JP3582426B2 - Coordinate generation method and coordinate generation circuit - Google Patents

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a coordinate generating method and a coordinate generating method for generating a coordinate value of a pixel after a scaling process in a scaling circuit for performing a scaling process for increasing the number of output pixels greater than the number of input pixels or a scaling process for reducing the number of output pixels less than the number of input pixels. It relates to a generation circuit.
[0002]
[Prior art]
Conventionally, fixed pixel display devices such as a liquid crystal display and a plasma display (hereinafter, referred to as PDP) have been known. The resolution of a device such as a computer for inputting a video signal to such a fixed pixel display device varies. Therefore, in order to support various input video signals, a scaling circuit for performing a scaling process (resolution conversion process or scaling process) of the input video signal is required.
[0003]
[Problems to be solved by the invention]
2. Description of the Related Art Scaling circuits for enlarging and reducing video signals have become popular, but in recent years it has been required to improve the image quality. In addition, liquid crystal displays and PDPs are often used for displaying guidance at stations and airports, for displaying stock prices, and for displaying computer screens. In these images, fine characters and ruled lines are often scrolled and displayed. In addition, it is considered that the number of images that cannot be displayed on one screen, such as homepages, are displayed with a function of automatically scrolling.
However, in the conventional scaling circuit, since the phase relationship between the input image before the scaling process and the output image after the scaling process is determined independently of scrolling, when the input image scrolls, the phase relationship changes with time. As a result, there is a problem that the image quality is deteriorated.
SUMMARY An advantage of some aspects of the invention is to provide a coordinate generation method and a coordinate generation circuit that can obtain a high-quality output image even when an input image scrolls.
[0004]
[Means for Solving the Problems]
According to the coordinate generation method of the present invention, the coordinate value is reset to an initial value each time the horizontal synchronization signal is input in the case of the horizontal scaling process, and the vertical synchronization signal is reset in the case of the vertical scaling process. The procedure of resetting the coordinate value to the initial value each time it is input, and adding the value obtained by multiplying the inter-pixel distance of the input pixel by the reciprocal of the enlargement ratio or reduction ratio to the coordinate value generated at the immediately preceding generation timing A procedure of repeating the addition value as a coordinate value at the current generation timing at each generation timing, and each time a certain number of vertical synchronization signals are input, the initial value is changed by a predetermined amount with respect to the immediately preceding initial value. And a changing procedure.
Further, as one configuration example of the coordinate generation method of the present invention, the method has a procedure of adding an offset value according to a motion vector of an input image to the coordinate value at the current generation timing.
[0005]
Further, the coordinate generation circuit of the present invention resets the coordinate value to an initial value each time a horizontal synchronization signal is input in the case of horizontal scaling processing, and resets the vertical synchronization in the case of vertical scaling processing. Reset means (5) for resetting coordinate values to initial values each time a signal is input, and a value obtained by multiplying the distance between input pixels by the reciprocal of the enlargement ratio or reduction ratio at the immediately preceding generation timing. Coordinate value generating means (1, 3, 4) that repeats adding to the coordinate value and using the value after the addition as the coordinate value at the current generation timing for each generation timing, and a certain number of vertical synchronization signals are input. An initial value generating means (2) for changing the initial value each time by a predetermined amount from the immediately preceding initial value.
As one configuration example of the coordinate generation circuit of the present invention, an offset generation means (10) for generating an offset value according to a motion vector of the input image, and the coordinates at the current generation timing generated by the coordinate value generation means. Adding means (11) for adding the offset value to the value.
[0006]
BEST MODE FOR CARRYING OUT THE INVENTION
[First Embodiment]
Next, embodiments of the present invention will be described in detail with reference to the drawings. FIG. 1 is a block diagram showing a configuration of a coordinate generation circuit according to a first embodiment of the present invention.
The coordinate generation circuit shown in FIG. 1 is for horizontal scaling or vertical scaling. Therefore, when performing the scaling process in both the horizontal and vertical directions, two coordinate generating circuits shown in FIG. 1 are required.
[0007]
Note that the scaling process of the present invention can be rephrased as a resolution conversion process. That is, the enlargement process is a process for increasing the resolution, and the reduction process is a process for decreasing the resolution. Therefore, the enlargement / reduction processing of the present invention does not necessarily mean enlargement or reduction of the image area.
[0008]
The coordinate generation circuit according to the present embodiment includes an enlargement / reduction ratio setting unit 1 that outputs a set value obtained by multiplying the distance between input pixels by the reciprocal of the enlargement ratio or reduction ratio of the scaling process. The set value output from the enlargement / reduction ratio setting unit 1 is supplied to one input of an adder 3. The output of the adder 3 is supplied to one input terminal of a switch 5 through a D flip-flop 4.
[0009]
The initial value output from the initial value generation unit 2 is input to the other input terminal of the switch 5 serving as reset means. The switch 5 performs a selecting operation according to a reset signal input from the reset input terminal 7 for resetting the output coordinate value to an initial value. The reset input terminal 7 receives a horizontal synchronization signal as a reset signal when the present coordinate generation circuit is a circuit for horizontal scaling processing, and a vertical synchronization signal when the present coordinate generation circuit is a circuit for vertical scaling processing. The signal is input as a reset signal.
[0010]
Further, a clock signal indicating the generation timing of the output coordinate value is input to the D flip-flop 4 from the clock input terminal 6. A clock signal is input to the clock input terminal 6 when the coordinate generation circuit is a circuit for horizontal scaling processing, and a horizontal synchronizing signal is supplied to the clock input terminal 6 when the coordinate generation circuit is a circuit for vertical scaling processing. Is entered as
[0011]
The output of the switch 5 is used as an output coordinate value to a scaling circuit (not shown). This coordinate value indicates the spatial position of the pixel to be interpolated by the scaling process. The integer part of this coordinate value indicates the address of the pixel in the horizontal direction in the input signal before scaling in the case of horizontal scaling processing, and indicates the address of the number of lines in the case of vertical scaling processing. become. The decimal part of the coordinate value indicates the spatial position of the pixel to be interpolated at these addresses.
[0012]
The output of the switch 5 is used as an output coordinate value to the scaling circuit, while being supplied to the other input of the adder 3.
Thus, the enlargement / reduction ratio setting unit 1, the adder 3, and the D flip-flop 4 constitute coordinate value generation means. The coordinate value generating means adds the set value from the enlargement / reduction ratio setting unit 1 to the output coordinate value generated at the immediately preceding generation timing, and sets the value after the addition as the output coordinate value at the current generation timing. It repeats every timing, that is, every clock if the present coordinate generation circuit is a circuit for horizontal scaling processing, and every one horizontal synchronization signal if it is a circuit for vertical scaling processing.
[0013]
The initial value generation unit 2 outputs an initial value that is equal to or greater than 0 and less than the pixel-to-pixel distance of the input pixels (for example, 0 to less than 1 when the pixel-to-pixel distance is 1). The initial value generator 2 receives the horizontal synchronizing signal HS, the vertical synchronizing signal VS, and the initial value control signal FC for increasing or decreasing the initial value at regular time intervals. A counter (not shown) for counting the number of the horizontal synchronizing signal HS or the vertical synchronizing signal VS is provided inside the initial value generating unit 2. The initial value control signal FC includes information indicating how much the initial value is changed when a certain number of vertical synchronization signals VS are counted, and the initial value generation unit 2 initializes the initial value based on this information. Change the value.
[0014]
Here, the case where the scaling circuit enlarges the video signal by a factor of 5/3 in the horizontal direction will be briefly described with reference to FIG.
2B in FIG. 2B indicates a spatial position of the input pixel before the scaling processing. The numerical value above the mark indicates the address (horizontal coordinate value) of the input pixel. The address at the time when the horizontal synchronization signal HS is input is set to 0, and each pixel has an address of 0, 1, 2, 3, 4,....
[0015]
2C indicate the spatial position of the output pixel after the scaling process. The numerical value above the mark indicates the address (output coordinate value) of the output pixel. When the initial value output from the initial value generation unit 2 is 0, the output coordinate value is reset to 0 when the horizontal synchronizing signal HS is input to the switch 5, and thereafter, from the enlargement / reduction ratio setting unit 1 every clock. The output set value 3/5 = 0.6 is added to the output coordinate value. Thus, the addresses of the output pixels have values of 0, 0.6, 1.2, 1.8, 2.4,....
[0016]
Next, a scaling process performed by the scaling circuit will be described. Here, the scaling process in the horizontal direction will be described as an example. For example, in a scaling process called linear interpolation, when an output pixel is generated at the position of the output coordinate value Xs shown in FIG. 3, the level of the input pixel of the coordinate value x is f (x), and the level of the input pixel of the coordinate value x + 1 is Is f (x + 1), the level F (Xs) of the output pixel to be obtained is obtained by the following equation.
F (Xs) = a × f (x + 1) + (1−a) × f (x) (1)
[0017]
“a” is a coefficient of the interpolation calculation determined based on the scaling ratio. For example, when the input image is enlarged by a factor of 5/3 as shown in FIG. 2, the level of the output pixel is determined by Expression (1) as follows.
F (0) = 1.0 × f (0)
F (0.6) = 0.6 × f (1) + 0.4 × f (0)
F (1.2) = 0.2 × f (2) + 0.8 × f (1)
F (1.8) = 0.8 × f (2) + 0.2 × f (1)
F (2.4) = 0.4 × f (3) + 0.6 × f (2)
F (3.0) = 1.0 × f (3) (2)
[0018]
In the scaling process described above, if the initial value of the output coordinate value is always constant, an image in which a thin vertical line scrolls horizontally or an image in which a horizontal line scrolls vertically can be obtained by linear interpolation. Since the position of the output pixel changes with respect to the input pixel, the image quality deteriorates such that the line width appears to change.
In the present invention, since the initial value is changed in accordance with the scroll image, it is possible to make the position of the output pixel and the level of the output pixel obtained by the scaling process close to the input pixel. The scaling process can be performed without occurrence.
[0019]
Hereinafter, the operation of the coordinate generation circuit according to the present embodiment will be described. First, the enlargement / reduction ratio setting unit 1 will be described. In the enlargement / reduction ratio setting unit 1, a value obtained by multiplying the reciprocal of the enlargement ratio by the reciprocal of the enlargement ratio is set as the set value in the case of the enlargement process. Is set as a set value.
[0020]
For example, when the distance between the input pixels is 1 and the input pixel is enlarged 1.6 times, 1 × 1 / 1.6 = 0.625 is set as the set value in the enlargement / reduction ratio setting unit 1 and the input pixel When the pixel-to-pixel distance is 1 and is reduced to 0.8 times, 1 × 1 / 0.8 = 1.25 is set as the set value. This set value indicates a spatial distance between output pixels after the scaling processing.
[0021]
The set value output from the enlargement / reduction ratio setting unit 1 is supplied to one input of an adder 3. The output of the adder 3 is supplied to one input terminal of a switch 5 through a D flip-flop 4. The output of the switch 5 is used as an output coordinate value to the scaling circuit, while being supplied to the other input of the adder 3.
The adder 3 generates output coordinate values to the scaling circuit by adding the set values output from the enlargement / reduction ratio setting unit 1 as needed.
[0022]
A clock signal is input to the D flip-flop 4 when the present coordinate generation circuit is a circuit for horizontal scaling processing, and when the coordinate generation circuit is a circuit for vertical scaling processing, the horizontal synchronizing signal is a clock signal. Is entered as
Therefore, in the case of horizontal scaling processing, the D flip-flop 4 takes in the output of the adder 3 every time a clock signal is input, outputs this as an output coordinate value (horizontal address), and outputs the vertical coordinate. In the case of the scaling process, the output of the adder 3 is fetched every time the horizontal synchronizing signal is input, and is output as an output coordinate value (vertical address).
[0023]
The other input terminal of the switch 5 receives the initial value output from the initial value generator 2. The switch 5 normally selects the output of the D flip-flop 4, but when a reset signal is input from the reset input terminal 7, selects and outputs this initial value. The reset input terminal 7 receives a horizontal synchronization signal as a reset signal when the present coordinate generation circuit is a circuit for horizontal scaling processing, and a vertical synchronization signal when the present coordinate generation circuit is a circuit for vertical scaling processing. The signal is input as a reset signal. Therefore, in the case of horizontal scaling processing, the output coordinate value is reset to the initial value each time a horizontal synchronization signal is input, and in the case of vertical scaling processing, the output coordinate value is reset each time a vertical synchronization signal is input. The value is reset to the initial value.
[0024]
The initial value generator 2 outputs an initial value that is equal to or greater than 0 and less than the distance between input pixels. The initial value generator 2 receives the horizontal synchronizing signal HS, the vertical synchronizing signal VS, and the initial value control signal FC for increasing or decreasing the initial value at regular time intervals. The initial value control signal FC includes information indicating how much the initial value is changed when a certain number of vertical synchronization signals VS are counted, and the initial value generation unit 2 initializes the initial value based on this information. Change the value.
[0025]
Next, the operation of the coordinate generation circuit according to the present embodiment will be specifically described with reference to FIGS. In FIG. 4, an input pixel indicated by a circle has a value of 255 when A / D converted to 8 bits, and an input pixel indicated by a circle has a value of 0 when A / D converted to 8 bits. You have. The numerical values above the circles and ● indicate the addresses of the input pixels. In FIG. 5 to FIG. 7, a mark “出力” indicates an output pixel after the scaling process. Numerical values above the crosses indicate the addresses (output coordinate values) of the output pixels. Numerical values below the triangles indicate the level of each output pixel obtained when linear interpolation is performed. Note that an output pixel having no numerical value below the triangle has a value of 255.
[0026]
First, it is assumed that one line in an input image for one frame is as shown in FIG. In the line shown in FIG. 4A, it is assumed that the pixel level sequentially changes as shown in FIGS. 4B, 4C, and 4D every 30 frames. When the value of the initial value output from the initial value generation unit 2 is fixed to 0 and the input image shown in FIG. , FIG. 4 (c), and FIG. 4 (d) are enlarged as shown in FIGS. 5 (a), 5 (b), 5 (c), and 5 (d), respectively.
[0027]
In the input image before the scaling processing shown in FIG. 4, the image at the pixel level 0 only moves horizontally, whereas in the output image after the scaling processing shown in FIG. Is changing. Such an image looks like an image having undulation, so that the image quality is deteriorated.
[0028]
Here, when the initial value control signal FC including information indicating that the initial value is increased by a predetermined amount (here, 0.2) every time the vertical synchronization signal is input 30 times is input to the initial value generation unit 2, Since the initial value of the output coordinate value changes as 0, 0.2, 0.4, 0.6,... Every 30 frames, FIG. 4 (a), FIG. 4 (b), FIG. The lines in (c) and FIG. 4 (d) are enlarged 5/3 times as shown in FIGS. 6 (a), 6 (b), 6 (c) and 6 (d), respectively. Since the output image that has been scaled as shown in FIG. 6 appears to be scrolled in the horizontal direction, the image quality does not deteriorate even if the input image is scrolled as shown in FIG.
[0029]
Further, the initial value is set to 0.2, and the initial value control signal FC including information indicating that the initial value is increased by 0.2 every time the vertical synchronization signal is input 30 times is input to the initial value generation unit 2. Then, the initial value of the output coordinate value changes as 0.2, 0.4, 0.6, 0.8,... Every 30 frames, so that FIG. 4B, 5C, and 4D are 5/3 times as shown in FIGS. 7A, 7B, 7C, and 7D, respectively. It is enlarged. In the example of FIG. 7, a high-quality output image closer to the input image of FIG. 4 is obtained.
[0030]
Since the initial value output from the initial value generation unit 2 is a value that is equal to or greater than 0 and less than the distance between pixels of the input pixel, the initial value is increased according to the initial value control signal FC. If the value is equal to or greater than the distance, a value obtained by subtracting the distance between pixels from this value is set as a new initial value.
[0031]
[Second Embodiment]
Next, a second embodiment of the present invention will be described. In this embodiment, the basic configuration is the same as that described in the first embodiment, but the scaling circuit may not need to perform real-time processing in some cases. That is, there is a method in which an output coordinate value is obtained in advance by a coordinate generation circuit, and the value is stored as a register value of a scaling circuit to perform a scaling process.
[0032]
In the case of the enlarging process, it is sufficient to prepare up to the number of registers corresponding to the number of addresses after the scaling process, and to update the value of the register each time the scaling process is performed. In the example of FIG. 5, with respect to the address 0 at the time of reset, the decimal part is an address that makes a round up to 3.0, and after 3.0, the value of this decimal part is repeated. It is only necessary to prepare
As described above, even when the output coordinate value is determined in advance by calculation, the horizontal synchronization signal and the vertical synchronization signal are counted, and the initial value of the output coordinate value is changed at regular time intervals. The result can be obtained.
[0033]
[Third Embodiment]
FIG. 8 is a block diagram showing a configuration of a coordinate generation circuit according to a third embodiment of the present invention. The same components as those in FIG. 1 are denoted by the same reference numerals. In the first embodiment, the effect is obtained only for an image in which the entire image of one frame is scrolled. However, in the present embodiment, the motion vector of the input image is detected, and the motion vector is detected according to the motion vector. By generating the offset value, a coordinate generation circuit capable of supporting partial scrolling of the input image is realized.
[0034]
The motion vector of the input image is input to the motion vector input terminal 9 from a motion vector detection circuit (not shown).
When a motion vector is input from the motion vector input terminal 9, the offset generation unit 10 generates an offset value based on the motion vector such that the position of the output pixel and the level of the output pixel are close to the input pixel.
[0035]
For example, when a motion vector that moves one pixel to the left every 30 frames is detected, an offset value that increases by 0.2 in synchronization with the generation of this motion vector may be generated. If a motion vector that moves two pixels to the left every 30 frames is detected, an offset value that increases by 0.4 in synchronization with the generation of the motion vector may be generated.
[0036]
The adder 11 adds an offset value to the output of the switch 5. In this way, the offset value is added to the output coordinate value, and it is possible to cope with partial scrolling of the input image.
Note that a scroll image in which only a part of the image is stopped can be used. In this case, the initial value change according to the entire scroll is performed by the initial value generation unit 2, and the offset generation unit 10 only needs to generate an offset value that cancels the change in the initial value only for the stop portion of the input image. .
[0037]
Further, it is also possible to cope with the scroll of the input image only with the offset generating unit 10 and the adder 11 without using the initial value changing function of the initial value generating unit 2. In this case, the offset value according to the scroll direction and the amount of movement can be automatically generated by the offset generation unit 10.
[0038]
【The invention's effect】
According to the present invention, every time a certain number of vertical synchronizing signals are input, by having a procedure of changing the initial value by a predetermined amount with respect to the immediately preceding initial value, the output pixel after the scaling process in accordance with the scroll image It is possible to change the initial value of the coordinate values of, so that the phase relationship between the input pixel before the scaling process and the output pixel after the scaling process can be matched, and the deterioration of the image quality due to the scrolling of the input image can be reduced. Can be reduced.
[0039]
Further, by having a procedure of adding an offset value according to the motion vector of the input image to the coordinate value at the current generation timing, it is possible to cope with partial scrolling of the input image, and only a part of the image is stopped. Such a scroll image can be supported.
[Brief description of the drawings]
FIG. 1 is a block diagram illustrating a configuration of a coordinate generation circuit according to a first embodiment of the present invention.
FIG. 2 is a diagram illustrating an example of a scaling process performed by a scaling circuit.
FIG. 3 is a diagram illustrating a scaling process.
FIG. 4 is a diagram illustrating an example of an input image.
5 is a diagram showing an output image when an enlargement process is performed on the input image of FIG. 4 while fixing an initial value of an output coordinate value.
6 is a diagram illustrating an example of an output image when an enlargement process is performed on the input image of FIG. 4 using the coordinate generation circuit of FIG. 1;
7 is a diagram illustrating another example of an output image in a case where an enlargement process is performed on the input image of FIG. 4 using the coordinate generation circuit of FIG. 1;
FIG. 8 is a block diagram illustrating a configuration of a coordinate generation circuit according to a third embodiment of the present invention.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 ... Enlargement / reduction rate setting part, 2 ... Initial value generation part, 3 ... Adder, 4 ... D flip-flop, 5 ... Switch, 6 ... Clock input terminal, 7 ... Reset input terminal, 8 ... Coordinate output terminal, 9 ... Motion vector input terminal, 10 ... offset generator, 11 ... adder

Claims (4)

In a scaling circuit for performing an enlargement process for increasing the number of output pixels from the number of input pixels or a reduction process for reducing the number of output pixels from the number of input pixels, a coordinate generation method for generating a coordinate value of a pixel after the enlargement / reduction process
In the case of horizontal scaling, the coordinate value is reset to an initial value each time a horizontal synchronization signal is input. In the case of vertical scaling, the coordinate value is reset each time a vertical synchronization signal is input. Steps to reset the values to their initial values,
A value obtained by multiplying the inter-pixel distance of the input pixel by the reciprocal of the enlargement ratio or the reduction ratio is added to the coordinate value generated at the immediately preceding generation timing, and the value after the addition is the coordinate value at the current generation timing. Steps to be repeated at each generation timing,
Changing the initial value by a predetermined amount from the immediately preceding initial value each time a certain number of vertical synchronization signals are input. The coordinate generation method according to claim 1,
A coordinate generation method comprising a step of adding an offset value according to a motion vector of an input image to the coordinate value at the current generation timing. In a scaling circuit that performs an enlargement process of increasing the number of output pixels from the number of input pixels or a reduction process of reducing the number of output pixels from the number of input pixels, a coordinate generation circuit that generates a coordinate value of the pixel after the enlargement / reduction process,
In the case of horizontal scaling, the coordinate value is reset to an initial value each time a horizontal synchronization signal is input. In the case of vertical scaling, the coordinate value is reset each time a vertical synchronization signal is input. Reset means for resetting the value to an initial value;
A value obtained by multiplying the inter-pixel distance of the input pixel by the reciprocal of the enlargement ratio or the reduction ratio is added to the coordinate value generated at the immediately preceding generation timing, and the value after the addition is the coordinate value at the current generation timing. Value generation means for repeating the operation for each generation timing,
An initial value generating means for changing the initial value by a predetermined amount from the immediately preceding initial value every time a certain number of vertical synchronizing signals are input. 4. The coordinate generating circuit according to claim 3,
Offset generating means for generating an offset value according to the motion vector of the input image,
A coordinate generating circuit comprising: an adding unit that adds the offset value to the coordinate value at the current generation timing generated by the coordinate value generating unit.

Images