JPH10126686A - Picture special effect device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide depth information without installing the hardware for operation and without increasing the burden of a central processing unit by using a parameter corresponding to a reduction factor used for the operation for obtaining the read address signal of a frame memory. SOLUTION: A value H showing the enlargement factor or the reduction factor of video deformed by perspective is used as depth information. For operating depth information, it is not necessary to actually operate the coordinate value of a Z-axis and therefore it is not necessary to use a high speed processor for three-dimensional conversion. Thus, the operation is executed even if a processor whose speed is comparatively slow is used. The parameter H is the value required at the time of operating the two-dimensional read address signals (X and Y) supplied to a frame memory 11 and a key information memory 16. Thus, it is not necessary to execute a special operation for obtaining the parameter H and the burden of the central processing unit 20 does not increase.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to enlargement and reduction of an image suitable for use in forming a television broadcast signal and the like.
The present invention relates to an image special effect device that performs deformation such as rotation and movement.

[0002]

2. Description of the Related Art
Enlarge, reduce, rotate, mosaic, and defocus images to form television broadcast signals,
Alternatively, there has been proposed an image special effect device which obtains a special effect such as turning a page of a book.

This conventional image special effect device stores an image signal in a frame memory, and performs transformation such as enlargement, reduction, rotation, and movement of the image when reading the image signal from the frame memory. is there.

In such an image special effect device, for example, when three-dimensional intersections between images are performed, depth information by three-dimensional linear conversion is required.

Conventionally, depth information z of a plane in this three-dimensional space can be obtained as follows: z = Ax + By + C if the equation of the plane is known.

Determining depth information by calculating this equation has the advantage that accurate depth information can be calculated. However, hardware for calculation for calculating this depth information z is required, and Since the coefficients A, B, and C must be calculated separately from the three-dimensional linear conversion, the burden on the central processing unit increases.

In view of the foregoing, it is an object of the present invention to provide depth information without providing hardware for calculating the depth information and without increasing the load on the central processing unit.

[0008]

The image special effect apparatus of the present invention stores an image signal in a frame memory, and enlarges, reduces, and enlarges the image when reading the image signal from the frame memory.
In an image special effect device that performs deformation such as rotation and movement, a parameter corresponding to a reduction ratio used in an operation for obtaining a read address signal of the frame memory for an output image projected on a plane is used as depth information. It is something to do.

According to the present invention, a parameter corresponding to a reduction ratio used in an operation for obtaining a read address signal of the frame memory for an output image projected on a plane is used as depth information. It does not require hardware and does not increase the load on the central processing unit.

[0010]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the image special effect apparatus of the present invention will be described below with reference to the drawings. In FIG. 1, reference numeral 10 denotes a video data input terminal to which digital video data such as a captured image signal is supplied. The video data supplied to the video data input terminal 10 is supplied to a frame memory 11.

The video data read from the frame memory 11 is supplied to a monitor 14 via an interpolation circuit 12 and a mixer 13. Further, 15 represents a key data input terminal key data K _A keying the image is input, as supplies key data K _A to be supplied to the key data input terminal 15 in the key information memory 16 of frame memory I do. The key data read from the key information memory 16 is supplied to the mixer 13 via the interpolation circuit 17, and the mixer 13 supplies a signal obtained by mixing the video data and the key data to the monitor 14. 13a is an output terminal for broadcast video data.

The interpolation circuits 12 and 17 perform interpolation so that images between key frames are continuous.

In this embodiment, the frame memory 11
When reading video data from the key information memory 1
The image obtained on the video screen 14a of the monitor 14 when the key data is read out from the monitor 6 has the effect of enlarging, reducing, rotating, moving, etc. the video data stored in the frame memory 11 and the key data image stored in the key information memory 16. Is added.

In FIG. 1, reference numeral 18 denotes a reference clock generation circuit.
The reference clock generation circuit 18 is connected to a central processing unit (CPU) 20 via the CPU 9 and outputs a clock signal CK, a horizontal synchronization signal HD, and a vertical synchronization signal VD corresponding to each pixel of the video screen 14a of the monitor 14. The clock signal CK, the horizontal synchronizing signal HD and the vertical synchronizing signal VD
Are supplied to the screen address generating circuit 21 respectively.

The screen address generating circuit 21 is connected to a central processing unit 20 via a bus 19, and the screen address generating circuit 21 converts the screen address signals xs and ys so as to correspond to a raster scan of the monitor 14. It is generated sequentially and supplied to the read address generation circuit 22. The screen address signals xs and ys correspond to the video screen 1 of the monitor 14.
4a are address signals supplied in an order corresponding to the raster scan of FIG. Also, this read address generation circuit 2
2 is connected to a central processing unit 20 via a bus 19.

The screen address signals xs and ys
Are output as read address signals X and Y without conversion in the read address generation circuit 22, and are stored in the video data (image signal) stored in the frame memory 11 and the key information memory 16. The key data is output as it is, and an image based on the video data and the key data is displayed on the video screen 14a of the monitor 14 as it is.

On the other hand, a tertiary spatial image in which an image of video data stored in the frame memory 11 and an image of key data stored in the key information memory 16 are enlarged, reduced, rotated and moved. When the conversion is performed, the read address generation circuit 2 is operated based on information input from the control panel 23 by the operation of the operator.
2 is a frame memory 11 and a key information memory 16 based on the standard address signals xs and ys and a plurality of conversion parameters.
To generate read address signals X and Y.

In this case, input information from the control panel 23 is supplied to the central processing unit 20 via the interface circuit 24 and the bus 19. In FIG. 1, reference numeral 25 denotes a RAM used for the operation of the central processing unit 20, and reference numeral 26 denotes a control display device for displaying items such as menus necessary for operating the control panel 23.

By supplying the generated two-dimensional read address signals X and Y to the frame memory 11 and the key information memory 16, the frame memory 11
The predetermined spatial image conversion is performed on the video data stored in the key information memory 16 and the key data stored in the key information memory 16.

Generally, the reading of the frame memory 11 and the key information memory 16 to obtain read address signals X and Y for enlarging, reducing, rotating, and moving an image of video data and key data is performed as follows. .

First, input video data (input key data)
The two coordinate systems are defined as shown in FIG. 2 in order to perform enlargement, reduction, movement, and rotation of the image, and to perform three-dimensional linear conversion. In other words, there are a world coordinate system of rectangular coordinates in XY and a coordinate system in which the projection plane (video screen of the monitor) is an XY plane and the depth direction is Z (plus).

As shown in FIG. 2, the image of the input video data is located on the XY plane of the world coordinate system, and if the center of the video screen 14a is defined as the center of the coordinate system, enlargement, reduction, movement, rotation, etc. The transformation of the image of 4x using homogeneous coordinates
4 can be represented by a transformation matrix T.

That is, a matrix representing the transformation of the movement and the rotation is T ₀
When a matrix representing enlargement, reduction, and perspective transformation is P ₀ , this transformation matrix is a 4-by-4 transformation matrix as follows.

[0024]

The 4 × 4 transformation matrix is a 4 × 4 matrix because rotation transformation and transformations of different dimensions such as enlargement and reduction are represented in the same coordinate system. Is called.

[0026] _{Here, P x = r 11 P 0x} + r 12 P 0y + r 13 P 0z P y = r 21 P 0x + r 22 P 0y + r 23 P 0z P z = r 31 P 0x + r 32 P 0y + r 33 P _{0z S = l x P 0x +} l y P 0y + l z P 0z + S 0 ‥‥ putting and (2),

[0027] It can be expressed as.

The conversion parameters r _{11 to} r ₃₃ used in the equations (1) and (2) are parameters related to scaling, rotation, and SKEW (shear) of each axis in a three-dimensional space of XYZ coordinates. _0x, P _0y and P _0z perspectives value for perspective transformation by applying the perspective when displayed on a video screen 14a, l _x, l _y and l _z
Is a parameter for translating the origin of the coordinate system in the X, Y and Z axis directions, and S ₀ is a parameter for enlarging or reducing the input video.

In this case, the read address generating circuit 22
In the equation (2), the perspective values P _0X , P _0y and P ₀ used for applying the perspective method
_0z is set to a value such that the value of P _0x = 0, P _0y = 0, P _0z = 1 / 16１６ (4) becomes a reference value, whereby the following equation (P _0x P _0y P _0z ) = (0 0 1/16) ‥‥ (5) means that the operator's viewpoint is at −16 on the Z coordinate.

The value of the coordinate value -16 is a value determined arbitrarily and is not a general value. XY
With the coordinates of the image screen 14a as a plane, the center of the image screen 14a as the origin, coordinate values of -4 to +4 are set on the X axis, and coordinate values of -3 to +3 are assumed on the Y axis. Is set.

Next, in equation (3), the video data read from the frame memory 11 and the video screen 14a
Since the data of the image displayed above is two-dimensional data, the parameters in the third row and third column of the equation (3) are not necessary when calculating a two-dimensional address. Therefore, if the T ₃₃ the transformation matrix excluding the third row and the third row of the parameter from the equation (3),

[0032]

However, this is a matrix for converting a two-dimensional point (X, Y) on the frame memory 11 into a point (x _S , y _S ) on the video screen 14a. That is, this transformation matrix T ₃₃ is shown in FIG.
Is a conversion formula for obtaining an image of the video screen 14a from the image signal stored in the frame memory 11 as shown in FIG. 7. In this example, the conversion is performed when the image signal is read from the frame memory 11. The conversion matrix used when reading the image signal is the inverse matrix T ₃₃ ^-1 of the conversion matrix T ₃₃ .

[0034] Here, as a vector of three-dimensional points in two-dimensional space (x ^* y ^*) and the point of the two-dimensional space on the video screen 14a a (x _S y _S) in the same coordinate system in the frame memory 11 It can be expressed as follows.

[0035] ^{(x * y *) = (} X Y H) (x S y S) = (x S y S l) ‥‥ (7)

This H is a value representing an enlargement ratio or a reduction ratio of an image deformed by the perspective method, and this value of H is used as depth information in this embodiment.

[0037] Returning again to equation (6) (7) above, to the vector on the frame memory 11 (X YH), by the action of the transformation matrix T _33, vector on the video screen 14a (x _S y _S 1), so the video screen 14a
Vector above, it will be defined by the formula _{(x S y S l) =} (X Y H) T 33 ‥‥ (8).

When spatial image conversion is performed, three-dimensional conversion is performed when video data is read from the frame memory 11, so that a screen address supplied sequentially to a raster scan is used. Therefore, it is necessary to specify an address on the frame memory. Thus, as can be seen from FIG. 2, (X Y H) = (x S y S l) by performing the calculation of the T ₃₃ ^-1 ‥‥ (9), the address on the video screen 14a (x _S, When y _s ) is specified according to the raster scan, the address signal (X, Y) of the frame memory 11 is specified. Therefore, by supplying the two-dimensional read address signal (X, Y) to the frame memory 11 and the key information memory 16, it is possible to obtain two-dimensional video data and key data subjected to spatial image conversion. it can.

Here, consider the matrix of equation (6).

[0040]

[0041]

By substituting equation (11) for equation (9),

[0043]

Which can be expressed as:

[0045] (X Y H) = (b 11 x s + b 12 y s + b 13 b 21 x s + b 22 y s + b 23 b 31 x s + b 32 y s + b 33) ‥‥ (13)

Therefore,

X = b ₁₁ x _S + b ₁₂ y _S + b ₁₃ Y = b ₂₁ x _S + b ₂₂ y _S + b ₂₃ H = b ₃₁ x _S + b ₃₂ y _S + b ₃₃ (14)

Is calculated. However, since the vector (XYH) on the frame memory 11 is a vector in the homogeneous coordinate system, it can be returned to the two-dimensional coordinate system by normalizing with the parameter H.

[0049]

Thus, the read address generation circuit 22 actually reads (X / H) in order to sequentially read data for the pixel corresponding to the screen address (x _S , y _S ) from the frame memory 11 and the key information memory 16. , Y / H) based on the read address signal (X, Y).

[0051] The next (11), solving the matrix equation T ₃₃ ^-1. Here, the parameters a _{11 to} a ₃₃ are obtained from the equations (10),

A ₁₁ = r ₁₁ , a ₁₂ = r ₁₂ , a ₁₃ = _Px a ₂₁ = r ₂₁ , a ₂₂ = r ₂₂ , a ₂₃ = P _y a ₃₁ = l _x , a ₃₂ = l _y , a ₃₃ = S ‥‥ (16)

Therefore,

[0054]

Is as follows.

Therefore, from equation (14), the parameter H as depth information is given by the following equation:

[0057]

Is obtained by

Next, according to the control parameters inputted from the control panel 23 by the operation of the operator,
(1) perspective value P _0x and P _0y associated with perspective of type parameters,

P _0x = 0, P _0y = 0 (19) and the enlargement / reduction ratio S ₀ is

Assuming that S ₀ = 1 ‥‥ (20) is set, from equation (2),

[0062] _{P x = r 13 P 0z P} y = r 23 P 0z S = l z P 0z +1 ‥‥ (21)

Is obtained. Here, the values of the expressions (19) and (20) are determined in order to simplify the expressions described later. In actual operation, P _0x =
0, P _0y = 0, S ₀ = 1 are often set.

When the values of these parameters P _x , P _y and S are substituted into equation (17), the following is obtained.

[0065]

In this example, the value of the parameter H obtained by the equation (22) is used as depth information. This is because the H is proportional to the actual Z coordinate value of the three-dimensional space at the time of three-dimensional conversion.

Further, this depth information will be described with reference to FIGS. 3 and 4 for a specific example. In the example shown in FIG.
This is an example in which the object 30 is translated in the Z-axis direction by a distance zo from the origin 0 with respect to a distance -pz) from the image to the viewpoint P.

In this case, from the viewpoint P to the Z of the object 30
Since the distance to the point C on the axis is zo-pz and the distance from the viewpoint P to the projection plane is -pz, the depth information z is as follows.

[0069]

A transformation matrix T ₃₃ ^-1 for obtaining an image as shown in FIG. 3 on the projection plane (video screen 14a) from the object 30 (image signal stored in the frame memory 11) is as follows. .

[0071]

Therefore, a parameter H = pz / (pz−) for enlarging or reducing the input image of the transformation matrix T ₃₃ ^-1.
zo) and the above-described depth information z are the same.

FIG. 4 shows an example in which the object 30 is rotated by an angle θ about the X axis with respect to the viewpoint P (the distance from the origin 0 to the viewpoint P−pz).

In this case, as shown in FIG. 4, the point B of the rotated object 30 is projected on the point A on the projection plane (the image screen 14a). Ask. This value can be obtained by the following simultaneous linear equation.

[0075] Here, y0 indicates the distance (0, A).

[0076]

Therefore, the depth information z is as follows.

[0078]

A transformation matrix T ₃₃ ^-1 for obtaining an image as shown in FIG. 4 on the projection plane (video screen 14a) from the object 30 (image signal stored in the frame memory 11) is as follows. . In the example of FIG. 4, the object 3 is set with respect to the viewpoint P (the distance from the origin 0 to the viewpoint P−pz).
Since 0 is rotated by θ about the X axis, the three-dimensional linear transformation matrix T is as follows.

[0080]

Excluding the elements in 3 rows and 3 columns, the following is the result.

[0082]

Further, when this inverse matrix is obtained, it is as follows.

[0084]

Therefore, the parameter H for enlarging or reducing the input image of the transformation matrix T ₃₃ ^-1 is as follows.

[0086]

It can be seen that this parameter H is the same as the depth information z.

As described above, in the present embodiment, the depth information does not use the actual Z-axis coordinate value of the three-dimensional space, but the parameter H obtained by the equation (22).
The use of the value has the following effects.

When calculating this depth information, it is not necessary to actually calculate the coordinate values of the Z-axis, so that the one-dimensional calculation can be omitted. As a result, it is not necessary to use a high-speed processor for three-dimensional conversion, and depth information can be calculated even when a relatively low-speed processor is used.

Since this parameter H is a value required when calculating the two-dimensional read address signal (X, Y) to be supplied to the frame memory 11 and the key information memory 16, this parameter H is There is an advantage that the calculation does not require any particular operation, does not increase the load on the central processing unit 20, and does not require hardware for this operation.

It is to be noted that the present invention is not limited to the above-described embodiment, and it goes without saying that various other configurations can be taken without departing from the gist of the present invention.

[0092]

According to the present invention, since the parameter H corresponding to the reduction ratio used in the operation for obtaining the read address signal of the frame memory is used as depth information, an operation is particularly required to obtain this depth information. Therefore, there is an advantage that the load on the central processing unit is not increased and the hardware for calculating the depth information is not required.

[Brief description of the drawings]

FIG. 1 is a configuration diagram showing an embodiment of an image special effect device of the present invention.

FIG. 2 is a diagram for describing the present invention.

FIG. 3 is a diagram for explanation of the present invention.

FIG. 4 is a diagram for describing the present invention.

[Explanation of symbols]

10 video data input terminal, 11 frame memory,
12 interpolation circuit, 14 monitors, 14a video screen, 19
Bus, 20 central processing unit, 22 read address generation circuit, 23 control panel, 25 RAM

Claims (1)

[Claims] 1. An image special effect device in which an image signal is stored in a frame memory and an image is deformed such as enlargement, reduction, rotation, movement or the like when the image signal is read out from the frame memory. An image special effect device using a parameter corresponding to a reduction ratio used in an operation for obtaining a read address signal on the frame memory for the output image obtained as depth information.