JP2002359775A - Telecine conversion method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a telecine conversion method that smoothes a motion of a video image. SOLUTION: This method smoothes a motion of a video image and converts a video image of a 24P movie film into a 60P video signal, and interpolated video images resulting from interpolating frame video images before and after a movie film are used for video images of prescribed frames after the conversion so that a display time integrated amount of each frame video image of the movie film is equal to each other after the telecine conversion.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a telecine conversion method for converting a picture of a 24P movie film into a 60P picture signal.

[0002]

2. Description of the Related Art Movie film of 24 frames per second (24P)
To a progressive video signal (60 frames per second)
As a conventional telecine conversion for converting to P), FIG.
As shown in (a), a 2-3 pull-down scheme is adopted in which one frame is alternately assigned to two frames and three frames.

[0003] In the 2-3 pull-down method, the image is stationary for two frames in a portion where one frame is allocated to two frames, and the video is stationary for three frames in a portion where one frame is allocated to three frames. Becomes Therefore, "jitter" occurs in the motion of the video, and the video has a strange feeling. In particular, in a fast-moving image, the uncomfortable feeling becomes remarkable.

[0004]

SUMMARY OF THE INVENTION An object of the present invention is to provide a telecine conversion method for smoothing the movement of an image.

Another object of the present invention is to provide a telecine conversion method capable of smoothing the motion of an image and avoiding the occurrence of flicker.

[0006]

According to a first telecine conversion method of the present invention, a picture of a 24P cinema film is converted to 60 pictures.
In the telecine conversion method of converting to a P video signal, frame images before and after the movie film are interpolated as converted predetermined frame images so that the integrated amount of display time of each frame image of the movie film becomes equal after the telecine conversion. A telecine conversion method, characterized in that a corrected interpolation image is used.

A second telecine conversion method according to the present invention is a telecine conversion method for converting an image of a 24P movie film into a 60P video signal, wherein a frame image before and after the movie film is interpolated as an image of a predetermined frame after the conversion. A telecine conversion method that uses an interpolated image obtained by dividing a screen into a plurality of blocks, calculates information on motion for each block, and generates an interpolated image by using a block for each block. , An interpolation coefficient for each block is determined on the basis of the information on the motion, and an interpolation image is generated for each block based on the determined interpolation coefficient for each block.

It is preferable that an interpolation coefficient for each pixel is calculated based on the interpolation coefficient determined for each block, and an interpolation image is generated for each pixel based on the calculated interpolation coefficient for each pixel.

A third telecine conversion method according to the present invention is a telecine conversion method for converting an image of a 24P movie film into a 60P video signal, wherein the frames before and after the movie film are converted as predetermined frame images. In a telecine conversion method in which an interpolated image obtained by interpolating an image is used, an interpolation coefficient obtained by summing both interpolation coefficients to be a value larger than 1 is used as an interpolation coefficient to be multiplied with the preceding and succeeding frame images. If the pixel-by-pixel interpolation value exceeds the larger of the two original frame image luminance values used to calculate the interpolation value, the two interpolation-based It is characterized in that correction is made so as to be equal to or less than the larger one of the luminance values of the frame images.

[0010] A fourth telecine conversion method according to the present invention is a telecine conversion method for converting an image of a 24P movie film into a 60P image signal. In a telecine conversion method in which an interpolated image obtained by interpolating an image is used, an interpolation coefficient in which the sum of both interpolation coefficients is 1 as an interpolation coefficient to be multiplied with the preceding and succeeding frame images, and a sum of the two interpolation coefficients is 1 or more. An interpolation coefficient having a large value is prepared, and when the area of the motion area is equal to or smaller than a predetermined value, an interpolation image is generated using an interpolation coefficient such that the sum of both interpolation coefficients becomes 1, If the area of the motion area is larger than a predetermined value, an interpolation image is generated using an interpolation coefficient such that the sum of both interpolation coefficients is larger than 1, and the sum of both interpolation coefficients becomes larger than 1. When an interpolated image is generated using such an interpolation coefficient, the obtained interpolated value in pixel units is the larger of the luminance values of the two original frame images used to calculate the interpolated value. When the brightness value exceeds the brightness value, the interpolation value is corrected so as to be equal to or less than the larger brightness value of the brightness values of the two original frame images.

A fifth telecine conversion method according to the present invention is a telecine conversion method for converting an image of a 24P movie film into a 60P video signal, wherein the frames before and after the movie film are converted as predetermined frame images. In a telecine conversion method in which an interpolated image obtained by interpolating an image is used, an interpolation coefficient in which the sum of both interpolation coefficients is 1 as an interpolation coefficient to be multiplied to the preceding and succeeding frame images, and a sum of the two interpolation coefficients is 1 or more. An interpolation coefficient having a large value is prepared, and when the amount of change in the luminance level of the motion area is equal to or smaller than a predetermined value, the interpolation video is used by using an interpolation coefficient such that the sum of both interpolation coefficients becomes 1. When the luminance level change amount of the motion area is larger than a predetermined value, the sum of both interpolation coefficients is 1
In the case where an interpolated video is generated using an interpolation coefficient having a larger value and an interpolated video is generated using an interpolation coefficient whose sum of both interpolation coefficients is larger than 1, the obtained pixel unit Is larger than the larger one of the luminance values of the two original frame images used to calculate the interpolated value, the interpolated value based on the interpolated value
It is characterized in that the correction is made so as to be equal to or less than the larger one of the luminance values of one frame image.

[0012]

Embodiments of the present invention will be described below with reference to the drawings.

[A] Description of First Embodiment

[1] Description of Telecine Conversion Method A telecine conversion method for converting a movie film (24P) at 24 frames per second into a progressive video signal (60P) at 60 frames per second using linear interpolation of the entire screen will be described.

FIGS. 1B, 1C and 1D show three concrete methods of a telecine conversion method using linear interpolation of the entire screen.

[1-1] Description of First Method in FIG. 1B In the first method, in the 2-3 pull-down method, for example, a portion where video is switched from video A to video B (before switching) _{Let the} video be Qn and the next video be Qn
_{n + 1} ), the next video Q _{n + 1} is not used as it is, but the video Q _n and Q _{n + 1} before and after the switching.
(PQ _n + qQ _{n + 1} ) is used.

The interpolation coefficients p and q are determined so that the sum thereof is 1 and the integrated value of the display time of each frame image is equal. That is, when the video is switched after two frames of the same video in the 2-3 pull-down scheme, such as switching between video A and video B in the 2-3 pull-down scheme, p = 0.75, q = 0. It will be 25. On the other hand, as in the case of switching between images B and C in the 2-3 pull-down method, the same image is
If the video switches after successive frames, p =
0.25, q = 0.75.

In a movie film, the display time of each frame image is the same. When a movie film is converted to 60P, it is difficult to make the display time of each frame image the same, so that the same value is used as the integrated value of the display time.
The interpolation coefficients p and q have been determined. Thus, the display time of each frame image is a time corresponding to 2.5 frames.

[1-2] Description of Second Method in FIG. 1C In the above-described first method in FIG. 1B, the same image is used in the 2-3 pull-down method like the images B and D. In a portion where three frames are continuous, the first frame is an interpolated video, but the two frames following the first frame are the same video.

When the video contains a fast motion, the video is stationary for two frames in a portion where the same video is continuous for two frames, and the motion of the video is "jittery".
Occurs.

Therefore, in the second method, in the 2-3 pull-down method, in addition to replacing an image switching portion with an interpolated image, one of continuous portions of the same image is replaced with an interpolated image.

As shown in FIG. 1 (c), the interpolation coefficients p and q are determined so that the sum thereof is 1 and the integrated amount of display time of each frame image is the same.

[1-3] Description of Third Method in FIG. 1 (d) In the third method, frame images (A, C, etc.) appear once in five frames, and the frames between them appear as: All are interpolated images (for example, pA + qB). As shown in FIG. 1D, the interpolation coefficients p and q of the respective interpolated images are determined so that the sum thereof is 1 and the integrated amount of display time of each frame image is the same.

The above first, second and third methods are
It is also possible to switch adaptively based on motion information or edge information of an input video (video of a movie film).
For example, the motion of the input image is detected. If the motion is small, the first method is used. If the motion is moderate, the second method is used. If the motion is large, the third method is used.

[2] Description of Telecine Conversion Device FIG. 2 shows a configuration of a telecine conversion device for realizing the above first, second or third method.

24P input image (movie film image)
Are connected to the frame memories 2, 3, 4 via the changeover switch 1.
Sent to one of The video read from each of the frame memories 2, 3, and 4 is sent to the selector 5.

As will be described later, the selector 5 normally has
Video signals read from the three frame memories are input. The selector 5 outputs one of the two input video signals from the output terminal (SEL0) as a first video signal and outputs the other video signal as a second video signal.
Output from (SEL1).

The first video signal output from the output terminal (SEL0) of the selector 5 is sent to the first multiplier 6, and the second video signal output from the output terminal (SEL1) of the selector 5 is output to the first multiplier 6. 2 to the multiplier 7. The first multiplier 6 multiplies the first video signal by an interpolation coefficient p. The second multiplier 7 multiplies the second video signal by an interpolation coefficient q.

The output of the first multiplier 6 and the output of the second multiplier 7 are added by the adder 8 and output as a 60P video signal.

A changeover switch 1, each frame memory 2,
3, 4 and the selector 5 are provided with a vertical synchronizing signal V
D is controlled by a 2-3 pull-down control circuit 11 input thereto. The interpolation coefficients p and q are generated by the interpolation coefficient control circuit 12. The interpolation coefficient control circuit 12 generates interpolation coefficients p and q based on the telecine cycle number sent from the 2-3 pull-down control circuit 11.

FIG. 3 shows signals and the like of each part in FIG.

In FIG. 3, there are five types of telecine cycle numbers "0" to "4" in frame units, and "0" to "4".
Repeat 4 ".

The changeover switch 1 is switched so as to cyclically select three frame memories (memory A, memory B, memory C) 2, 3, and 4 in synchronization with the switching timing of the input image.

Among the videos stored in the frame memories 2, 3, and 4, the latest readable video and its 1
The video that was previously readable is read and sent to the selector 5. The latest readable image is output from the terminal (SEL1) of the selector 5 as a second image. The video that has become readable immediately before is output as the first video from the terminal (SEL0) of the selector 5.

The correction coefficient is determined as follows based on the telecine cycle number.

(1) In the case of the above first method Telecine cycle number "0": p = 0, q = 1 Telecine cycle number "1": p = 0, q = 1 Telecine cycle number "2": p = 0 .25, q = 0.75 Telecine cycle number "3": p = 0, q = 1 Telecine cycle number "4": p = 0.75, q = 0.25

(2) In the case of the above second method Telecine cycle number "0": p = 0.5, q = 0.5 Telecine cycle number "1": p = 0, q = 1 Telecine cycle number "2" : P = 0.75, q = 0.25 Telecine cycle number “3”: p = 0.25, q = 0.75 Telecine cycle number “4”: p = 1, q = 0

(3) In the case of the above third method Telecine cycle number "0": p = 0.5, q = 0.5 Telecine cycle number "1": p = 0.25, q = 0.75 Telecine cycle Number “2”: p = 0.75, q = 0.25 Telecine cycle number “3”: p = 0.5, q = 0.5 Telecine cycle number “4”: p = 1, q = 0

[B] Description of the Second Embodiment

[1] Outline of Telecine Conversion Method

Using linear interpolation for each block in the screen,
A description will be given of a telecine conversion method for converting a movie film (24P) at 24 frames per second into a progressive video signal (60P) at 60 frames per second.

In the second embodiment, the telecine conversion is performed by obtaining an interpolation coefficient for each block in the screen, instead of interpolating the entire screen as in the first embodiment.

That is, the screen is divided into a plurality of blocks,
An interpolation coefficient is calculated for each block based on information on the motion of the video in that block. In this embodiment, the interpolation coefficients p and q for each pixel are further calculated based on the interpolation coefficients obtained for each block. An interpolated image is generated using the calculated interpolation coefficients p and q for each pixel.

By the way, when one screen is divided into a plurality of blocks, an image having no sense of incongruity is obtained in a block having a small motion even when the conventional 2-3 pull-down system is used. However, for a moving block, the conventional 2-
In the 3-pull-down system, "jitter" occurs in the movement. The “rack” of this movement increases as the movement speed increases.

Therefore, in the second embodiment, for a moving block, an interpolation coefficient is calculated based on information on the movement, and an interpolated image is generated based on the calculated interpolation coefficient. In adjacent blocks,
If the interpolation coefficients are different, the continuity of the subject is interrupted at the block boundary. Therefore, the interpolation coefficients p and q for each pixel are obtained based on the interpolation coefficients obtained for each block.

[2] Description of Telecine Converter

FIG. 4 shows the configuration of a telecine conversion device for performing telecine conversion using linear interpolation for each block in the screen.

In FIG. 4, those corresponding to FIG.
The same reference numerals as in FIG. 2 are used.

The difference between the telecine converter of FIG. 4 and the telecine converter of FIG. 2 lies in the interpolation coefficient control circuit 12.
The interpolation coefficient control circuit 12 receives the telecine cycle number from the 2-3 pull-down control circuit 11 and the first video signal and the second video signal output from the selector 5.

[3] Description of interpolation coefficient control circuit 12

FIG. 5 shows the configuration of the interpolation coefficient control circuit 12. The interpolation coefficient control circuit 12 includes a motion information calculation circuit 21, a target interpolation amount determination unit 22, a region interpolation amount determination unit 23,
A telecine cycle adjustment area interpolation coefficient calculation circuit 24 and an interpolation coefficient calculation circuit 25 for each pixel are provided.

The motion information calculation circuit 21 calculates information on motion based on the first video signal and the second video signal output from the selector 5.

The target interpolation amount determining section 22 calculates a target interpolation coefficient based on the information on the motion.

The area interpolation amount determining section 23 calculates a new area interpolation coefficient based on the target interpolation coefficient and the previous area interpolation coefficient.

The area interpolation coefficient calculation circuit 24 with telecine cycle adjustment determines an area interpolation coefficient based on the telecine cycle.

The interpolation coefficient calculating circuit 25 for each pixel calculates an interpolation coefficient for each pixel based on the area interpolation coefficient.

Hereinafter, each of these units will be described.

[3-1] Description of Motion Information Calculation Circuit 21 Here, it is assumed that one screen area is divided into a plurality of blocks. The motion information calculation circuit 21 calculates, for each block, information (hereinafter, motion information) Det relating to motion represented by the following equation (1).

Det = Wa · (edge change amount) + Wb · (brightness change amount) + Wc · (contrast change amount) (1)

Wa, Wb and Wc are weighting factors set in advance. How to determine the edge change amount, the luminance change amount, and the contrast change amount will be described.

FIG. 6 shows a circuit for calculating the edge change amount.

The first video signal is sent to the first Laplacian calculator 31. The first Laplacian calculator 31 calculates
Laplacian is obtained for each pixel using the filter of FIG. The second video signal is sent to the second Laplacian calculator 32. The second Laplacian calculating unit 32 obtains a Laplacian for each pixel using the filter of FIG.

The Laplacian for the same pixel obtained by each of the Laplacian calculating units 31 and 32 is added to the adder 3
3, the difference is calculated. The output of the adder 33 is
It is sent to the block-by-block integration circuit 34. The block-by-block integrating circuit 34 integrates the output of the adder 33 for each block,
The absolute value of the integrated value for each block is output as the edge change amount of the block.

FIG. 8 shows a circuit for calculating the luminance change amount.

The luminance signal of the first video signal is sent to the first block-by-block luminance integrating circuit 41. The first block-by-block luminance integrating circuit 41 calculates a luminance integrated value for each block.

The luminance signal of the second video signal is sent to a second block-by-block luminance integrating circuit 42. The second block-by-block luminance integrating circuit 42 calculates a luminance integrated value for each block.

The integrated luminance value of each block calculated by the first luminance integrating circuit 41 and the integrated luminance value of each block calculated by the second luminance integrating circuit 42 are added to an adder 43. Sent. Adder 43
Calculates the absolute value of the difference between the two luminance integrated values for the same block, and outputs it as the luminance change amount of that block.

FIG. 9 shows a circuit for calculating the amount of change in contrast.

The luminance signal of the first video signal is sent to the first block-by-block contrast calculation circuit 51. The first per-block contrast calculation circuit 51 calculates the difference (contrast) between the maximum luminance value and the minimum luminance value for each block.

The luminance signal of the second video signal is sent to the second block-by-block contrast calculation circuit 52. The second per-block contrast calculation circuit 52 calculates a difference (contrast) between the maximum luminance value and the minimum luminance value for each block.

First block-by-block contrast calculation circuit 5
The contrast for each block calculated by 1 and the contrast for each block calculated by the second block-contrast calculation circuit 52 are sent to the adder 53. The adder 53 calculates the absolute value of the difference between the two contrasts for the same block, and outputs the calculated difference as the contrast change amount of the block.

[3-2] Description of Target Interpolation Amount Determining Unit 22 The target interpolation amount deciding unit 22 calculates the motion information De for each block.
The target interpolation value MK is determined for each block based on a table storing the relationship between t and the target interpolation value MK for the motion information Det as shown in FIG.

[3-3] Description of Area Interpolation Amount Determining Unit 23 FIG.

The area interpolation amount determining section 23 calculates an area interpolation value RK for each block as follows.

First, the first adder 61 calculates a difference (MK−RK) between the target interpolation value MK and the previous area interpolation value RK for the block.

Next, the increase / decrease value KK of the interpolation coefficient is calculated by the KK calculation ROM 62. The KK calculation ROM 62 is
The difference (MK-RK) calculated by the first adder 61
And KK based on the relationship between KK and the difference (target interpolation value MK-previous area interpolation value RK) as shown in FIG.

Next, the second adder 63 adds the previous area interpolation value RK for the block to KK calculated by the KK calculation ROM 62.

The result of the addition (KK + RK) is sent to the comparison circuit 64. The comparison circuit 64 calculates the sum (KK
+ RK) is 0 or more, this addition result (KK + R
K) is output as the current region interpolation value RK. On the other hand, if the addition result (KK + RK) is smaller than 0, 0 is output as the current region interpolation value RK.

[3-4] Description of Telecine Period Adjustment Area Interpolation Coefficient Calculation Circuit 24

The area interpolation coefficient calculating circuit 24 with the telecine cycle adjustment calculates the area interpolation value R for each block for each block.
An interpolation coefficient α for each telecine cycle number is generated and output based on K and the telecine cycle number.

The interpolation coefficient α for the combination of RK and telecine cycle number is determined based on the table shown in FIG. The interpolation coefficient α for the combination of RK and telecine cycle number may be determined based on the table in FIG.

[3-5] Interpolation coefficient calculation circuit 25 for each pixel
The interpolation coefficient calculation circuit 25 for each pixel obtains an interpolation coefficient p for each pixel based on the interpolation coefficient α for each block in the screen.

A method for obtaining the interpolation coefficient p for each pixel will be described with reference to FIG.

FIG. 15 shows four blocks for convenience of explanation.
Ku Z₀₀, Z_Ten, Z₀₁, Z₁₁Only one is shown. Bro
Check Z₀₀The interpolation coefficient calculated for α is
Ku Z _TenThe interpolation coefficient calculated with respect to
Z₀₁The interpolation coefficient calculated with respect to
₁₁Let α11 be the interpolation coefficient calculated for.

[0085] The coordinates of the center of the block Z ₀₀ (x00, y
00), and the coordinates of the center of the block Z ₁₀ (x10, y
10), and the coordinates of the center of the block Z ₀₁ (x01, y
01), and the coordinates of the center of the block Z ₁₁ (x11, y
11).

The interpolation coefficient of the pixel at the coordinates (x, y) is expressed by p
Assuming that (x, y) and q (x, y), p (x, y) is
It is calculated based on the following equation (2). Note that q (x, y)
Is determined by 1-p (x, y).

P (x, y) = (α00 + Δh0 * (x-x00)) + [｛(α01 + Δh1 * (x-x0)-(α00 + Δh0 * (x-x00))｝ / (y01-y00)] * (y-y00) h0 = (α10-α00) / (x10-x00) h1 = (α11-α01) / (x11-x01)… (2)

FIG. 16 shows an example of signals and the like of each section in FIG.

In FIG. 16, the 60P output indicates a case where the area interpolation coefficient calculation circuit 24 with the telecine cycle adjustment determines the interpolation coefficient α based on the table of FIG.

[C] Description of Third Embodiment

According to the first or second embodiment, the motion of the video is smoother than that of the conventional 2-3 pull-down system. However, when the video after the telecine conversion is displayed on the CRT, the video becomes smoother. It has been found that, depending on the type, flicker, that is, a phenomenon in which an image appears to flicker due to a change in luminance level occurs.

The occurrence of flicker is caused by the interpolation coefficients p,
Since the sum of them is 1 as q,
It is considered that the cause is that the luminance level fluctuates between a video obtained by interpolation and a video not subjected to interpolation. This will be described more specifically.

As shown in FIG. 17, it is assumed that a high-luminance moving object (hatched portion) X exists between a frame image A and a frame image B. The frame image A and the second frame image B are
By performing interpolation using two interpolation coefficients p and q such that the sum becomes 1, an interpolated image (pA +
qB) is obtained. In this interpolated image (pA + qB), the luminance is high in a portion (hatched portion) Y where the moving object portion in the frame image A and the moving object portion in the frame image B overlap, but the luminance decreases in the non-overlapping portion Z. I do. That is, in the interpolated video (pA + qB), the area of the high-brightness portion decreases in the high-brightness moving object portion. As a result, the luminance level fluctuates between the interpolated video (pA + qB) and the non-interpolated videos A and B.

[1] Description of Telecine Conversion Method A telecine conversion method for converting a movie film (24P) at 24 frames per second into a progressive video signal (60P) at 60 frames per second using linear interpolation of the entire screen will be described.

FIGS. 18 (b), (c) and (d) show three specific methods of a telecine conversion method using linear interpolation of the entire screen. FIG. 18A shows the conventional 2
3 shows a pull-down method.

[1-1] Description of First Method in FIG. 18B In the first method, in the 2-3 pull-down method, a portion where the video is switched from video A to video B, for example (before the switching) _{Let the} video be Qn and the next video be Qn
_{n + 1} ), the next video Q _{n + 1} is not used as it is, but the video Q _n and Q _{n + 1} before and after the switching.
(PQ _n + qQ _{n + 1} ) is used.

The interpolation coefficients p and q are determined so that the sum (p + q) of them becomes larger than 1 in order to compensate for the decrease in the luminance of the interpolated video and make the flicker less noticeable.

As described above, when the sum of the interpolation coefficients p and q is determined to be a value larger than 1, the obtained pixel-based interpolation value (p
Q _n + qQ _{n + 1} ) may exceed the larger luminance value of the original two frame images Q _n and Q _{n + 1} used to calculate the interpolation value. When this occurs, the luminance of the interpolated video becomes too large, and the reverse flicker may occur.

Therefore, as described later, in this embodiment, the obtained interpolated value (pQ _n + qQ _{n + 1} ) in pixel units is obtained.
Exceeds the larger luminance value of the original two frame images Q _n and Q _{n + 1} used for calculating the interpolation value, the two interpolation-based two Correction is made so as to be equal to or less than the larger luminance value of the frame images Q _n and Q _{n + 1} . Such a correction is similarly performed in the second method and the third method.

In the first method, as in the case of switching between the images A and B in the 2-3 pull-down system, when the same video is switched after two consecutive frames in the 2-3 pull-down system, p = 24 / 32, q = 9/32. On the other hand, when the video is switched after three frames of the same video have been consecutively performed in the 2-3 pull-down system, as in the case of switching between the video images B and C in the 2-3 pull-down system, p = 8/32, q = 25 / It becomes 32.

In this example, the interpolation coefficients p and q are determined so that the integrated value of the display time of each frame image becomes equal. Accordingly, the display time of each frame image is a time corresponding to {2+ (17/32)} frames. In this example, the interpolation coefficients p and q are determined so that the integrated value of the display time of each frame image is equal.
It is not necessary to make the integrated value of the display time of each frame image equal.

[1-2] Description of Second Method in FIG. 18 (c) In the first method in FIG. 18 (b), the same video is In a portion where three frames are continuous, the first frame is an interpolated video, but the two frames following the first frame are the same video.

When the video includes a fast motion, in a portion where the same video continues for two frames, the video is stationary for two frames, and the motion of the video is “jittery”.
Occurs.

Therefore, in the second method, in the 2-3 pull-down method, in addition to replacing the portion where the video is switched with the interpolated video, one of the portions where the same video continues is replaced with the interpolated video.

As shown in FIG. 18 (c), the interpolation coefficients p and q are determined so that the sum thereof is greater than 1 and the display time integration amounts of the respective frame images are the same. ing.

[1-3] Description of Third Method in FIG. 18 (d) In the third method, frame images (A, C, etc.) appear once every five frames, and the frames between them appear as: All are interpolated images (for example, pA + qB). As shown in FIG. 18D, the interpolation coefficients p and q of the respective interpolated images are determined so that the sum thereof is greater than 1 and the display time integration amounts of the respective frame images are the same. Have been.

[2] Description of Telecine Conversion Device FIG. 19 shows the configuration of a telecine conversion device for realizing the first, second or third method.

24P input video (movie film video)
Are connected to the frame memories 2, 3, 4 via the changeover switch 1.
Sent to one of The video read from each of the frame memories 2, 3, and 4 is sent to the selector 5.

As described later, the selector 5 normally has
Video signals read from the three frame memories are input. The selector 5 outputs one of the two input video signals from the output terminal (SEL0) as a first video signal and outputs the other video signal as a second video signal.
Output from (SEL1).

The first video signal output from the output terminal (SEL0) of the selector 5 is sent to the first multiplier 6, and the second video signal output from the output terminal (SEL1) of the selector 5 is output to the first multiplier 6. 2 to the multiplier 7. The first multiplier 6 multiplies the first video signal by an interpolation coefficient p. The second multiplier 7 multiplies the second video signal by an interpolation coefficient q.

The output of the first multiplier 6 and the output of the second multiplier 7 are added by the adder 8. The output of the adder 8 is sent to the minimum value output circuit 10.

The first video signal output from the output terminal (SEL0) of the selector 5 and the second video signal output from the output terminal (SEL1) of the selector 5 are also sent to the maximum value output circuit 9. The maximum value output circuit 9 outputs the one of the first video signal and the second video signal that has higher luminance. The output of the maximum value output circuit 9 is the minimum value output circuit 1
Sent to 0.

The minimum value output circuit 10 outputs the video signal having the smaller luminance value between the video signal sent from the adder 8 and the video signal sent from the maximum value output circuit 9. The output signal of the minimum value output circuit 10 is output as a 60P video signal.

Maximum value output circuit 9 and minimum value output circuit 1
0 indicates that the pixel-based interpolation value obtained by using the interpolation coefficients p and q exceeds the larger one of the luminance values of the original two frame images used to calculate the interpolation value. In this case, the interpolation value is provided to correct the interpolation value to be equal to or less than the larger one of the luminance values of the two original frame images.

The changeover switch 1, each frame memory 2,
3, 4 and the selector 5 are provided with a vertical synchronizing signal V
D is controlled by a 2-3 pull-down control circuit 11 input thereto. The interpolation coefficients p and q are generated by the interpolation coefficient control circuit 12. The interpolation coefficient control circuit 12 generates interpolation coefficients p and q based on the telecine cycle number sent from the 2-3 pull-down control circuit 11.

FIG. 20 shows the signals and the like of each part in FIG.

In FIG. 20, there are five types of telecine cycle numbers “0” to “4” in frame units, and “0” to “4”.
Repeat 4 ".

The changeover switch 1 is switched so as to cyclically select three frame memories (memory A, memory B, memory C) 2, 3, and 4 in synchronization with the switching timing of the input video.

Among the videos stored in the frame memories 2, 3, and 4, the latest readable video and its 1
The video that was previously readable is read and sent to the selector 5. The latest readable image is output from the terminal (SEL1) of the selector 5 as a second image. The video that has become readable immediately before is output as the first video from the terminal (SEL0) of the selector 5.

The correction coefficient is determined as follows based on the telecine cycle number.

(1) In the case of the above first method Telecine cycle number "0": p = 0, q = 1 Telecine cycle number "1": p = 0, q = 1 Telecine cycle number "2": p = 8 / 32, q = 25/3
2 Telecine cycle number “3”: p = 0, q = 1 Telecine cycle number “4”: p = 24/32, q = 9/3
2

(2) In the case of the above second method Telecine cycle number "0": p = 33/64, q = 33 /
64 Telecine cycle number "1": p = 0, q = 1 Telecine cycle number "2": p = 25/32, q = 8/3
2 Telecine cycle number “3”: p = 8/32, q = 25/3
2 Telecine cycle number “4”: p = 1, q = 0

(3) In the case of the above third method Telecine cycle number "0": p = 33/64, q = 33 /
64 telecine cycle number "1": p = 8/32, q = 25/3
2 Telecine cycle number “2”: p = 24/32, q = 9/3
2 Telecine cycle number “3”: p = 33/64, q = 33 /
64 Telecine cycle number "4": p = 1, q = 0

[D] Description of Fourth Embodiment

In the fourth embodiment, when the area of the motion region is small, an interpolated image is generated using the coefficients p and q whose sum is 1, and when the area of the motion region is large, the sum is calculated. The coefficient is adaptively controlled according to the area of the motion region, such as generating an interpolated video using the coefficients p and q having a value larger than 1.

FIG. 21 shows the configuration of the telecine converter.

In FIG. 21, parts corresponding to those in FIG. 19 are denoted by the same reference numerals as in FIG.

The difference between the telecine conversion device shown in FIG. 21 and the telecine conversion device shown in FIG. 19 is that the motion information calculation circuit 13 is provided and the interpolation coefficient control circuit 12 is calculated by the motion information calculation circuit 13. Depending on the area of the moving region, the sum of the coefficients p and q is set to 1 and the sum is set to 1
The point is to switch between those having a larger value.

The interpolation coefficient control circuit 12 is, for example, shown in FIG.
Two types of coefficient tables as shown in FIG. The first coefficient table shown in FIG. 22A stores the values of coefficients p and q corresponding to the telecine cycle number when the area of the motion area is small. The second coefficient table shown in FIG. 22B stores coefficients p and q according to the telecine cycle number when the area of the motion area is large.

The values of the coefficients p and q corresponding to the telecine cycle number when the area of the motion area is small are set to values whose sum is 1. The values of the coefficients p and q corresponding to the telecine cycle number when the area of the motion region is large are set to values whose sum is greater than 1.

FIG. 23 shows the configuration of the motion information calculation circuit 13.

The first video signal and the second video signal output from the selector 5 are input to the motion information calculation circuit 13.

The motion information calculation circuit 13 includes a difference circuit 12
1, an absolute value calculation circuit 122, a binarization circuit 123, and a motion area counter 124.

The difference circuit 121 calculates a luminance difference between the first video signal and the second video signal for each pixel. The absolute value calculation circuit 122 calculates the absolute value of the luminance difference for each pixel obtained by the difference circuit 121.

The binarization circuit 123 binarizes the absolute value of the luminance difference for each pixel based on a predetermined threshold. That is, the binarization circuit 123 determines whether the pixel has a motion or has no motion. If the absolute value of the luminance difference is larger than the threshold, it is determined that the pixel is a pixel with motion. The motion area counter 124 counts the number of pixels determined to have motion for each field. The number of pixels having motion counted by the motion area counter 124, that is, the area of the motion area is determined by the interpolation coefficient control circuit 12
Sent to

When the area of the motion area with respect to the previous field is equal to or smaller than a predetermined threshold, the interpolation coefficient control circuit 12 determines the coefficients p, p based on the first coefficient table shown in FIG. q is determined, and the coefficients p and q are sent to the corresponding multipliers 6 and 7.

When the area of the motion area with respect to the previous field is larger than the predetermined threshold, the interpolation coefficient control circuit 12 calculates the coefficients p and q based on the second coefficient table shown in FIG. And sends the coefficients p and q to the corresponding multipliers 6 and 7.

[E] Description of Modification of Fourth Embodiment

In the fourth embodiment, the coefficient is adaptively controlled according to the area of the moving area. However, the coefficient is controlled according to the luminance change amount of the moving area (luminance level difference of the moving area). May be adaptively controlled.

That is, as the motion information calculation circuit 13 in FIG. 23, a circuit for calculating the luminance change amount of the motion area is used.

FIG. 24 shows the configuration of the motion information calculation circuit 13 for calculating the amount of change in the luminance of a motion area.

The motion information calculation circuit 13 includes a difference circuit 12
1, an absolute value calculation circuit 122, a binarization circuit 123, and a motion area luminance average calculation circuit 125.

The difference circuit 121 calculates a luminance difference between the first video signal and the second video signal for each pixel. The absolute value calculation circuit 122 calculates the absolute value of the luminance difference for each pixel obtained by the difference circuit 121. The binarization circuit 123 binarizes the absolute value of the luminance difference for each pixel based on a predetermined threshold. That is, the binarization circuit 123 determines whether the pixel has a motion or has no motion. If the absolute value of the luminance difference is larger than the threshold, it is determined that the pixel is a pixel with motion.

The output of the absolute value calculating circuit 122 (the absolute value of the luminance difference for each pixel) is supplied to the moving area luminance average calculating circuit 125.
And the output of the binarization circuit 123 (movement presence / absence determination signal) are input. The motion area luminance average calculation circuit 125 calculates, for each field, the sum of the luminance difference absolute values of the pixels determined to have motion, and divides the calculated sum by the number of pixels determined to have motion. To calculate the luminance change amount (average value) of the motion area. The luminance change amount (average value) of the motion area calculated by the motion area luminance average calculation circuit 125 is sent to the interpolation coefficient control circuit 12.

For example, the interpolation coefficient control circuit 12 shown in FIG.
Two types of coefficient tables as shown in FIG. The first coefficient table shown in FIG. 22A stores the values of the coefficients p and q according to the telecine cycle number when the brightness level difference in the motion area is small. The second coefficient table shown in FIG. 22B stores the coefficients p and q corresponding to the telecine cycle number when the luminance level difference in the motion area is large.

When the luminance change amount of the motion area with respect to the previous field is equal to or smaller than a predetermined threshold, the interpolation coefficient control circuit 12 calculates the coefficient based on the first coefficient table shown in FIG. p and q are determined, and coefficients p and
q is sent to the corresponding multipliers 6,7.

When the luminance change amount of the motion area with respect to the previous field is larger than a predetermined threshold value, the interpolation coefficient control circuit 12 calculates the coefficient p based on the second coefficient table shown in FIG. , Q, and the coefficients p,
q is sent to the corresponding multipliers 6,7.

As the motion information calculation circuit 13, a circuit that calculates both the area of the motion area and the luminance level difference of the motion area is used. When the area of the motion area is large and the luminance change amount of the motion area is large, May employ coefficients p and q whose sum is greater than 1, and otherwise employ coefficients p and q whose sum is 1.

[0149]

According to the present invention, a telecine conversion method for smoothing the motion of an image is realized. Further, according to the present invention, a telecine conversion method is realized in which the motion of an image becomes smooth and the occurrence of flicker can be avoided.

[Brief description of the drawings]

FIG. 1 is a time chart showing a conventional method (2-3 pull-down method) and a telecine conversion method according to a first embodiment.

FIG. 2 is a block diagram showing a configuration of a telecine conversion device for realizing the telecine conversion method according to the first embodiment.

FIG. 3 is a time chart showing signals and the like of respective units in FIG. 2;

FIG. 4 is a block diagram showing a configuration of a telecine conversion device for realizing a telecine conversion method according to a second embodiment.

FIG. 5 is a block diagram showing a configuration of an interpolation coefficient control circuit 12.

FIG. 6 is a circuit diagram showing a circuit for calculating an edge change amount.

FIG. 7 is a schematic diagram showing a filter for calculating Laplacian.

FIG. 8 is a circuit diagram showing a circuit for calculating a luminance change amount.

FIG. 9 is a circuit diagram illustrating a circuit for calculating a contrast change amount.

FIG. 10 is a graph showing a relationship between motion information Det and a target interpolation value MK.

FIG. 11 is a circuit diagram showing a configuration of a region interpolation amount determination unit 23.

FIG. 12 (Target interpolation value MK-previous area interpolation value RK)
9 is a graph showing the relationship between the increase and decrease value KK of the interpolation coefficient with respect to.

FIG. 13 is a schematic diagram showing an interpolation coefficient α for a combination of RK and a telecine cycle number.

FIG. 14 is a schematic diagram showing another example of an interpolation coefficient α for a combination of RK and a telecine cycle number.

FIG. 15 is a schematic diagram for explaining a method of obtaining an interpolation coefficient p for each pixel based on an interpolation coefficient α for each block in the screen.

FIG. 16 is a time chart showing signals and the like of each unit in FIG. 4;

FIG. 17 is an explanatory diagram for explaining a cause of occurrence of flicker when telecine conversion is performed according to the first embodiment or the second embodiment.

FIG. 18 is a time chart showing a conventional 2-3 pull-down method and a telecine conversion method according to the third embodiment.

FIG. 19 is a block diagram showing a configuration of a telecine conversion device for realizing the telecine conversion method according to the third embodiment.

FIG. 20 is a time chart showing signals and the like of respective units in FIG. 19;

FIG. 21 is a block diagram showing a configuration of a telecine conversion device for realizing a telecine conversion method according to a fourth embodiment.

FIG. 22 is a schematic diagram illustrating an example of a first coefficient table and a second coefficient table.

FIG. 23 is a block diagram illustrating a configuration of a motion information calculation circuit.

FIG. 24 is a block diagram illustrating another configuration example of the motion information calculation circuit.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Changeover switch 2, 3, 4 Frame memory 5 Selector 6, 7 Multiplier 8 Adder 9 Maximum value calculation circuit 10 Minimum value calculation circuit 11 2-3 Pulldown control circuit 12 Interpolation coefficient control circuit 13 Motion information calculation circuit

 ──────────────────────────────────────────────────続 き Continuing on the front page (72) Inventor Shugo Yamashita 2-5-5 Keihanhondori, Moriguchi-shi, Osaka Sanyo Electric Co., Ltd. (72) Inventor Haruhiko Murata 2-chome, Keihanhondori, Moriguchi-shi, Osaka No.5-5 Sanyo Electric Co., Ltd. F term (reference) 5C022 BA13 BA17 BA19 5C053 FA30 GA10 GA14 GB40 5C063 BA20 CA05 CA07

Claims (6)

[Claims] 1. A telecine conversion method for converting an image of a 24P movie film into a 60P video signal, wherein after the telecine conversion, a predetermined time after the conversion is such that the integrated amount of display time of each frame image of the movie film becomes equal. A telecine conversion method characterized in that an interpolated image obtained by interpolating frame images before and after a movie film is used as the image of the frame. 2. A telecine conversion method for converting a picture of a 24P movie film into a picture signal of 60P, wherein an interpolated picture obtained by interpolating frame pictures before and after the movie film is used as a picture of a predetermined frame after the conversion. A telecine conversion method, wherein one screen is divided into a plurality of blocks,
When calculating information on motion for each block and generating an interpolated video, an interpolation coefficient for each block is determined based on information on motion of the block for each block, and an interpolation coefficient for each determined block is determined. A telecine conversion method, wherein an interpolated video is generated for each block based on a coefficient. 3. An interpolation coefficient for each pixel is calculated based on an interpolation coefficient determined for each block, and an interpolated image is generated for each pixel based on the calculated interpolation coefficient for each pixel. 3. The telecine conversion method according to claim 2, wherein: 4. A telecine conversion method for converting a picture of a 24P movie film into a picture signal of 60P, wherein an interpolated picture obtained by interpolating frame pictures before and after the movie film is used as a picture of a predetermined frame after the conversion. In the telecine conversion method described above, an interpolation coefficient for which the sum of the two interpolation coefficients is greater than 1 is used as an interpolation coefficient for multiplying the preceding and succeeding frame images, and the obtained interpolation value in pixel units is the interpolation value. If the luminance value of a larger one of the luminance values of the original two frame images used for calculating the luminance value exceeds the larger one of the luminance values of the two frame images based on the interpolation value, A telecine conversion method, wherein the correction is made to be less than or equal to the luminance value of the other. 5. A telecine conversion method for converting a 24P movie film image into a 60P video signal, wherein an interpolated image obtained by interpolating frame images before and after the movie film is used as a converted predetermined frame image. In the telecine conversion method described above, as an interpolation coefficient for multiplying the preceding and succeeding frame images, an interpolation coefficient such that the sum of both interpolation coefficients is 1 and an interpolation coefficient such that the sum of both interpolation coefficients is greater than 1 If the area of the motion area is equal to or smaller than a predetermined value, an interpolation image is generated using an interpolation coefficient such that the sum of both interpolation coefficients becomes 1, and the area of the motion area is larger than the predetermined value. In this case, an interpolated image is generated using an interpolation coefficient such that the sum of both interpolation coefficients is greater than 1, and an interpolated image is generated using an interpolation coefficient such that the sum of both interpolation coefficients is greater than 1. Raw If the obtained pixel-based interpolation value exceeds the larger one of the luminance values of the two original frame images used to calculate the interpolation value, A telecine conversion method, wherein the value is corrected so as to be equal to or less than the larger one of the luminance values of the two original frame images. 6. A telecine conversion method for converting an image of a 24P movie film into a 60P video signal, wherein an interpolated image obtained by interpolating frame images before and after the movie film is used as a converted predetermined frame image. In the telecine conversion method described above, as an interpolation coefficient for multiplying the preceding and succeeding frame images, an interpolation coefficient in which the sum of the two interpolation coefficients is 1 and an interpolation coefficient in which the sum of both the interpolation coefficients is greater than 1 If the amount of change in the brightness level of the motion area is equal to or less than a predetermined value,
An interpolated video is generated using an interpolation coefficient such that the sum of both interpolation coefficients becomes 1, and when the amount of change in the luminance level of the motion area is larger than a predetermined value, the sum of both interpolation coefficients becomes a value larger than 1. When an interpolated image is generated by using such an interpolation coefficient, and an interpolated image is generated by using an interpolation coefficient such that the sum of both interpolation coefficients is greater than 1, the obtained interpolated value in pixel units is If the luminance value of a larger one of the luminance values of the two original frame images used to calculate the interpolation value exceeds the luminance value of the two frame images based on the interpolation value, The telecine conversion method, wherein the correction is made so as to be equal to or less than the luminance value of the larger one.