JPH07262398A - Method and device for processing image - Google Patents

Abstract

PURPOSE:To display a dynamic image such as a portrait or an animation image in colors at low cost with a little data. CONSTITUTION:When a start flag SF is turned to [1] (step S40) in the case of preparing the dynamic image for the color portrait, closed curved corresponding to the respective sections of the portrait are prepared (step S41) and next, deformation information data are read out to deform the prepared closed curved (steps 42 and 43). Afterwards, the boundary and inside of closed curves A to F are painted out (step S44) and the former display is cleared and changed into the prepared display (step S45). This processing is repeated until an address pointer AD to designate the deformation information is turned to the final address. Thus, even when displaying the dynamic image of the portrait in colors, it is not necessary to provide data for the entire part or one part of the portrait or the like as many as the number of screens to move the image but it is enough to simply provide closed curve data, color data and sequence data for changing the expression of the portrait.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing method and an apparatus therefor, and more particularly to an image processing method capable of drawing a moving image such as an animation or a portrait with a small amount of data and an apparatus for realizing the method.

[0002]

2. Description of the Related Art Conventionally, image data in a bit array format is often used in animation, games, etc., and characters and background data are displayed by this image data. The image data in the bit array format is composed of dots, and each dot has a display color number or a palette number. That is, in order to display a portrait or an animation image in color on a computer screen or the like, it has color data for each pixel. Further, when displaying a moving image of a portrait or an animation image on a computer screen or the like in a conventional image processing device, the image data corresponding to each moving image is provided for the number of moving screens.

[0003]

By the way, in the case of the conventional image processing apparatus, when a moving image of a portrait or an animation image is displayed in color on a computer screen or a television screen, at least all or one of the portrait or the animation image is displayed. There is a problem in that it is necessary to have as many copies of data as the number of screens that move the image, and inevitably a large amount of data is required. In addition, the capacity of the memory device for processing a large amount of data is increased, and the cost is increased in this respect.

Therefore, an object of the present invention is to provide an image processing method and apparatus capable of displaying moving images such as portraits and animation images in color with a small amount of data and at low cost.

[0005]

In order to achieve the above object, the image processing method according to the present invention provides a predetermined basic image with a plurality of closed curves whose colors are omitted and which have a predetermined priority. Create a combination of them and create a plurality of deformed images by deforming the closed curves that make up the created basic image according to the deformation information. It is characterized in that a plurality of predetermined color images are created by filling in with.

As a preferred mode, for example, the closed curve may be formed by one or more of a straight line and a Bezier curve. For example, as described in claim 3, in the process of creating the predetermined basic image with a plurality of closed curves in which colors are omitted, closed curve data including predetermined coordinate data of each closed curve is created to create each of the plurality of closed curves. In the processing of filling the boundaries of the plurality of closed curves and the inside thereof with the designated color, color data for designating the filling color may be provided for each closed curve. For example, as described in claim 4, the process of creating the closed curve by a straight line is to form a straight line by a large number of dots, determine two dot positions forming both ends of the straight line, and The process of calculating the slope,
A process of accumulating an error when the x-coordinate or the y-coordinate is increased by one unit according to this inclination, and changing the position of the x-coordinate or the y-coordinate of the next dot only when the error exceeds a predetermined value. And a process of subtracting a constant value from the accumulated error.

For example, as described in claim 5, a plurality of sequence data is prepared which is a combination of a plurality of types of deformation information for deforming the closed curve of the basic image, and one selected sequence data among them is constituted. You may make it deform | transform sequentially the closed curve which comprises a basic image based on deformation information. For example, as described in claim 6, a plurality of pieces of deformation information for deforming the closed curve of the basic image are prepared inside the small sequence data that is a combination of a plurality of types, and a required small amount is selected from the plurality of small sequence data. The sequence data may be created by selecting and combining a plurality of sequence data, and the closed curve forming the basic image may be sequentially deformed based on the deformation information forming the sequence data. For example, as described in claim 7,
The deformation information for deforming the closed curve of the basic image may be created by an input operation, and the closed curve forming the basic image may be modified according to the created deformation information.

According to another aspect of the image processing apparatus of the present invention, a closed curve creating means for creating a predetermined basic image by combining a plurality of closed curves whose colors are omitted and which has a predetermined priority, and the created basic image. Deformation information setting means for setting deformation information for deforming each of the closed curves forming the image, image deforming means for deforming the closed curve of the basic image according to the deformation information to create a plurality of deformed images, and a plurality of created deformations. And a color image creating means for creating a plurality of predetermined color images by filling the boundary and the inside of each closed curve in the image with a predetermined color in accordance with the priority order.

Further, for example, as described in claim 9, the closed curve creating means is one of a straight line and a Bezier curve.
A closed curve may be created by one or more. For example, as described in claim 10, the closed curve creating means,
Created by closed curve data consisting of predetermined coordinate data of each closed curve for creating each of a plurality of closed curves forming a predetermined basic image, the color image creating means,
Color data corresponding to each closed curve may be stored, and the boundary of each closed curve and the inside thereof may be filled with a color corresponding to the stored color data. For example, as described in claim 11, a display means for displaying a predetermined color image created by the color image creating means may be further provided. For example, as described in claim 12, a printing means for printing a plurality of color images created by the color image creating means may be further provided. For example, when the closed curve creating means creates a closed curve by a straight line composed of a large number of dots, the closed curve creating means determines two dot positions forming both ends of the straight line and calculates an inclination of the dot position. Means for accumulating the error when the x-coordinate or the y-coordinate is increased by one unit according to the inclination, and the x-coordinate or the y-coordinate of the next dot only when the error exceeds a predetermined value. And a means for subtracting a constant value from the accumulated error.

For example, as described in claim 14, the deformation information setting means includes sequence data composed of a combination of a plurality of deformation information, sequence data storage means for storing a plurality of sequence data, and a plurality of sequence data. Sequence data selecting means for selecting one sequence data may be provided. For example, as described in claim 15, the modification information setting means includes a small sequence data storage means for storing a plurality of small sequence data composed of a combination of a plurality of modification information and a necessary one from the plurality of small sequence data. Sequence data creating means for creating sequence data by selecting and combining them may be provided. For example, as described in claim 16, the deformation information setting unit includes a deformation information creating unit that creates deformation information by an input operation, and a deformation information storage unit that stores the deformation information created by the deformation information creating unit. It may be provided.

[0011]

In the present invention, when a moving image of a color image (for example, a portrait or an animation image) is created, first, an image is created by a plurality of closed curves in which colors are omitted, and then
The created closed curve is deformed to prepare a plurality of images, and the boundary and the inside of the deformed closed curve corresponding to the plurality of images are filled to create a moving image of a color image. Therefore, even when displaying a portrait video in color, it is not necessary to have all or part of the data of the portrait, etc. for the number of screens that move the image as in the past, but simply changing the closed curve data, color data and facial expression. It is possible to display a moving image in color if it has sequence data (a plurality of pieces of deformation information) to be performed. As a result, a moving image such as a portrait or an animation image can be displayed in color at a low cost while the amount of data is small and the capacity of the memory device is small.

[0012]

Embodiments of the present invention will be described below with reference to the drawings. Configuration of Portrait Making Device FIG. 1 is a configuration diagram showing a first embodiment of a portrait making device (corresponding to an image processing device) for realizing an image processing method according to the present invention. In FIG. 1, the portrait creation device is roughly divided into a CPU 1, a ROM 2, a RAM 3, a switch section 4, a display section 5, a color printer 6 and a bus 7 connecting the respective sections. The CPU 1 controls the entire apparatus, and when a drawing command is input from the switch unit 4, a process necessary to create a portrait based on a control program stored in the ROM 2 corresponding to the command information ( For example, the data calculation of the straight line or the Bezier curve necessary for drawing the outline of the portrait or each part, the filling process, the graphic transformation process, and the data interpolation process) are performed. Also, CPU
Reference numeral 1 has an internal register, and stores the values of flags and pointers in the internal register.

The ROM 2 stores a control program of the CPU 1 and parameters such as closed curve creation data, closed curve color data, sequence data, and deformation information representing the contour of the portrait and each part of the face as shown in FIG. 3 described later. Is stored in advance. The RAM 3 is used as a work area when the CPU 1 performs arithmetic processing, and temporarily stores data. The switch unit 4 is operated by an operator, and is operated when starting a portrait creation process and a sequence selection switch (sequence data selecting means) operated when selecting one of a plurality of portrait moving images. It has a start switch. In addition,
Each operation switch may be a single operation push switch, or a switch board composed of multiple switches,
A keyboard or the like may be used. As the switch unit 4, a mouse, a trackball, or the like may be used instead of the switch board or the like.

The display unit (display means) 5 displays a portrait made by the arithmetic processing of the CPU 1 for each process, and for example, an image signal generating circuit (Video Display Pr).
osseser: hereinafter referred to as VDP), VRAM, and TV display. The screen may be displayed on the LCD instead of the TV display as long as color display is possible. The color printer (printing means) 6 prints an image on a predetermined sheet, and can print the portrait displayed on the display unit 5 in color, for example. The CPU 1, ROM 2 and RAM 3 together constitute a closed curve creating means 21, an image transforming means 22, and a color image creating means 23. Further, the ROM 2 constitutes a deformation information storage means and a sequence data storage means.

Next, the operation will be described. Main Program FIG. 2 is a flowchart showing the main program of the portrait creation process. When this program starts, first, in step S10, initialization (initialization processing) is performed. Thereby, for example, various registers in the CPU 1 are initialized, the work area of the RAM 3 is cleared, a subroutine is initialized, and flags are reset. Next, in step S12, it is determined whether or not the sequence selection switch of the switch unit 4 is turned on. If the sequence selection switch is on, step S1
In step 4, the sequence number is changed and the process proceeds to step S16.
As a result, the sequence number corresponding to one of the plurality of portrait moving images is selected. On the other hand, if the sequence selection switch is not turned on, step S
Then, jump 14 and proceed to step S16. Therefore, at this time, the sequence number corresponding to the moving image of the same portrait is continuously selected.

Here, as shown in FIG. 3, the ROM 2 stores portrait data for creating a reference portrait and a plurality of sequence data (1) to (n) for displaying various moving images of the portrait. Has been done. The portrait data is the closed curve creation data A to F representing the contour of the portrait and each part of the face, and the closed curve color data A that specifies the color of the closed curve.
It is stored in a predetermined area by being divided into ~ F. The closed curve color data A to F are the closed curve creation data A.
Each color of ~ F (color that fills the boundary and the inside of the closed curve) is designated. Further, the plurality of sequence data (1) to (n) is data indicating in what kind of moving image a reference portrait is displayed, and is composed of a plurality of data of deformation information (1) to (n), respectively. , Are stored in a predetermined area. For example, the sequence data (1) is data for displaying a facial expression of a portrait, such as a laughed portrait, and the sequence data (2) is an angry portrait. In this case, the sequence data can be freely set and the content of the sequence data (deformation information) can be freely set depending on the expression of the portrait moving image. In step S14, for example, when the sequence number = (1) is selected, the contents of the sequence data (1) stored in the ROM 2 (transformation information (1) to
(N)) will be read.

Next, in step S16, it is determined whether or not the start switch of the switch section 4 is turned on. If the start switch is not turned on, the process returns to step S12. If the start switch is turned on, step S
In step 18, it is determined whether or not the start flag SF is [1]. The start flag SF is set in step S described later.
22. As shown in step S24, [1] and [0] are alternately repeated according to the ON operation of the start switch.
When SF = [1], a portrait moving image can be displayed. For example, SF = [0] in the first routine, and therefore the determination result of step S18 is NO at this time and the process branches to step S20. On the other hand, in the second and subsequent routines, for example, if the start switch is turned on when SF = [1], the process branches to step S22, and SF = [0] is returned to step S12. That is, a portrait moving image is displayed or the moving image is stopped by turning on the start switch.

When the process branches to step S20, the display on the display unit 5 is cleared here. Here, in the first routine, for example, the initial screen may be displayed, some operation screen may be displayed, or nothing may be displayed, but in any case, in step S20
The display is cleared once with. Further, for example, when a portrait is already displayed, the previous portrait is cleared. Next, in step S24, the start flag SF is set to [1] and the modification information address AD of the sequence data is set to the address [0]. As a result, the modification information of the selected sequence data is read from the contents of the address [0] (for example, modification information (1)). In this case, there is a relation that the modification information (1) = AD [0] address to the modification information (n) = AD [final] address. Next, in step S26, the closed curve creation data A to F and the closed curve color data A to F are read from the ROM 2 to create a portrait that is the basis of the moving image, and RA
Load on M3.

Here, the RAM 3 is provided with various work areas as shown in FIG. 4, and the data read from the ROM 2 is loaded into the corresponding area. RA
The work area of M3 will be described below. Data for creating closed curve A ... Area for storing data for creating closed curve A Data for creating closed curve B ... Area for storing data for creating closed curve B Data for creating closed curve C ... Data for creating closed curve C Area for storing closed curve D creation data: Area for storing data for creating closed curve D Closed curve E creation data: Area for storing data for creating closed curve E Closed curve F creation data: Create closed curve F Area to store data for

Closed curve A color data: Area for storing data that specifies the color (color) for filling closed curve A. Closed curve B color data: Area for storing data that specifies the color (color) for filling closed curve B. Closed curve C color Data: Area for storing data that specifies the color (color) for filling closed curve C Color data for closed curve D: Area for storing data that specifies the color (color) for filling closed curve D Color data for closed curve E ... Area that stores data that specifies the color (color) to fill Closed curve F color data ... Area that stores data that specifies the color (color) to fill closed curve F

Color status flag (A): an area for storing a flag for determining whether the closed curve A is designated (that is, filled with a color) in the color setting area. Color status flag (B): the closed curve B is colored. Area that stores a flag that determines whether to specify (that is, paint with a color) in the setting area Color status flag (C) ... Whether to specify the closed curve C in the color setting area (that is, paint with a color) Area for storing a flag for determining the color state flag (D) ... Area for storing a flag for determining whether the closed curve D is designated (that is, filled with a color) in the color setting area (E) ... The area color state flag (F that stores a flag that determines whether or not the closed curve E is designated (that is, filled with a color) in the color setting area ...... Area that stores a flag that determines whether to specify (that is, paint with a color) the closed curve F in the color setting area Background color number …… Data that specifies the color (color) to paint the background part Area to store

Sequence No. Area modification information (1) that stores the sequence number selected by the sequence selection switch Sequence No. Area for storing a part of the data (1) of the moving picture of the portrait selected by step Transformation information (2) ... Sequence No. Area for storing a part of the data (2) of the moving picture of the portrait selected by step Transformation information (3) ... Sequence NO. Area for storing a part of data (3) of the portrait moving image selected by ... Deformation information (n) ... Sequence NO. Areas for storing a part of the data (n) of the moving picture of the portrait selected by are also included areas 11 to 16 for storing closed curves corresponding to the respective parts of the created portrait. For example, area 11 is an area for storing bangs, area 12 is an area for storing hairstyles, area 13 is an area for storing glossy parts of hair, area 14 is an area for storing outlines of faces, and area 15 is parts for faces. The storage area, area 16, is an area for storing the neck.

Returning to the description of the flow chart again, after step S26, the process proceeds to the following step S28, in which closed curves A to F are created based on the loaded closed curve creation data A to F (details will be described later in a subroutine). In addition, the closed curve A created in step S30
A process of filling F (details will be described later in a subroutine) is performed. As a result, closed curves A to F corresponding to respective parts of the basic portrait are created, and the created closed curves A to F are filled with a predetermined color to finally create a portrait. . Next, a display process is performed in step S32. As a result, the created portrait is displayed on the display unit 5.
Is displayed in. After step S32, step S1
Return to 2 and repeat the same loop. In this way, the portrait that is the basis of the moving image is created and displayed.

Timer Interrupt Processing FIG. 5 is a flow chart showing the timer interrupt processing. This timer interrupt process is repeated by applying an interrupt (for example, execution at an interrupt time suitable for displaying a moving image) at regular intervals. In the timer interrupt process, first, in step S40, it is determined whether or not the start flag SF is [1].
If the start flag SF is not set to [1], it is determined that moving image display by turning on the start switch is not requested this time, and the process returns. Therefore, the moving image is not displayed at this time. If the start flag SF is [1], the process proceeds to step S41 to perform a process of creating each of the closed curves A to F (details will be described later in a subroutine). Then, in step S42, the (AD) th modification information data of the corresponding sequence data is stored in the ROM.
Read from 2. The read deformation information data is RAM3
Loaded in. Then, in step S43, the closed curve A to
Processing for interpolating F is performed according to the read deformation information. As a result, the closed curves A to F are deformed (that is, each part of the portrait is deformed corresponding to the read deformation information).

Next, the process of filling the deformed closed curves A to F in step S44 (details will be described later in a subroutine).
I do. As a result, the deformed closed curves A to F corresponding to the respective parts of the basic portrait are filled with a predetermined color. Then, in step S45, the previous display is cleared to change to the newly created display (that is, a portrait moving image). Next, in step S46, it is determined whether the modification information address AD of the sequence data is the final address. In the routine that first makes the determination in step S46,
Since the modification information address AD is not the final address, N
The process branches to O and advances to step S47 to advance the deformation information address AD by [1] and return (to the next address.) Then, the same loop is repeated. As a result, in the next routine, the (AD + 1) th deformation information data of the sequence data is read from the ROM 2 and loaded into the RAM 3, and the closed curves A to F are interpolated according to the finally read deformation information. Deform (that is, each part of the portrait is transformed corresponding to the read deformation information). After that, the same loop is repeated, and when the modified information address AD becomes equal to the final address in step S46, the process branches to YES this time and proceeds to step S48 to return the start flag SF to [0] and return. In this way, each part of the basic portrait is deformed according to the deformation information of the corresponding sequence data by the ON operation of the start switch, and as a result, the moving image of the portrait is displayed on the display unit 5. When the start switch is turned on again, the portrait moving image display is stopped.

Subroutine for Creating Closed Curves A to F FIG. 6 is a flowchart showing a subroutine for creating closed curves A to F of the main program (step S28) and to create closed curves A to F of the timer interrupt process (step S41). Is. After shifting to this subroutine, the closed curve A is created in step S50. For example, when the closed curve A corresponds to the front hair, the closed curve of the front hair (the image stored in the area 11 in FIG. 4) is created. Next, in step S52, a process of creating the closed curve B is performed. For example, when the closed curve B corresponds to the hairstyle, a closed hairstyle curve (the image stored in the area 12 in FIG. 4) is created. Then, in step S54, the closed curve C
Process to create. For example, when the closed curve C corresponds to the glossy part of the hair, the closed curve of the glossy part (see FIG.
Image stored in the area 13) is created. Next, in step S56, the process of creating the closed curve D is performed.
For example, when the closed curve D corresponds to the contour of the face, a closed contour curve (the image stored in the area 14 of FIG. 4) is created. Next, in step S58, a process of creating the closed curve E is performed. For example, when the closed curve E corresponds to a face part, a closed curve of the face part (an image stored in the area 15 of FIG. 4) is created. The face parts include eyebrows, eyes, nose, and mouth. Then, step S60
The process of creating the closed curve F is performed. For example, when the closed curve F corresponds to the neck, the closed curve of the neck (area 16 in FIG.
Image stored in) is created. After step S60, the process returns to the main program (or timer interrupt process). In this way, the closed curves A to F for creating the portrait are created.

Subroutine of Dot Calculation Processing FIGS. 7 and 8 are flowcharts showing a subroutine of dot calculation processing when the closed curves A to F are created. This processing is performed in order to display a straight line connecting the two points on the TV display of the display unit 5 (e.g., a computer screen or a television screen), and the straight line connecting the two points calculates a large number of dots. It is displayed by tying together. This technique is an application of the so-called "integer Bresenham" algorithm used for straight line drawing of computers. In the usual "Bresenham" algorithm, it is necessary to perform addition and subtraction and division of floating point to judge the inclination of the straight line and the error, but the integer calculation is used and the processing speed of the algorithm is improved by eliminating the division. Is the "integer Bresenham" algorithm. In the following description of the flow chart, the description method of "C language" used in a computer is used.
The contents of each step will be described in "C language" as necessary. The same applies to each of the flowcharts described below.

In this subroutine, first, step S1
00 is the difference between the start and end coordinates of the line (that is, the slope)
Is calculated. The starting point and the ending point of the straight line are represented by using the x coordinate and the y coordinate. For example, assume that the starting point and the ending point of the straight line are (x ₁ , y ₁ ) and (x ₂ , y ₂ ), respectively, and they are not equal to each other. All variables are integers. Then, x = x ₁ y = y ₁ Δx = x ₂ −x ₁ Δy = y ₂ −y ₁ is set, and the error term e of the algorithm is set to e = 2eΔx (initial value = −1 / 2) to proceed the calculation. ,
When the slope of the straight line is 1/2 or more, the variable y is increased to the next point separated by one unit, and when the slope of the straight line is less than 1/2, the variable y is not increased to the next point. Note that when the variable y is increased to the next point separated by one unit, “1” is subtracted from the error term e in order to re-initialize the error term e before determining the next pixel. .

Now, in step S100, the starting point of the straight line,
As the difference between the coordinates of the end points, gainx is calculated for the x coordinate and gainy is calculated for the y coordinate. In this case, the x coordinate of the starting point is fromx, the y coordinate of the starting point is fromy,
The x coordinate of the end point is represented by to.x, and the y coordinate of the end point is represented by to.y. Then, the calculation formula of the difference about the x coordinate is x at the end point.
It is represented as a value obtained by subtracting the x coordinate fromx of the starting point from the coordinate to.x, that is, (to.x-fromx). Similarly, the arithmetic expression of the difference about the y coordinate is obtained by subtracting the y coordinate fromy of the start point from the y coordinate to.y of the end point, that is, (to.y-fromy)
Expressed as Next, in step S102, the absolute value Δ (de) of the difference (that is, the slope) between the coordinates of the start point and the end point of the straight line
lta) for each of the x and y coordinates. For example, in the x coordinate, deltax = abs (gainx) is calculated. deltax corresponds to Δx = x ₂ −x ₁ . In the y coordinate, deltay = abs (gainy) is calculated. delta
y corresponds to Δy = y ₂ −y ₁ .

Then, in step S104, the starting point of the straight line,
The sign of the difference gainx of the x coordinate of the coordinates of the end point is determined. gain
When x is positive, the process proceeds to step S106 to set ratex = 1, when gainx is “0”, the process proceeds to step S108 to set ratex = 0, and when gainx is negative, the process proceeds to step S110. Set ratex = -1. After any of steps S106 to S110, the process proceeds to step S112. Similarly, in step S112, the sign of the gainy difference between the y-coordinates of the start and end coordinates of the straight line is determined. When gainy is positive, the process proceeds to step S114 to set rate = 1, when gainy is "0", the process proceeds to step S116 to set ratey = 0, and when gainy is negative, the process proceeds to step S118. ratey
Set to -1. Step S114 to Step S1
When any of 18 is passed, the process proceeds to step S120.

In step S120, the absolute value deltax of the inclination of the straight line in the x-coordinate direction and the absolute value del of the inclination of the y-coordinate direction del
It is compared with tay to determine whether deltax is larger than deltay. This is to determine which of the x-coordinate direction and the y-coordinate direction should be the main error term. delta
When x> deltay (that is, when TRUE, it corresponds to so-called YES), the process proceeds to step S122, and the main error term (main delta) is set to deltax. Although an underbar is added as (main_delta) in the drawing, this underbar is omitted because it becomes complicated in the text. Then, in step S124, the sub error term (su
b delta) to deltay. In the drawing, (sub_de
lta) is added with an underbar, but it is complicated in the text, so this underbar is omitted. Next, in step S126, the main determination flag (flag) is set to T.
Set it to RUE and proceed to step S134 in FIG. The main decision flag (flag) delt the main error term (main delta)
It indicates whether it is set to ax or deltay. On the other hand, when deltax ≦ deltay (that is, FAL
In SE, so-called NO) corresponds to step S12.
In step 8, the main error term (main delta) is set to deltay. Next, in step S130, the sub error term (sub delt
Let a) be deltax. Next, in step S132, the main determination flag (flag) is set to FALSE, and the process proceeds to step S134 in FIG.

Turning to FIG. 8, in step S134, the dot dotx [0] corresponding to the x coordinate of the starting point of the straight line is set to the x coordinate fromx of the starting point, and the dot doty [0 corresponding to the y coordinate of the starting point of the straight line is set. ] To the y coordinate fromy of the start point. As a result, each coordinate of the starting point dot is determined. Next, in step S136, similarly, the parameter px in the x coordinate direction is set to the x coordinate fromx of the starting point, and the parameter py in the y coordinate direction is set to the y coordinate fr of the starting point.
Set to omy. Next, in step S138, the error term e is obtained from the expression error = 2 × (sub delta) − (main delta). According to this formula, for example, e = 2 * Δy−Δ
It becomes like x.

Then, in step S140, the pointer i is set to the initial value [0]. The pointer i sequentially designates dots forming a straight line, and is incremented by an integer of [1]. Then, step S14
At 2, it is determined whether the pointer i is smaller than the main error term (main delta). If TRUE, step S14
In step 4, it is determined whether the error term error is error ≧ 0. If error ≧ 0, the main determination flag (flag) is determined in step S146. Main decision flag (fl
If ag) is TRUE, the x coordinate is the main error term (ma
in delta), the process proceeds to step S148, and the parameter py in the y-coordinate direction of this routine is calculated according to the following equation. py = py + ratey

Next, at step S150, the error term e of this routine is calculated according to the following equation. error = error−2 × (main delta) As a result, the error term e changes depending on the main error. Then, the process returns to step S144 again to repeat the same loop. On the other hand, if the determination result of the main determination flag (flag) is FALSE in step S146, it is determined that the y coordinate is set as the main error term (main delta), and the process branches from step S144 to step S152. , The parameter px in the x-coordinate direction of this routine is calculated according to the following equation. px = px + ratex Then, the process proceeds to step S150, and the error term e of this routine is calculated in accordance with the above-described calculation formula. Then, it returns to step S144 and repeats the same loop. And
When the error term error becomes error <0 in step S144,
Exit to step S154. Step S144-
By the process of step S152, the main error term (main d
elta) processing is performed.

Next, the processing of the sub error term (sub delta) is performed. First, in step S154, the main determination flag (flag) is determined, and the main determination flag (flag) is set to TRU.
If E, the x coordinate is the main error term (main delta)
Therefore, the process proceeds to step S156, and the parameter px in the x-coordinate direction of this routine is calculated according to the following equation. px = px + ratex Next, in step S160, the error term e of this routine is set.
Is calculated according to the following equation. error = error + 2 × (sub delta) As a result, the error term e changes depending on the sub error sub delta. Next, in step S162, the value of the parameter px of this routine in the x coordinate direction is set to the dot dotx [i + 1] corresponding to the x coordinate of the straight line. Since i = 0 in the first routine, dotx [i + 1] = dotx [1].
That is, the dot do of the x coordinate that is one dot away from the starting point
tx is calculated. Next, in step S164, the pointer i is incremented by the integer [1], and the process returns to step S142 to repeat the loop.

On the other hand, if the determination result of the main determination flag (flag) is FALSE in step S154, it is determined that the y coordinate is set as the main error term (main delta), and steps S154 to S158 are performed. And the parameter py in the y-coordinate direction of this routine is calculated according to the following equation. py = py + ratey Then, it progresses to step S160, the sub error term e of this routine is calculated according to the said arithmetic expression, Then, through step S162 and step S164, it returns to step S142 again and repeats a loop. Then, the pointer i is sequentially incremented and the above loop is repeated,
In step S142, the pointer i is set to the main error term (main d
(elta) When this is over, this subroutine ends. In this way, the straight line dot connecting the two points is calculated,
By connecting these dots, a process of displaying a straight line connecting the two points is performed. In this case, in this embodiment, since the integer calculation is mainly performed using the “integer Bresenham” algorithm, the processing speed of the algorithm is high.

Subroutine of Bezier Curve Data Calculation Processing FIGS. 9 and 10 are flowcharts showing a subroutine of data calculation processing when a Bezier curve is created. Bezier curve has two control points (P ₂ , P ₃ ),
It is defined by two anchor points (P ₁ , P ₄ ), and is generally expressed by the following equation. Here, the anchor points (P ₁ , P ₄ ) are two points at both ends of the curve, and the coordinates of each point are, for example, P ₁ (x ₁ , y ₁ ) and P ₄ (x ₄ , y ₄ ). Represented by. In addition, the control point (P ₂ ,
P ₃ ) is two points that control the bending of the curve, and the coordinates of each point are, for example, P ₂ (x ₂ , y ₂ ), P ₃ (x ₃ , y ₃ ).
It is expressed as. B (t) = (1- t) 3 · P 1 + 3t · (1-t) 2 · P 2
+ 3t ² · (1−t) · P ₃ + t ³ · P ₄ However, 0 ≦ t ≦ 1 t can be changed arbitrarily between 0 and 1. Therefore, by dividing between 0 and 1 finely, the Bezier curve The upper point is calculated with high accuracy. Since the number of calculations increases, an appropriate number of points on the curve are calculated, and then a method of interpolating them with a straight line is adopted. Note that, for example, substituting t = 0 into the above equation, P ₁ (x ₁ ,
y ₁ ), which represents one end point. Also, t = 1
Is substituted into the above equation, P ₄ (x ₄ , y ₄ ), which represents the other end point. The straight line can be displayed as a straight Bezier curve with no width.

First, the pointer i is set to [0] in step S200, and the variable t is set to t = i in step S202.
Calculate as / n. Since the pointer i is incremented for each [1], the variable t is divided by n so as to be finely divided between 0 and 1.
Then, in step S204, the deviation tn is tn = 1.0-
It is calculated by the formula t. Then, step S206
And the term to the power of 0 of the variable t of the Bezier curve B (t) (ie,
The x coordinate of t = 1) is calculated according to the following equation. sx [0] = tn × tn × tn × px [0] = (1-t) 3 · px [0] = (1-t) 3 · P 1 (x) px [0] is the anchor point P ₁ The x coordinate. Then, in step S208, the x-coordinate of the first term of the variable t (that is, t) of the Bezier curve B (t) is calculated according to the following equation. sx [1] = 3.0 × t × tn × tn × px [1] = 3t · (1-t) ² · px [1] = 3t · (1-t) ² · P ₂ (x) px [ 1] is the x coordinate of the control point P ₂ .

Next, in step S210, Bezier curve B
The x-coordinate of the variable t squared term of (t) (that is, t ² ) is calculated according to the following equation. sx [2] = 3.0 × t × t × tn × px [2] = 3t ² · (1-t) · px [2] = 3t ² · (1-t) · P ₃ (x) px [ 2] is the x coordinate of the control point P ₃ . Next, in step S212, the variable t of the Bezier curve B (t) is
The x-coordinate of the third power term (ie, t ³ ) is calculated according to the following equation. sx [3] = t × t × t × px [3] = t 3 · px [3] = t 3 · P 4 (x) px [3] is the x-coordinate of the anchor point P _4.

Then, steps S214 to S2
At 00, the same calculation as above is performed for the y coordinate of the variable t of the Bezier curve B (t). That is, step S2
In step 14, the y coordinate of the 0th power of the variable t of the Bezier curve B (t) (that is, t = 1) is calculated according to the following equation. sy [0] = tn × tn × tn × py [0] = (1-t) 3 · py [0] = (1-t) 3 · P 1 (y) py [0] is the anchor point P ₁ It is the y coordinate. Next, in step S216, the y coordinate of the first term of the variable t of the Bezier curve B (t) (that is, t) is calculated according to the following equation. sy [1] = 3.0 × t × tn × tn × py [1] = 3t · (1-t) ² · py [1] = 3t · (1-t) ² · P ₂ (y) py [ 1] is the y coordinate of the control point P ₂ .

Then, in step S218, Bezier curve B
The y-coordinate of the variable t squared of (t) (that is, t ² ) is calculated according to the following equation. sy [2] = 3.0 × t × t × tn × py [2] = 3t ² · (1-t) · py [2] = 3t ² · (1-t) · P ₃ (y) py [ 2] is the y coordinate of the control point P ₃ . Next, in step S220, the variable t of the Bezier curve B (t) is
The y-coordinate of the third power of the term (that is, t ³ ) is calculated according to the following equation. sy [3] = t × t × t × py [3] = t 3 · py [3] = t 3 · P 4 (y) py [3] is the y coordinate of the anchor point P _4.

Turning to FIG. 10, the x-coordinate bx [i] and the y-coordinate b of the interpolation constants for interpolating points on the Bezier curve B (t) with a straight line in step S222.
Initially, both y [i] are set to bx [i] = by [i] = 0. Next, in step S224, the pointer j is set to the initial value [0], and in the following step S226, the x-coordinate bx [i] and the y-coordinate by [i] of the interpolation constant are calculated according to the following equations. bx [i] = bx [i] + sx [j] by [i] = by [i] + sy [j] Then, in step S228, it is determined whether the pointer j is equal to or more than [4], and [4] If less than, step S
The routine proceeds to 230, the pointer j is incremented by [1], the routine returns to step S226, and the loop is repeated. Then, when the pointer j becomes [4] or more in step S228, the process exits to step S232.

Next, at step S232, the pointer i
If it is less than n, it is judged whether it is more than n.
In step S234, the pointer i is incremented by [1].
And returns to step S202 in FIG. 8 to perform a similar loop.
repeat. Then, in step S232, the pointer i
When it becomes n or more, this subroutine is finished. like this
And increment pointer i from [0] to n
Two control points (P₂,
P₃) And two anchor points (P 1, P_Four) By
The point on the Bezier curve B (t) defined by
When t is divided into 0 to 1 and finely changed
Therefore, the points on the Bezier curve can be calculated accurately.
It Also, points and points on the Bezier curve B (t)
By properly interpolating between the four points, the accuracy can be further improved.
To be enhanced.

Color Setting Subroutine FIG. 11 and FIG. 12 show step S in the main program.
24 is a flowchart showing a subroutine of color setting processing in the filling processing of step S44 in the filling processing of 24 and the timer interrupt processing.
This process is to paint a specified color on the created closed curve. In this subroutine, the color area is first cleared in step S300. As a result, at first, all areas to be colored are once cleared and become colorless. Next, in step S302, the color state flag Cflag is initialized. The color state flag Cflag is a flag that determines whether or not the corresponding closed curve is designated (that is, filled with a color) in the color setting area, and when initialized, as shown in FIG. 1]. The color status flag = [− 1] is to set a color that cannot be a color number that specifies a color.

In the example of FIG. 13, the colors of the object 1 to the object n (each object corresponds to each part of a portrait, for example) are all set to the initial state, and only the background color has some color number. Comes in and is painted in color. Then, every time the color switching flag of each object becomes [1], the corresponding color state flag is inverted (for example, [-1]
To [1]). Step S
At 304, the pointer i is cleared to [0], and step S
At 306, the pointer j is cleared to [0]. Pointer i
Is to sequentially specify lines on the screen (for example, there are lines from 0 to 524), and the pointer j is to sequentially specify dots on the line. Pointer i =
Line 0 is designated by setting [0], and dot 0 on line 0 is designated by setting pointer j = [0].

Then, in step S308, the line color is set.
The lcolor is set to [-1] and the line number (line number) lnum is set to [-1]. The line color lcolor specifies the color of the closed curve. The line number lnum designates the boundary line number of the closed curve. Both are set to the initial state by being set to [-1]. Next, in step S310, the closed curve number k is set to (n-1). For example, if there are six closed curves, the closed curve number k is set to [5]. The smaller the closed curve number k, the higher the priority. Therefore, k = 0 has the highest priority, and subsequently becomes lower. As a result, step S3
In the case of 10, the state is set to the second lowest priority state.

Then, in step S312, the closed curve function C
Judge [k] [i] [j]. Closed curve function C [k]
[I] and [j] can determine the priority of the closed curve number k and whether or not the closed curve this time is in any closed curve (that is, which closed curve boundary line is detected). Is. When the determination result of the closed curve function C [k] [i] [j] is TRUE, the process proceeds to step S314 and the color state flag Cflag [k] is inverted. As a result, the color state flag of the closed curve specified by the closed curve number k this time is inverted from [-1] to [1]. Therefore, the corresponding closed curve is colored. Next, in step S316, the color number is entered in the line color lcolor, and the closed curve number k is entered in the line number lnum. As a result, the color number (which specifies the color to be painted) is given to the closed curve this time, and the priority of the closed curve is given. Next, in step S318, the closed curve number k is decremented by [1]. This gives only one higher priority.

On the other hand, in step S312, the closed curve function C
When the judgment result of [k] [i] [j] is FALSE,
The process jumps from step S314 and step S316 and proceeds to step S318. Therefore, at this time, the closed curve number k is immediately decremented by [1], and the priority is moved up by one. After step S318, it is determined in the following step S320 whether the closed curve number k is less than [0] (that is, it is in a state where it has not yet become the highest priority). If k = 0 (priority is not highest in the first routine), the process branches to TRUE and returns to step S312 to repeat the loop. By repeating the loop, the closed curve number k sequentially increases from 5 to 4, 3, 2, 1, and when k = 0, it is determined that the highest priority is given, and the process proceeds to step S322 in FIG. Get out. In this way, for each closed curve, the color state flag Cflag
By reversing [k], the permission to paint the color is given, the color to be painted is designated by the color number, and the priority of the closed curve is given.

Turning to FIG. 12, the process of processing dot by dot is executed for the line of the closed curve. First, in step S322, the closed curve number k is set to (n). For example, if there are six closed curves, the closed curve number k is set to [6]. The smaller the closed curve number k, the higher the priority. Therefore, step S3
In 22, the state has the lowest priority. Next, in step S324, the color status flag Cfl
It is determined whether ag [k] is not equal to [−1], that is, whether the closed curve is filled with color. Note that this determination is based on Cflag
[K]! The sign "!" Represents "not" in "C language". Step S3
The determination of 24 means that the color state flag Cflag [k] is determined from the closed curve of low priority.

The color status flag Cflag [k] is [-1].
If not equal to (TRUE), step S3
Go to step 26 and put the closed curve number k in the color buffer Cbuf.
Put in. As a result, the closed curve of the closed curve number k designated this time is colored. Then
In step S328, the closed curve number k is decremented by [1]. This gives only one higher priority. Next, in step S330, it is determined whether or not the closed curve number k is less than [0] (that is, in a state in which the highest priority is not yet reached). If k = 0 (priority is not the highest in the first routine), the process branches to TRUE and returns to step S324 to repeat the loop. By repeating the loop, the closed curve number k sequentially increases from 6 to 4, 3, 2, 1, and when k = 0, it is determined that the highest priority has been reached, and the process proceeds to step S332. On the other hand, if the color state flag Cflag [k] is equal to [−1] (FALSE) in step S324, step S326.
And jump to step S328. Therefore,
At this time, the closed curve number k is not put in the color buffer Cbuf, and the color is not painted on the closed curve of the closed curve number k designated this time. In this way
The closed curve number k is put into the color buffer Cbuf from the closed curve having the lower priority.

In step S332, the line color lcolor
Is not equal to [−1], that is, whether the color of the closed curve is designated. Note that this determination is lcolor! = -1. If the line color lcolor is not equal to [−1] (if the color of the closed curve is specified), the process proceeds to step S334 and the line number (line number) lnum.
Is smaller than the color buffer Cbuf.
Since the color of the curve having a high priority is stored in the color buffer Cbuf, if the line number (line number) lnum is smaller than the color buffer Cbuf, the process branches to TRUE and proceeds to step S336, and is designated by the pointers i and j. The line and dot colors color [i] [j] (hereinafter referred to as pointer designated colors) are set to the line color lcolor with high priority, and then the process proceeds to step S338. As a result, the color of the higher priority curve is painted on the closed curve.

On the other hand, in step S332, the line color lc
If olor is equal to [−1] (the color of the closed curve is not designated), the process branches to step S340 and the pointer designated color color [i] [j] is set in the color state flag Cflag [Cbuf]. Then step S338
Proceed to. The color state flag Cflag [Cbuf] is a flag indicating a state in which the background color is applied.
Therefore, in this case, the closed curve is painted with the background color. Further, in step S334, the line number (line number) lnum is the color buffer C.
If buf or more, branch to FALSE and step S
In step 340, the pointer designation color color [i] is similarly set.
[J] is set to the color status flag Cflag [Cbuf], and the process proceeds to step S338. Therefore, also in this case, the background color is applied to the closed curve.
In this way, the line and dot designated by the pointers i and j this time are painted with the color of the closed curve having a high priority or the color of the background (background).

Then, in step S338, the pointer j is incremented by [1] to move to the next dot.
Next, in step S342, the pointer j after increment is ndot (for example, ndot = 25 at the maximum value of one line).
(6 dots) is determined, and if not reached, the process returns to step S308 in FIG. 11 and the same routine is repeated. As a result, in the next routine, similar processing is performed on the next dot on the same line. Then, when the pointer j reaches ndot (for example, 256 dots) in step S342, it is determined that the color setting process has been completed for all the dots in one line, and the process exits to step S344. In step S344, the pointer i is incremented by [1] to move to the next line. Next, in step S346, the pointer i after being incremented is nline (for example, the maximum value of one screen is nline.
= 525 lines) is determined, and if not reached, the process returns to step S306 in FIG. 11 and the pointer j for designating the dot is returned to [0] and the same routine is repeated. As a result, in the next routine, the same process is performed by shifting to the next line. Then, when the pointer i reaches nline (for example, 525 lines) in step S346, it is determined that the color setting process has been completed for all lines, and this routine is ended.

In this way, one screen is divided into lines and dots and color setting processing is performed, and all closed curves drawn at this time (for example, closed curves corresponding to each part of a portrait)
After finishing the flag setting for, the color number is set in the color setting area with reference to the flags of all the closed curves. In this case, when the boundary line of any closed curve is detected, if the boundary line is inside a closed curve having a higher priority than the closed curve to which it belongs, a high priority internal color is set in the color setting area. The color status flag is controlled so that Therefore, it is always filled with the color of the closed curve having a high priority, and the portrait is created with a simple data structure. Specifically, for example, as shown in FIG. 4, a closed curve corresponding to each part (forelock, hairstyle, hair gloss, face outline, face part, neck) of a caricature is created, and a color setting area for each part is created. When the color number is set to, and is displayed on the screen, the closed curve is filled with the color of each part having a high priority. As a result, a portrait completion screen is displayed on the display unit 5.

Explanation of Principle of Graphic Transformation Process Next, the principle of the graphic transformation process used for creating different facial expressions for the same person with a small amount of data will be described. This principle is applied when transforming the facial expression of a portrait according to the transform information of the sequence data described above. Here, as shown in FIG. 14, for example, a rectangle 1
The principle of dividing 00 into a plurality of triangles and deforming the figure according to the deformation of the triangle will be described. A method of dividing a triangle in the rectangle 100 is as shown in FIG. 14, and the triangle is divided into eight triangles to. Split so that First, as a transformation condition, the number of data is reduced by not moving the center point TP of the rectangle 100 as the division center, and the position transformation 1 as shown in FIG. 15 is applied to the transformation of the right triangle. Right triangle ...
Position transformation 2 as shown in FIG. 16 is applied to the modification of FIG.

A. Position Transformation 1 by Framing Right Angle Triangle FIG. 15 shows an example of position transformation 1 by framing a right triangle, in which a right triangle 101 having T ₁ , T ₂ , and T ₃ as vertices is replaced by a vertex T ₃ . Don't change, T ₁ ',
It is transformed into a triangle 102 having T ₂ 'and T ₃ as vertices. At this time, in the position conversion from the right triangle 101 to the triangle 102, the calculation described below is performed. The x-coordinate of a specific point P inside the right triangle 101 is represented by V ₁ x = V ₂ x = Px when corresponding to the points V ₁ and V ₂ on each side. In addition, the relative position v between the specific point P and the vertex T ₁
The relation of r is vr = V ₁ x / T ₁ x = Px / T ₁ x. Here, the y coordinate of the point V ₁ on the side is V ₁ y = 0. On the other hand, when the points V ₁ and V ₂ on each side of the right triangle 101 are made to correspond to the relation of the relative position vr between the specific point P and the vertex T ₂ , it is expressed by V ₂ y = T ₂ y · vr. Therefore, tables in the relative position pr of the particular point _{P, pr = (V 2 y-} V 1 y) / (Py-V 1 y) = V 2 y / Py = T 2 y · Px / T 2 x · Py To be done. Further, the coordinates of the points V ₁ ′ and V ₂ ′ on each side of the deformed triangle 102 can be expressed by the following equation. _{V 1 'x = T 1'} x · vr V 1 'y = T 1' y · vr V 2 'x = T 2' x · vr V 2 'y = T 2' y · vr

From the above data, the transformed triangle 102
The x coordinate of the specific point P ′ can be calculated by the following arithmetic processing. First, since the following relationship holds that _{(V 2 'x-V 1} ' x) :( P'x-V 1 'x) = (V 2 x-V 1 x) :( Px-V 1 x), (V _{2 'x-V 1' x} ) / (P'x-V 1 'x) = relationship pr is established. Therefore, when the x coordinate of the specific point P ′ is obtained using this pr, it becomes as follows.
_{P'x = {(V 2 'x} -V 1' x) / Pr} + V 1 'x =
_{(T 2 'x-T 1} ' x) × (vr / Pr) + T 1 'x · v
_{r = (T 2 'x-} T 1' x) × {(T 1 x · Py) / (T 2
y · Px)} × (Px / T 1 x) + T 1 'x · (Px / T
_{1 x) = (T 1 '} x / T 1 x) · Px + {(T 2' x-
T ₁ 'x) / T ₂ y} · Py

Similarly, the y coordinate of the specific point P'of the deformed triangle 102 can be calculated by the following arithmetic processing.
First, since the following relationship holds that _{(V 2 'y-V 1} ' y) :( P'y-V 1 'y) = (V 2 y-V 1 y) :( Py-V 1 y), (V _{2 'y-V 1' y} ) / (P'y-V 1 'y) = relationship pr is established. Therefore, when the y coordinate of the specific point P ′ is obtained using this pr, it becomes as follows.
_{P'y = {(V 2 'y} -V 1' y) / Pr} + V 1 'y =
_{(T 2 'y-T 1} ' y) × (vr / Pr) + T 1 'y · v
r = (T ₂ 'y-T ₁ ' y) × {(T ₁ x · Py) / (T ₂
y · Px)} × (Px / T 1 x) + T 1 'y · (Px / T
_{1 x) = (T 1 '} y / T 1 x) · Px + {(T 2' y-
T ₁ 'y) / T ₂ y} Py

As described above, the right triangle 101 having T ₁ , T ₂ and T ₃ as vertices is not changed in position of the vertex T ₃
When transforming into a triangle 102 having T ₁ ′, T ₂ ′ and T ₃ as vertices, x of a specific point P inside the right triangle 101 before transformation
The coordinates and the y-coordinate can be calculated as the deformed specific point P ′ in association with the points V ₁ and V ₂ on each side. In other words, this means that when the specific point P is applied to each undeformed dot forming the portrait, each deformed dot can be obtained as the specific point P ′. Moreover, since the position of the vertex T ₃ is not changed at this time, the data required for the deformation can be reduced. Therefore, by such a transformation process, it becomes possible to change the facial expression of the portrait while ensuring the sameness as the same person.

B. Those shown in position conversion 2 16 by framing of a right triangle is an example of position transformation 2 by framing of a right triangle, a right triangle 111 whose vertices T _1, T _2, T _3, the position of the vertex T ₃ Don't change, T ₁ ',
It is transformed into a triangle 112 having T ₂ 'and T ₃ as vertices. At this time, in the position conversion from the right triangle 111 to the triangle 112, the following calculation is performed. The y coordinate of a specific point P inside the right triangle 111 is represented by V ₁ y = V ₂ y = Py when corresponding to the points V ₁ and V ₂ on each side. In addition, the relative position v between the specific point P and the vertex T ₁
The relation of r is vr = V ₁ y / T ₁ y = Py / T ₁ y.

Here, the x coordinate of the point V ₁ on the side is V ₁ x = 0
Is. On the other hand, the points V ₁ and V ₂ on each side of the right triangle 111
Is related to the relationship of the relative position vr between the specific point P and the apex T ₂ , it is expressed by V ₂ x = T ₂ x · vr. Therefore, tables in the relative position pr of the particular point _{P, pr = (V 2 x-} V 1 x) / (Px-V 1 x) = V 2 x / Px = T 2 x · Py / T 1 y · Px To be done. Further, the coordinates of the points V ₁ 'and V ₂ ' on each side of the deformed triangle 112 can be expressed by the following equation. _{V 1 'x = T 1'} x · vr V 1 'y = T 1' y · vr V 2 'x = T 2' x · vr V 2 'y = T 2' y · vr

From the above data, the deformed triangle 112
The x coordinate of the specific point P ′ can be calculated by the following arithmetic processing. First, since the following relationship holds that _{(V 2 'x-V 1} ' x) :( P'x-V 1 'x) = (V 2 x-V 1 x) :( Px-V 1 x), (V _{2 'x-V 1' x} ) / (P'x-V 1 'x) = (V 2 x-V 1 x) / (Px-V 1 x) = established relationship V ₂ x / Px = pr To do. Therefore, when the x coordinate of the specific point P ′ is obtained using this pr, it becomes as follows. _{P'x = {(V 2 'x} -V 1' x) / Pr} + V 1 'x = (T 2' x-T 1 'x) × (vr / Pr) + T 1' x · vr = (T _{2 'x-T 1' x} ) × {(T 1 y · Px) / (T 2 x · Py)} × (Py / T 1 y) + T 1 'x · (Py / T 1 y) = (T _{2 'x-T 1' x} ) / T 2 x) · Px + (T 1 'x / T 1 y) · Py

Similarly, the y coordinate of the specific point P'of the deformed triangle 112 can be calculated by the following arithmetic processing. 'Because relationship (y = pr is satisfied, therefore, the particular point P using the _{pr (V 2' y-V} 1 'y) / P'y-V 1)' When determining the y-coordinate of the following become that way. _{P'y = {(V 2 'y} -V 1' y) / Pr} + V 1 'y = (T 2' y-T 1 'y) × (vr / Pr) + T 1' y · vr = (T _{2 'y-T 1' y} ) × {(T 1 y · Px) / (T 2 x · Py)} × (Py / T 1 y) + T 1 'y · (Py / T 1 y) = (T _{2 'y-T 1' y} ) / T 2 x) · Px + (T 1 'y / T 1 y) · Py

As described above, the right triangle 111 having the vertices of T ₁ , T ₂ and T ₃ does not change the position of the vertex T ₃ ,
When transforming into a triangle 112 having T ₁ ′, T ₂ ′ and T ₃ as vertices, x of a specific point P inside the right triangle 111 before transformation
The coordinates and the y-coordinate can be calculated as the deformed specific point P ′ in association with the points V ₁ and V ₂ on each side. In other words, this means that when the specific point P is applied to each undeformed dot forming the portrait, each deformed dot can be obtained as the specific point P ′. Moreover, since the position of the vertex T ₃ is not changed at this time, the data required for the deformation can be reduced. Therefore, by such a transformation process, it becomes possible to change the facial expression of the portrait while ensuring the sameness as the same person.

As described above, the inside of the rectangle 100 is divided into a plurality of (eight in this embodiment) right-angled triangles ~.
By deforming the figure according to the deformation of the triangle, the center point T of the rectangle 100 is obtained as data necessary for the deformation.
It is possible to easily change the facial expression of the portrait by using P (data of the center of division), data of the part to be deformed (data of the part), original data of the rectangle 100, and less data that the data after the deformation is sufficient. In this embodiment, a plurality of portraits using such a transformation method are prepared by executing the above-described programs, and the facial expressions are interpolated to display a moving image. Specifically, the main program changes the sequence number to create a basic portrait. Then, in the timer interrupt process, a moving image in which the facial expression of the portrait is changed is created and displayed according to the deformation information of the corresponding sequence number.

FIG. 17 is a diagram showing an example of a case where a moving image is created by changing the facial expression of a portrait. As shown in FIG. 17, for example, portraits 202 to 206 obtained by transforming the portrait 201 according to the modification information of the sequence number with respect to the basic portrait 201 are created by the timer interrupt process, and are sequentially displayed as moving images on the display unit 5. This makes it possible to realize a moving image in which the facial expression of the basic portrait 201 is variously changed. In this case, it is not necessary to have all or part of the data of the portrait for the number of screens for moving the image as in the conventional case, but it is only necessary to have the closed curve data, the color data and the sequence data (a plurality of deformation information) for changing the facial expression. It becomes possible to display the video in color. As a result, it is possible to display a moving image of a portrait in color at low cost while reducing the amount of data, reducing the capacity of the memory device. Further, the same effect as that of the present embodiment can be obtained when a moving image of an animation image is displayed in color instead of a portrait. Note that, for example, how many images are used to interpolate each facial expression in the example of FIG.
Must be specified in advance. As an interpolation method, for example, a process of interpolating between the data may be performed after performing the deformation from the current deformation information and the next deformation information. By doing so, the facial expression can be changed more finely with less data.

Next, FIG. 18 shows the second embodiment of the present invention.
~ It demonstrates with reference to FIG. In the present embodiment, a plurality of pieces of sequence data (small sequence data) each including a combination of a small number of pieces of deformation information are prepared, and necessary small sequence data are selected from the plurality of pieces of sequence data and combined to make the whole piece of deformation information. The hardware block configuration of this embodiment is similar to that shown in FIG. 1 of the previous embodiment, except that the switch unit 4 includes a sequence set mode switch and a selection switch. The sequence set mode switch is operated when selecting a mode for selecting and combining a plurality of small sequence data, and the selection switch is operated when selecting a desired one from a plurality of small sequence data. Is. In addition, ROM
Reference numeral 2 constitutes deformation information storage means and small sequence data storage means. The different program parts will be described below. Main Program FIG. 18 is a flowchart showing the main program of the portrait creation process. When this program starts,
First, in step S400, initialization (initialization processing)
I do. As a result, for example, various registers in the CPU 1 are initialized, the work area of the RAM 3 is cleared,
Subroutine initialization, flag resetting, etc. are performed.

Then, in step S402, it is determined whether or not the sequence set mode switch is turned on. A. In the sequence set mode If the sequence set mode switch is on, the small sequence data (1) to (n) are displayed in step S404. If the sequence set mode switch is not turned on, the process branches to step S416 described below. Here, the ROM 2 stores portrait data for creating a reference portrait and a plurality of small sequence data (1) to (n) for displaying various moving images of the portrait as shown in FIG. There is. The caricature data is closed curve creation data A to represent the contour of the caricature and each part of the face.
F and closed curve color data A to F that specify the color of the closed curve
And are stored in a predetermined area. In addition,
The closed curve color data A to F are the closed curve creation data A to F.
Each of the colors (the color that fills the boundary and the inside of the closed curve) is specified.

Further, a plurality of small sequence data (1) to
(N) is data indicating in what kind of moving image a reference portrait is displayed, and in particular, in this embodiment, each is composed of a small amount of deformation information of three and is stored in a predetermined area. For example, small sequence data (1)
Is transformation information (A), (B), (C), small sequence data (2) is transformation information (D), (E), (F), and small sequence data (3) is transformation information ( G),
(H) and (I). Similarly, any small sequence data is composed of three pieces of modification information, and the last small sequence data (n) is composed of modification information (α), (β), and (γ). ing. Such small sequence data (1)-
Various combinations of (n) serve as data for displaying the facial expressions of a portrait such as a laughed portrait or an angry portrait in a moving image. In this case, the small sequence data can be freely combined and the content (deformation information) of the small sequence data can be freely set depending on the expression of the portrait moving image. .

After returning to the program again and entering the sequence set mode to display the small sequence data (1) to (n) in step S404, it is determined in step S406 whether or not the selection switch is turned on. If the selection switch is not turned on, the process returns to step S402. If the selection switch is on, step S408
From (AD1) of the sequence data area of RAM3
The selected small sequence data is stored in the address {(AD1) +2}. Here, the RAM 3 is provided with various work areas as shown in FIG. 20, and the data read from the ROM 2 is loaded into the corresponding area. The work area of the RAM 3 will be described below. The areas for storing the closed curve A creation data to the closed curve F creation data, the closed curve A color data to the closed curve F color data, the color status flag (A) to the color status flag (F), and the background color number are the same as in the above embodiment. is there.

In addition, three small sequence data storage areas are used as a unit (for example, (AD1) to {(AD
1) +2} are specified). And, in the storage area of each small sequence data,
Three pieces of deformation information corresponding to the small sequence data selected by the selection switch are stored. Further, there are areas 11 to 16 that store closed curves corresponding to the respective parts of the created portrait, for example, area 1
Reference numeral 1 is an area for storing bangs, area 12 is an area for storing hairstyles, area 13 is an area for storing glossy parts of hair, area 14 is an area for storing facial contours, and area 15 is an area for storing facial parts. , Area 16 is an area for storing the neck.

Returning to the explanation of the flow chart again, in step S408, (A in the sequence data area of the RAM 3 is
After storing the selected small sequence data in the addresses D1) to {(AD1) +2}, whether or not the sequence data area (AD1) +2 is equal to n (the number of the selected small sequence data) in the subsequent step S410. To determine.
For example, when three small sequence data are selected, n = 3, and when (AD1) +2 is equal to 3, YE
Branch to S, and if not equal, branch to NO. If NO in step S410, the flow advances to step S412 to increment the address pointer (AD1) of the sequence data area to (AD1) +3, and then step S402.
Return to. As a result, the address pointer (AD1) is increased by 3 all at once, which is because the next small sequence data is entered. Then, when the sequence data area (AD1) +2 becomes equal to n in step S410 in the loop after the next time (for example, when the deformation information of the three small sequence data is completely stored), the process goes to step S414 and the sequence data area The address pointer (AD1) is returned to [0] again, and the process returns to step S402. In this way, the small sequence data of the number selected by the selection switch and its contents (deformation information) are sequentially stored in the RAM 3, and the information for deforming the portrait is prepared.

B. When not in the sequence set mode If the sequence set mode switch is not on, the determination result of step S402 is NO, and the process branches to step S416. In step S416, the address (AD1) of the sequence data area is returned to [0].
As a result, the preparation for selecting the small sequence data next time is prepared. Next, in step S418, it is determined whether or not the start switch of the switch unit 4 has been turned on. If the start switch is not on, step S40
Return to 2. C. When the start switch is turned on When the start switch is turned on, step S42
The process proceeds to 0 and it is determined whether or not the start flag SF is [1]. The start flag SF alternately repeats [1] and [0] according to the ON operation of the start switch, and when SF = [1], a portrait moving image can be displayed.

For example, in the routine that first goes through step S420, SF = [0]. Therefore, at this time, the determination result of step S420 is NO and the process branches to step S422. On the other hand, in the routine after the second time,
For example, if the start switch is turned on when SF = [1], the process branches to step S424, SF = [0], and the process returns to step S402. That is, a portrait moving image is displayed or the moving image is stopped by turning on the start switch.

When the process branches to step S422, the display on the display unit 5 is cleared here. Then, step S
At 426, the start flag SF is set to [1] and the modification information address AD of the sequence data is set to the address [0]. As a result, the modification information of the selected small sequence data is read from the contents of the address [0] (for example, when the small sequence data (1) is selected, the modification information (A)). In this case, modification information (A) = address AD [0] to modification information (C)
= AD [final] address. Next, in step S428, the closed curve creation data A to F and the closed curve color data A to F are read from the ROM 2 and loaded into the RAM 3 in order to create a portrait that is the basis of the moving image.

Next, in step S430, the closed curves A to F are created based on the loaded closed curve creation data A to F (subroutine contents are the same), and the created closed curves A to F are created in step S432.
Is performed (the contents of the subroutine are the same). As a result, closed curves A to F corresponding to respective parts of the basic portrait are created, and the created closed curves A to F are filled with a predetermined color to finally create a portrait. . Then, a display process is performed in step S434. As a result, the created portrait is displayed on the display unit 5. After step S434, the process returns to step S402 and the same loop is repeated. In this way, the portrait that is the basis of the moving image is created and displayed.

In addition, the timer interrupt process is repeated by interrupting at regular intervals. As a result, a portrait moving image is displayed. In this case, in the second embodiment, small sequence data (1)
~ (N) are prepared in advance, and at least one desired small sequence data is selected from them to be combined into the whole deformation information, and each part of the basic caricature is deformed according to the deformation information, and the result is obtained. A moving image of a portrait is displayed on the display unit 5. Therefore, there is an advantage that the operator can freely set the contents of the deformation and the degree of freedom in displaying the moving image is increased.

Next, FIG. 21 shows the third embodiment of the present invention.
~ It demonstrates with reference to FIG. In this embodiment, the operator himself inputs transformation information for transforming a portrait. The hardware block configuration of this embodiment is the same as that shown in FIG. 1 of the previous embodiment, except that the switch unit 4 includes a sequence set mode switch, an input switch and a step switch. The sequence set mode switch is operated when selecting a mode for inputting deformation information, and the input switch (deformation information creating means) is for inputting deformation information for deforming a portrait, and further advances. The switch advances the input of deformation information to the next (that is, turns on when inputting the next deformation information). The ROM 2 constitutes a deformation information storage means. The different program parts will be described below. Main Program FIG. 21 is a flowchart showing the main program of the portrait creation process. When this program starts,
First, in step S500, initialization (initialization processing)
I do. As a result, for example, various registers in the CPU 1 are initialized, the work area of the RAM 3 is cleared,
Subroutine initialization, flag resetting, etc. are performed.

Then, in step S502, it is determined whether or not the sequence set mode switch is turned on. A. In the sequence set mode If the sequence set mode switch is on, the process proceeds to step S504, and it is determined whether or not the input switch is operated. Here, the ROM 2 stores caricature data for creating a reference caricature as shown in FIG. 22, and the caricature data includes closed curve creation data A to F representing the contour of the caricature and each part of the face. It is divided into closed curve color data A to F for designating the color of the closed curve and stored in a predetermined area. The closed curve color data A to
F designates each color of the closed curve creation data A to F (color for filling the boundary and the inside of the closed curve).

Returning to the program again, when the input switch is operated by the switch 504, step S50
In step 6, the deformation information input to the address (AD1) of the sequence data area of the RAM 3 is stored. In this case, the input switch can input information for transforming the portrait, and various transformations such as transforming the shape of the part or changing the position of the part in order to change the facial expression, for example. Information is input by a switch operation.
As a result, even if the ROM 2 does not have the deformation information in advance, it is possible to create the deformation information as desired by the operator and specify the deformation order. The input switch may be a keyboard or a mouse-like type. Here, the RAM 3 is provided with various work areas as shown in FIG. 23, and the data read from the ROM 2 is loaded into the corresponding area. The work area of the RAM 3 will be described below.
The areas for storing the closed curve A creation data to the closed curve F creation data, the closed curve A color data to the closed curve F color data, the color status flag (A) to the color status flag (F), and the background color number are the same as in the above embodiment. is there.

In addition, a plurality of areas for storing deformation information input by operating the input switch are arranged, and these areas are sequence data. Further, the address of the area for storing the respective deformation information is designated by the address pointer (AD1). Further, areas 11 to 1 storing closed curves corresponding to the respective parts of the created portrait.
There are six areas, for example, area 11 is an area for storing bangs, area 12 is an area for storing hairstyles, area 13 is an area for storing glossy parts of hair, area 14 is an area for storing the outline of a face, and area 15 is Area 16 is an area for storing face parts, and area 16 is an area for storing a neck.

Returning to the explanation of the flow chart again, in step S506, (A of the sequence data area of the RAM 3 is
After the deformation information input to the address D1) is stored, it is determined in step S508 whether the step switch is on. If the step switch is not turned on, the process returns to step S502 and the same loop is repeated. When the step switch is turned on, the input of this deformation information is completed,
It is determined that the next deformation information is input, and step S51 is performed.
The process proceeds to 0, and the address pointer (AD1) is incremented by [1] (becomes AD1 + 1). Next, in step S512, it is determined whether or not the address pointer (AD1) has reached (final address + 1). If AD1 = (final address + 1), the process returns to step S502 and the same loop is repeated. Therefore, in the subsequent loops, a plurality of pieces of deformation information are sequentially input, and the addresses (AD1 + 1) and (AD1 +) in the sequence data area shown in FIG.
2) The address is stored ... Then, in step S512, the address pointer (AD1)
Becomes equal to (final address + 1), step S5
After returning to step 14, the address pointer (AD1) is returned to [0] and the process returns to step S502. In this way, a plurality of pieces of transformation information are input, and the preparation of data for transforming a portrait is completed.

B. When not in the sequence set mode If the sequence set mode switch is not turned on, the determination result of step S502 is NO, and the process branches to step S516. In step S516, an address pointer (AD1) for designating the sequence data area
Is returned to [0]. As a result, the preparation for selecting the deformation information next time is prepared. Next, in step S518, it is determined whether or not the start switch of the switch unit 4 has been turned on. If the start switch is not turned on, the process returns to step S502. C. When the start switch is turned on When the start switch is turned on, step S52
The process proceeds to 0 and it is determined whether or not the start flag SF is [1]. The start flag SF alternately repeats [1] and [0] according to the ON operation of the start switch, and when SF = [1], a portrait moving image can be displayed. For example, SF = [0] in the routine that first passes through step S520, and therefore the determination result of step S520 is NO at this time and step S520
It branches to 522. On the other hand, in the second and subsequent routines, if the start switch is turned on when SF = [1], for example, the process branches to step S526, and SF = [0] is returned to step S502. That is, a portrait moving image is displayed or the moving image is stopped by turning on the start switch.

When the process branches to step S522, the display on the display unit 5 is cleared here. Then, step S
At 524, the start flag SF is set to [1] and the modification information address AD of the sequence data is set to the address [0]. As a result, the modification information of the sequence data is read from the contents of the address [0]. Next, in step S528, closed curve creation data A to F are created in order to create a portrait that is the basis of the moving image.
And the closed curve color data A to F are read from the ROM 2 and loaded into the RAM 3.

Next, in step S530, a process of creating closed curves A to F based on the loaded closed curve creation data A to F (subroutine contents are the same) is performed, and in step S532 the created closed curves A to F are created.
Is performed (the contents of the subroutine are the same). As a result, closed curves A to F corresponding to respective parts of the basic portrait are created, and the created closed curves A to F are filled with a predetermined color to finally create a portrait. . Next, a display process is performed in step S534. As a result, the created portrait is displayed on the display unit 5. After step S534, the process returns to step S502 and the same loop is repeated. In this way, the portrait that is the basis of the moving image is created and displayed.

In addition, the timer interrupt process is repeated by interrupting at regular intervals. As a result, a portrait moving image is displayed. In this case, in the third embodiment, the operator can freely input and set the deformation information as sequence data, and each part of the basic caricature is deformed according to the deformation information, and as a result, the caricature is The moving image is displayed on the display unit 5. Therefore, there is an advantage that the operator can freely set the contents of the deformation and the degree of freedom in displaying the moving image is increased.

In each of the above embodiments, the Bezier curve parameter is used as a curve parameter when creating a closed curve, but the present invention is not limited to this, and an arbitrary curve such as a B-spline curve can also be used. is there. In addition to these, parameters such as parabola, hyperbola, trigonometric function and the like can be used as appropriate. In this case, an appropriate mathematical formula such as a parabola, a hyperbola, a trigonometric function or the like may be used depending on the shape of the closed curve to be created. By doing this, it is possible to create an appropriate closed curve depending on the occasion.
Further, the color image to be displayed is not limited to a portrait or an animation image, but may be various images used in a game or the like, a character, or background data. Furthermore, the present invention is not limited to the case where a moving image of a portrait or an animation image is displayed in color on a computer screen or a television screen, but can be applied to other fields and moving image display of other images.

[0088]

According to the present invention, when a moving image of a color image (for example, a portrait or an animation image) is created, first, an image is created by a plurality of closed curves in which colors are omitted, and then the created closed curve is transformed. Since a plurality of images are prepared and the boundary and the inside of the deformed closed curve corresponding to the plurality of images are filled to create a moving image of a color image, the following effects can be obtained. For example, when displaying a portrait video in color,
It is not necessary to have all or part of the data such as portraits for the number of screens that move the image as in the past, but if you simply have closed curve data, color data and sequence data (multiple deformation information) that changes facial expressions. Video can be displayed in color. As a result, it is possible to display a moving image such as a portrait or an animation image in color while reducing the amount of data, reducing the capacity of the memory device, and at low cost.

[Brief description of drawings]

FIG. 1 is a configuration diagram showing a first embodiment of a portrait creation device according to the present invention.

FIG. 2 is a flowchart showing a main program of a portrait creation process of the embodiment.

FIG. 3 is a diagram showing data stored in a ROM of the same embodiment.

FIG. 4 is a diagram showing data stored in a RAM of the same embodiment.

FIG. 5 is a flowchart showing a timer interrupt process of the embodiment.

FIG. 6 is a flowchart showing a subroutine of a process of creating closed curves A to F of the same embodiment.

FIG. 7 is a flowchart showing a subroutine of dot calculation processing of the same embodiment.

FIG. 8 is a flowchart showing a subroutine of dot calculation processing of the same embodiment.

FIG. 9 is a flowchart showing a subroutine of data calculation processing when creating a Bezier curve of the same embodiment.

FIG. 10 is a flowchart showing a subroutine of a data calculation process when creating a Bezier curve according to the same embodiment.

FIG. 11 is a flowchart showing a subroutine of color setting processing of the embodiment.

FIG. 12 is a flowchart showing a subroutine of color setting processing of the embodiment.

FIG. 13 is a diagram illustrating a color state flag of the embodiment.

FIG. 14 is a diagram showing a rectangle of a graphic transformation process of the embodiment.

FIG. 15 is a diagram illustrating a graphic transformation process of the embodiment.

FIG. 16 is a diagram illustrating a graphic transformation process of the embodiment.

FIG. 17 is a diagram showing an example of a moving image display of a portrait in the embodiment.

FIG. 18 is a flow chart showing a main program for portrait creation processing according to the second embodiment of the present invention.

FIG. 19 is a diagram showing data stored in a ROM of the same example.

FIG. 20 is a diagram showing data stored in the RAM of the same embodiment.

FIG. 21 is a flow chart showing a main program for portrait creation processing according to the third embodiment of the present invention.

FIG. 22 is a diagram showing data stored in a ROM of the same example.

FIG. 23 is a diagram showing data stored in the RAM of the same embodiment.

[Explanation of symbols]

1 CPU 2 ROM (deformation information storage means, sequence data storage means, small sequence data storage means) 3 RAM 4 switch section 5 display section (display means) 6 printing section (printing means) 21 closed curve creating means 22 image transforming means 23 color Image creation means

Claims (16)

[Claims] 1. The color of a predetermined basic image is omitted,
In addition, a plurality of closed curves having a predetermined priority are combined and created, and the closed curves constituting the created basic image are modified according to the modification information to create a plurality of modified images, and the closed curves in the plurality of modified images are created. An image processing method characterized in that a plurality of predetermined color images are created by filling the boundary and the inside of the same with a predetermined color in accordance with the priority order. 2. The image processing method according to claim 1, wherein the closed curve is created by one or more of a straight line and a Bezier curve. 3. In the process of creating the predetermined image with a plurality of closed curves in which colors are omitted, closed curve data including predetermined coordinate data of each closed curve is used to create each of the plurality of closed curves. 2. The image processing method according to claim 1, wherein in the process of filling the boundary of the closed curve and the inside thereof with a designated color, color data for designating the color of the filling is provided for each closed curve. 4. The process of creating the closed curve by a straight line is to form a straight line by a large number of dots. The two dot positions forming both ends of the straight line are determined, and the inclination of the dot position is calculated. The process and the process of accumulating the error when the x-coordinate or the y-coordinate is increased by one unit according to this inclination, and the x-dot of the next dot only when the error exceeds a predetermined value.
The image processing method according to claim 1, further comprising: a process of changing a position of a coordinate or ay coordinate and subtracting a constant value from the accumulated error. 5. A plurality of sequence data prepared by combining a plurality of types of deformation information for deforming the closed curve of the basic image are prepared, and the basic sequence is sequentially based on the deformation information constituting one selected sequence data among them. The image processing method according to claim 1, wherein the closed curve forming the image is modified. 6. A plurality of pieces of deformation information for deforming the closed curve of the basic image are prepared inside small sequence data formed by combining a plurality of types, and necessary small sequence data is selected from the plurality of small sequence data. 2. The image processing method according to claim 1, wherein sequence data is created by combining a plurality of them, and the closed curve forming the basic image is sequentially deformed based on the deformation information forming the sequence data. 7. The deformation information for deforming the closed curve of the basic image is created by an input operation, and the closed curve forming the basic image is modified according to the created deformation information. The described image processing method. 8. A predetermined basic image, the color of which is omitted,
And a closed curve creating means for creating a combination of a plurality of closed curves having a predetermined priority, a deformation information setting means for setting deformation information for deforming each closed curve constituting the created basic image, and a basic information according to the deformation information. An image transforming means for transforming a closed curve of an image to create a plurality of deformed images, and a border and the inside of each closed curve in the created deformed images are filled with a predetermined color according to a priority order to create a plurality of predetermined color images. An image processing apparatus, comprising: 9. The image processing apparatus according to claim 8, wherein the closed curve creating means creates the closed curve by one or more of a straight line and a Bezier curve. 10. The closed curve creating means creates closed curve data composed of predetermined coordinate data of each closed curve for creating each of a plurality of closed curves forming a predetermined basic image, and the color image creating means comprises: 9. The image processing apparatus according to claim 8, wherein the color data corresponding to each closed curve is stored, and the boundary of each closed curve and the inside thereof are filled with a color corresponding to the stored color data. 11. The image processing apparatus according to claim 8, further comprising display means for displaying a plurality of color images created by said color image creating means. 12. The image processing apparatus according to claim 8, further comprising a printing unit that prints a plurality of color images created by the color image creating unit. 13. The closed curve creating means, when creating a closed curve by a straight line composed of a large number of dots, determines two dot positions forming both ends of the straight line, and calculates an inclination of the dot position. Means for accumulating the error when the x coordinate or the y coordinate is increased by one unit according to this inclination, and the x of the next dot only when the error exceeds a predetermined value.
9. The image processing apparatus according to claim 8, further comprising means for changing a position of a coordinate or ay coordinate and subtracting a constant value from the accumulated error. 14. The modification information setting means selects sequence data composed of a combination of a plurality of modification information, sequence data storage means storing a plurality of sequence data, and one sequence data from the plurality of sequence data. 9. The image processing apparatus according to claim 8, further comprising a sequence data selection unit. 15. The modification information setting means is a small sequence data storage means for storing a plurality of small sequence data composed of a combination of a plurality of modification information, and a necessary one is selected from the plurality of small sequence data for combination. 9. The image processing apparatus according to claim 8, further comprising a sequence data creating unit that creates sequence data. 16. The deformation information setting means includes deformation information creating means for creating deformation information by an input operation, and deformation information storing means for storing the deformation information created by the deformation information creating means. 9. The image processing apparatus according to claim 8, wherein: