vector V is high. THH control can be started. Thus, even when a detection error of the motion vector V occurs, the threshold value THH can be controlled appropriately.

(4) Since the threshold value THH is controlled to control the ease with which the linear field interpolation signal S2 is selected, the adaptive motion interpolation switching circuit 31
Can be used as it is only by making the threshold value THH variable.

FIG. 8 is a block diagram showing the configuration of the second embodiment of the present invention. In FIG. 8, the same parts as those in FIG. 1 are denoted by the same reference numerals, and detailed description will be omitted.

In the above embodiment, the field interpolation signal S
A case has been described in which the present invention is applied to an apparatus that generates a true field interpolation signal S3 by weight-adding them according to the image quality of S1 and S2.

On the other hand, in this embodiment, an apparatus for selectively selecting one of the field interpolation signals S1 and S2 according to the image quality of the field interpolation signals S1 and S2 is provided.
This shows a case where the present invention is applied.

That is, in FIG. 8, reference numeral 51 denotes field interpolation signals S1, S based on the adaptive motion interpolation coefficient β.
2 is a selection circuit for selectively selecting one of the two.
Reference numeral 52 denotes an adaptive motion interpolation coefficient control circuit that calculates a difference ΔD between the inter-field difference values D1 and D2, and generates an adaptive motion interpolation coefficient β based on the difference ΔD.

As shown in FIG. 9, the adaptive motion interpolation control circuit 52 sets the adaptive motion interpolation coefficient β to 0 when the difference ΔD is smaller than a predetermined threshold value TH, and when the difference ΔD is larger than the threshold value TH, Set to 1. Thereby, as shown in FIG. 10, when the difference ΔD is smaller than the threshold value TH, the linear field interpolation signal S2 is selected as the true interpolation signal S3. When the difference ΔD exceeds the threshold value TH, the motion compensation field interpolation signal is selected. S1 is selected.

In such a configuration, the threshold control circuit 4
4 is the magnitude of the motion vector V as shown in FIG.
When V | exceeds the threshold TH, the threshold TH is controlled so that the threshold THV increases with an increase in the difference ΔV between the two.

Thus, the control characteristic of the adaptive motion interpolation coefficient β is shifted as the difference ΔV increases, as shown by the broken line in FIG. As a result, the linear field interpolation signal S2
Is more easily selected than when the magnitude | V | of the motion vector V does not exceed the threshold value THV. Therefore,
In this embodiment, the same effects as in the previous embodiment can be obtained.

Although the two embodiments of the present invention have been described in detail, the present invention is not limited to such embodiments.

(1) For example, in the above embodiment, a case has been described in which the present invention is applied to an apparatus for performing an interpolation process on a predetermined block basis. However, the present invention can also be applied to an apparatus that performs an interpolation process on a pixel-by-pixel basis. However, in this case, it is necessary to consider an interpolation error due to noise as described above.

(2) In the above embodiment, a case was described in which the present invention was applied to an apparatus using the inter-field difference values D1, D2 as parameters indicating the image quality of the field interpolation signals S1, S2. However, the present invention can be applied to an apparatus using other parameters.

(3) In the above embodiment, a case has been described in which the present invention is applied to an apparatus using a linear interpolation method as a field interpolation method using no motion vector.
However, the present invention can be applied to, for example, an apparatus using a four-field interpolation method. Also, the present invention can be applied to an apparatus using an interpolation method other than the interpolation method in the time axis direction such as a linear interpolation method or a four-field interpolation method.

(4) In the above embodiment, the case where the present invention is applied to the field interpolation processing has been described.
However, the present invention can also be applied to other interpolation processing in the time axis direction, for example, inter-frame interpolation processing.

(5) Further, in the above embodiment, the case where the present invention is applied to conversion of the number of fields has been described. However, the present invention is also applicable to other interpolation processing, for example, an interpolation processing for reproducing a field decimated on the transmission side on the reception side in a high-efficiency coding scheme, and an interpolation processing in a high-definition system. can do.

(5) In addition, it goes without saying that the present invention can be variously modified and implemented without departing from the scope of the invention.

[0100]

As described above in detail, according to the present invention, an adaptive motion interpolation signal generating apparatus using a motion vector capable of suppressing image distortion caused by a sudden change in the magnitude of the motion vector. Can be provided.

[Brief description of the drawings]

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

FIG. 2 is a block diagram showing a configuration of a conventional device.

FIG. 3 is a diagram for explaining a conventional problem.

FIG. 4 is a characteristic diagram illustrating control characteristics of an adaptive motion interpolation coefficient β according to the first embodiment.

FIG. 5 is a characteristic diagram illustrating switching characteristics of a field interpolation signal according to the first embodiment.

FIG. 6 is a characteristic diagram illustrating control characteristics of a threshold value THH according to the first embodiment.

FIG. 7 is a diagram for explaining the reason for setting the threshold value THV to a value smaller than | V | max.

FIG. 8 is a block diagram showing a configuration of a second embodiment of the present invention.

FIG. 9 is a characteristic diagram illustrating control characteristics of an adaptive motion interpolation coefficient β according to the second embodiment.

FIG. 10 is a characteristic diagram illustrating switching characteristics of a field interpolation signal according to the second embodiment.

FIG. 11 is a characteristic diagram illustrating control characteristics of a threshold value THH according to the second embodiment.

[Explanation of symbols]

11, 12 ... input terminals, 13 ... motion vector detection circuit,
14: motion vector correction circuit, 15, 16: motion correction memory, 17, 18, 20, 21, 23, 24: multiplication circuit, 19, 22, 25 ... addition circuit, 26: output terminal, 2
7, 29 ... subtraction circuit, 28, 30 ... absolute value conversion / accumulation circuit, 31, 52 ... adaptive motion interpolation switching control circuit, 51 ...
Selection circuit.

Claims (1)

(57) [Claims] 1. An interpolation process using a motion vector,
A first inner interpolation signals generating means for generating a No. 1 inner挿信by interpolation processing not using a motion vector, a second inner interpolation signals generating means for generating a No. 2 of the挿信, the first A first evaluation value giving the image quality of the interpolation signal of 1 and
The difference from the second evaluation value giving the image quality of the second interpolation signal is
After that, the interpolation coefficient corresponding to the difference is applied to the characteristic curve.
Read out, and based on the read out interpolation coefficient,
The interpolation processing between the first interpolation signal and the second interpolation signal is performed.
To determine the third interpolated signal, and use this as the true interpolated signal.
A third interpolation signal generating means for outputting the motion vector,
When the threshold is exceeded, the motion vector
The third difference increases as the difference between the torque and the predetermined threshold increases.
Cut the characteristic curve of the interpolation signal generation means to one with a small increase rate.
In other words, the second interpolation signal occupying the third interpolation signal
Characteristic curve that controls so that the interpolation ratio of
An adaptive motion interpolation signal generation apparatus using a motion vector, comprising: a switching control unit .