JP3449563B2 - How to edit image data - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [0001] BACKGROUND OF THE INVENTION 1. Field of the Invention
Image editing such as copying machineGatheringLaw, especially for gray scale models.
Image data of the original image, such as a black and white photo or color photo
Image suitable for reading data and deforming it into a predetermined shape
EditionGatheringAbout the law. [0002] 2. Description of the Related Art Generally, the shape of an original image such as a character or a figure is changed.
Read and use this as original image data on a computer, etc.
Capture, transform this original image data into the desired shape,
Draw on an output device such as a CRT, plotter, etc.
Image data editing, such as recording on storage media, is
Has been done. [0003] Such conventional image data editing is performed.
In this case, a substantially rectangular original image (not shown) is shown in FIG.
A projective deformation for deforming into such a shape has been transferred. [0004] However, as described above,
In conventional image data editing using projective transformation,
As the image after deformation gradually increases in the deformation direction,
A large space is formed on the tip side in the deformation direction, and the image after deformation
There is a problem that the quality of the image is inferior. Also like this
To edit the original image data using
However, there is a problem that it cannot be transformed into a curved shape.
Was. In order to deal with such a problem, the present invention
The applicant is required to use the image data described in Japanese Patent Application No. 4-264858.
A data editing device and method were proposed. [0006] However, Japanese Patent Application No. 4-264858.
In the image data editing method described in
Image data (vector data) of the original image that can be binarized
(1) is intended to be transformed into a curved shape. [0007] Therefore, a depth image is further provided as image data.
A set of points (pixels) having image information (density values)
Or a black-and-white photo with a gradation composed of
Transforms color photographs into curved shapes
I wanted something that could be done. [0008] The present invention has been made in view of these points.
Is a monochrome or color photograph with gradation
Image editing that can freely deform the shape of the original image such asGathering
The purpose is to provide the law. [0009] The above-mentioned object is achieved.
The image data editing of the present invention according to claim 1The method is
Original image data as image information of the original image
A set of pixels in units that can be processed by
Xel is an integer message whose coordinate axis is a substantially rectangular reference outline frame.
This coordinate is stored in the reference outer frame as
The coordinate value of each pixel stored in the frame is determined by
Pixel at the position farthest away from the origin of the reference axis for each coordinate axis
Are converted to parameter coordinate values, where each coordinate value is 1.
The maximum value of each coordinate axis on the mesh coordinates
It is expanded to a substantially rectangular parameter coordinate set to 1 and out of the standard
The desired shape can be described by at least a function.
The shape is changed to a shape frame, and this deformed shape frame is used as a coordinate axis.
Coordinate, and this real coordinate is calculated from the origin of the coordinate axis of the real coordinate.
The coordinate value of the most distant position is set to 1 for each coordinate axis.
And the deformation parameter coordinates
The parameter coordinate values are
Convert each pixel to the actual
If you expand it as an integer float mesh coordinate on the benchmark
In addition, the area in the real coordinates is defined as a plurality of minimum units of the deformation processing.
And divide it into blocks.
Original image of one of the pixels located in the
And obtains an image density value of
For each rectangle on the real coordinates surrounding the lock,
It is characterized by filling with the image density value of the image
You. [0010] [0011] The image data editing of the present invention having the above configuration
According to the method, Storing the original image in a predetermined reference outer frame,
Determining the relative position of the original image to the reference outline
it can. Then, the reference outer frame is moved outside the desired shape.
When deformed into a shape frame, the original image is
Maintains the same positional relationship as the relative position to the reference outline frame.
With the original image deformed in the deformed outline frame with
The density value of the original image in the deformed outline frame.
it can. [0012] FIG. 1 to FIG. 24 show an embodiment of the present invention.
Will be described. This embodiment is directed to an image data editing system according to the present invention.Collection
OneMethod is used to transform a photo manuscript (photo transformation)
FIG. 1 shows a configuration of a main part of the image data editing apparatus.
FIG. As shown in FIG. 1, the image data of this embodiment
The editing device 1 includes an original image data reading unit 2 and a parameter
Data coordinate forming means 3 and deformed outer frame shape reading means 4
And deformation parameter coordinate forming means 5 and real coordinate calculating means
6, an image transformation / drawing unit setting means 7, and an original image density value
It is formed by the obtaining means 8 and the image drawing means 9. Next, the configuration of each part will be further described.
The original image data reading means 2 has a rectangular shape, for example.
Original pictures such as photosBarrel original imageImage information (original image data
Pixel) and pixel attributes (density values)
Scanner, memory, C to store in appropriate memory, etc.
It is formed from PU or the like (not shown).Therefore,
The original image is read by the original image data reading means 2 such as a scanner.
When scanning, each dot (image) of the read original image data is
Element) is coordinate data on rectangular coordinates, as is conventionally known.
To be stored in the memory of the original image data reading means 2.
And The parameter coordinate forming means 3 comprises:
The original read by the original image data reading means 2
Image data pixels are processed by a computer (not shown)
A collection of possible units of pixels, and each pixel
Integer mesh coordinates with the reference outline of the rectangle as the coordinate axis and
And store it in an appropriate memory, etc.
Developed into parameter coordinates with the large value as 1, and appropriate memory
Formed from memory, CPU, etc. (not shown) to be stored in etc.
Have been. That is, the parameter coordinate forming means 3
Original image data read by the image data reading means 2
Pixel of data that can be processed by computer
And these pixels are defined as a substantially rectangular reference outline frame.
The reference outer shape as an integer mesh coordinate with
The position of each pixel stored in this frame
The standard value is calculated for each coordinate axis from the origin of the coordinate axes of the reference outline frame.
The coordinate values of pixels at positions separated from each other are set to 1
Each coordinate on the mesh coordinates is converted to parameter coordinate values.
A roughly rectangular parameter with the maximum value of the coordinate axes as 1
It expands to coordinates. The computer
It is conventionally known that the unit pixel can be processed by
For example, the computer is independent on the display
Is the smallest unit that can be controlled by 8 bit units
In this case, X and X on the rectangular coordinates of the original image data are
Eight pixels (dots) in the Y-axis direction are defined as one pixel.
You. Of course, the computer is independent on the display
If the minimum unit that can be controlled by
Indicates that one bit of the read original image data is one pixel
Becomes Therefore, in the parameter coordinates, the coordinates
The coordinate value furthest from the origin of the axis is 1. Further explanation
Then, the “unit that can be processed by computer” is CC
Input data from a D-scanner or digital cameraU
WhatOf the density information and color information corresponding to each point of the image.
That means.Specifically, the image element stored in the memory
(Pixel) position data and its position
Color information (or shading information) data.example
For example, when expressing in black and white with 256 levels,
8-bit data corresponding to each point of the image is required.
16.7 million colors (167
7215 colors) corresponds to each point of the image
R: 8 bits, G: 8 bits, B: 8 bits
Site data is required. And the pixel is this
This means data corresponding to each point.For example,
When represented by 256, (x,
y) The coordinate data is processed by an 8-bit computer.
Can be treated as possible data. Also, this position
Color component data corresponding to coordinates is handled by computer
Can be handled as 8-bit x3 data
Wear. The deformed outer frame shape reading means 4
Is described by at least a function, such as a cubic Bezier function.
A reference outer shape frame transformed into a desired shape that can be described
Image data (shape data) of all deformed outline frames
The element is read by a scanner (not shown) or given in advance.
By selecting from a plurality of obtained desired deformed outline frames
Suitable memory etc.MemorizeScanner, memory, CP
U and the like (not shown). The deformation parameter coordinate forming means 5
Is read by the deformed outer frame shape reading means 4.
The pixels of the shape data (contour)Compu
DataProcessable units of pixelsAggregation ofToshiko
Is defined as a coordinate axis.
After storing the actual coordinates in a suitable memory or the like,
Expand to deformation parameter coordinates with the maximum value of the real coordinate axis set to 1
Memory, CPU, etc., which are stored in an appropriate memory, etc.
(Not shown).That is, the deformation parameters
The coordinate forming means 5 transforms the reference outer frame into a deformed outer frame.
Then, use this deformed outer frame as the actual coordinates
After storing the actual coordinates in the appropriate memory, etc.
Coordinates of the position farthest away from the origin of each coordinate axis for each coordinate axis
As appropriate, as the deformation parameter coordinates each having a value of 1,
Memory, CPU, etc. (not shown) to be stored in a simple memory, etc.
Is formed from. Further, the real coordinate calculating means 6,
Xel parameter coordinate value in transformation parameter coordinate
After converting to the transformation parameter coordinate valuesDesired processing
Each pixel through the processing accuracy setting unit 10
Develop on the bench as integer float mesh coordinates, and
Memory, CPU, etc. (not shown) to be stored in a simple memory, etc.
Is formed from.As is well known, a computer
In the operation using
A "floating point number" that can move several points
There is And "float mesh coordinates"
Coordinate representing the value on the benchmark in this "floating point number".
The coordinates with the reference outline frame as the coordinate axis
Mesh coordinates "
Expressed as "integer float mesh coordinates". In addition,
Since it is a number, the value after the decimal point is 0. The image transformation / drawing unit setting means 7
Was developed on the float mesh coordinates of the real coordinates
Considering the image quality and processing speed,
Column processing, and then divide each column into rows.
By the deformation outline frame (float mesh seat of real coordinates)
The area within the mark is formed by the desired four pixels.
Divided into appropriate minimum units (blocks)
From memory, CPU, etc. (not shown)
Is formed. The original image density value acquiring means 8 is provided
4 pixels forming each blockOne of
Pixel originalDensity value (integer mesh of the reference outline frame)
Density values of four pixels on the menu coordinates: in this embodiment
Is the upper left one of the four pixels on the mesh coordinates
To store the density value of the pixel in a suitable memory or the like.
It is formed from a memory, a CPU and the like (not shown). The image drawing means 9 comprises:The actual coordinates
Each rectangle on the real coordinates surrounding the block
The image density value of the acquired original image, that is, the image density value
Density value for each block acquired by acquisition means 8Based on
To form the deformed image data
The transformed image data on a CRT and plotter.
Memory to output to an appropriate output device such as
(Not shown). Note that the processing accuracy setting unit 10 is provided as needed.
May be attached to each means in the first place.
The configuration is not limited to the example. In addition, each memo
And all CPUs etc. should be provided collectively in the control unit etc.
Can also. Next, in the present embodiment having the above-described configuration,
Creation of image data editing method using image data editing device 1
The use will be described. An outline of the image data editing method according to the present embodiment will be described.
If you explain it,Of the original imageImage information
Pixels and pixel attributes (density) as (original image data)
Value) can be processed by a computer (not shown).
A cell, and its pixels are aligned with the reference outline frame as the coordinate axis.
Stored in the reference outer frame as mesh coordinates of
Parameter coordinates of the quasi-outer frame with the maximum value of its coordinate axis as 1
And expand the reference outline to a function.
Deformed into a desired shape that can be described by d-function
Frame, and the actual coordinates using the deformed outer frame as a coordinate axis.
To the transformation parameter coordinates with the maximum value of the real coordinate axes
The pixel on the parameter coordinates is transformed
Data on the coordinates of the
Unfolded on real coordinates as integer float mesh coordinates
And the area within the deformed outline (actual coordinates)
Divide into blocks as the minimum unit for deformation processing, and
Mark on the reference outline frame equal to the relative position of each block.
To fill with cell density value (image density value)
It is designed to transform original pictures such as photographs.
That is, original image data as image information of the original image is copied.
A set of pixels in units that can be processed by the computer,
Using these pixels as the coordinate axes of the substantially rectangular reference outline frame
It is stored in the reference outer frame as integer mesh coordinates, and
The coordinate value of each pixel stored in the quasi
The pixel at the position furthest away from the coordinate axis origin for each coordinate axis
Are converted to parameter coordinate values with each coordinate value of 1
Each pixel on the mesh coordinates to 1
To the parameter coordinates of a substantially rectangular
A deformed contour of a desired shape that can be described by at least a function
To the actual coordinates with the deformed outer frame as the coordinate axis.
Then, the actual coordinates are moved from the origin of the coordinate axes of the actual coordinates to the coordinate axes.
Deformation where the coordinate value of the most distant position is 1 for each
Parameter coordinates, then the parameters for each pixel
Deformation parameter coordinates in transformation parameter coordinates
Converts each pixel to an integer value on the real coordinates
It is developed as tomesh coordinates,
Divide the area into multiple blocks as the minimum unit for deformation processing
And for each block the pixels located within that block
The image density value of the original image of one of the pixels is obtained and
The area in the coordinates is divided into rectangles on the actual coordinates surrounding the block,
With the image density value of the acquired original image By filling
In addition, the original image such as a photograph is deformed. I
Therefore,The reference outline frame containing the original image data is
Deform into a deformed outline frame of a desired shape that can be described by a number
Thus, the original image data is converted. FIG. 2 shows such an image data editing method.
This will be described with reference to the flowchart shown. First, in step ST10, a photograph
Pixels and pixel attributes (density) that are the image information of the original image
Appropriate original image data reading means such as a scanner
2 and read with the desired resolution.
The process proceeds to step ST11. Next, in step ST11, the original image
Of the image information by the appropriate parameter coordinate forming means 3
Pixels of a unit that can be processed by the image data editing device 1
The pixel from the integer mesh coordinates.
After storing in the reference outline frame,
Data coordinates, and proceeds to the next step ST12.
That is, original image data as image information of the original image is copied.
A set of pixels in units that can be processed by the computer,
Using these pixels as the coordinate axes of the substantially rectangular reference outline frame
It is stored in the reference outer frame as integer mesh coordinates, and
The coordinate value of each pixel stored in the quasi
The pixel at the position furthest away from the coordinate axis origin for each coordinate axis
Are converted to parameter coordinate values with each coordinate value of 1
Each pixel on the mesh coordinates to 1
To the parameter coordinates of a substantially rectangular This
Store the original image in the specified reference outline frame, and
The position relative to the form can be determined. Next, in step ST12,
Read the shape of the deformed outer frame as appropriate.
Read by the take-out means 4 and proceed to the next step ST13.
proceed. Next, in step ST13, the outside of the deformation
The transformation parameter coordinates for the form
Formed by the data coordinate forming means 5 and the next step ST14
Proceed to.That is, the reference outer frame is defined as at least a function
Into a desired shape that can be described by
After using the deformed outer frame as the actual coordinates as coordinate axes,
The actual coordinates are calculated from the origin of the coordinate axes of the actual coordinates for each coordinate axis.
The coordinate values of the separated positions are each set to 1. Next, in step ST14, an appropriate
Pixels on the parameter coordinates are calculated by the actual coordinate calculating means 6.
The transformation parameters on the transformation parameter coordinates, and
Float mesh of pixels converted on data coordinates
The coordinates are expanded to real coordinates (deformed outline frame), and the next
Proceed to step ST15.That is, the parameter of each pixel
Deformation parameter in transformation parameter coordinates
And convert each pixel to an integer coordinate on real coordinates.
Develop as rotomesh coordinates. Next, in step ST15, the actual coordinates
(Float mesh coordinates)
Minimal deformation processing by appropriate image deformation drawing unit setting means 7
Divide into units (blocks) and proceed to the next step ST16.
Run.In other words, the area within the actual coordinates is
Divide into blocks as small units. Next, in step ST16, an appropriate
The original image density value acquisition means 8 assigns pixels to each block.
The density value of the pixel on the corresponding reference outline (the original density
Value) and proceeds to the next step ST17.sand
In other words, for each block, the pixels located in that block
The original image density value of one of the pixels
You. Details will be described later. Next, in step ST17, the outside of the deformation
Each block in the frame (actual coordinates) is converted to an appropriate image drawing means 9
To fill with the original density value and the desired output
A series of processes is terminated by outputting the data to the device.Details
Details will be described later. The above-described image data editing method
The main part will be specifically described in detail. First, out of the standard of the image data editing of this embodiment.
About the combination of line segments when the form is deformed
h. FIG. 3 shows a combination of line segments when the reference outer frame is deformed.
FIG. 4 is an explanatory diagram for explaining an example, in which a indicates an example of an original image.
B indicates a reference outer frame, and c indicates a vertical change of the reference outer frame.
Shows the combination of line segments at the time of shaping, and d is the deformation after vertical deformation.
E indicates a vertically deformed image, f indicates
Shows the combination of line segments when the reference outer frame is deformed in the horizontal direction,
g indicates the deformed outline frame after the horizontal deformation, and h indicates the horizontal deformation.
The image shown is shown. The image information of the original picture (photograph) shown in FIG.
A point A, a point as shown in FIG.
A-B-C- of a rectangular shape composed of B, point C, and point D
A reference outer frame 11 made of D is formed. And Baek
Considering the direction of the torque, the reference outer frame 11 shown in FIG.
In the figure, upper and lower line segments AB: A (t) and line segments DC: B
Using the combination with (t), as shown in FIG.
By deforming the reference outer frame 11 in the vertical direction,
When forming and obtaining an image deformed in the vertical direction shown in FIG.
In the case shown in FIG.
Between the right line segment AD: A (t) and the line segment BC: B (t)
Using the combination, as shown in FIG.
A deformed outer frame 12 is formed by deforming the quasi-outer frame 11, and FIG.
Either of obtaining vertically deformed image shown in 3h
The original image is transformed using one of the combinations. Next, the source of deformation as image data editing
As an example, the vertical deformation is explained with reference to FIGS.
explain. FIG. 4 is an explanatory view for explaining vertical deformation.
And a represents a pixel stored in the reference outer frame,
b indicates a pixel on the reference outer frame, and c indicates a pixel on the reference outer frame.
Where d is the pixel on the parameter coordinates
Indicates the cell position, e indicates the deformation outline frame, and f indicates the deformation parameter.
Indicates the position of the pixel on the meter coordinates, g is the deformed outline frame
Shows the location of the transformed pixel above. As shown in FIG. 4A, a rectangular AB
The point A at the upper left of the reference outer frame 11 composed of -CD is defined as the origin.
And the line segment AB direction is defined as a coordinate axis x (x axis), and the x axis
Let XS be the maximum value of (the number of pixels = distance)
And the line segment A-D direction as a coordinate axis y (y-axis), and the y-axis
(The number of pixels = distance) is YS. Toes
The coordinate value of the point A forming the reference outer frame 11 is (0,
0), the coordinates of point B are (XS, 0), and the coordinates of point D are
(0, YS), and the x-axis direction of the pixel is xbase.
And the y-axis direction of the pixel is referred to as ybase.
The line segment in the deformation direction is pxlcnt, and the line segment in the straight line direction is
It is called base. And formed by the x-axis and the y-axis.
Mesh coordinates (x−
Pixels as image information of the original image
Will be delivered. Pixels (mesh) on this reference outline frame 11
(Coordinates) are shown in FIG. Here, as shown in FIG.
At the mesh coordinates (xy coordinates) 13 of the pixel P
The coordinate value obtained is P = (Px, Py). Further explanation
And a rectangular pixel constituting the image as shown in FIG.
The upper left corner is A, the upper right corner is B, the lower left corner is D, and the lower right corner is C
Then, the position of the pixel constituting the image is AB
Is the x-axis and A-D is the y-axis, in two-dimensional coordinates (x, y).
Can be represented. For example, taking point A as the origin,
The pixel coordinates are (0,0), are on the AB axis and
The coordinates of the third pixel from can be expressed as (2,0)
it can. This is the mesh coordinates. Coordinate values x and y are
It is a numerical value (see c in FIG. 4). In addition, "standard outer frame"
Is a rectangular shape in FIG.
Refers to the form ABCD. This rectangle is called
It is referred to as “quasi-outer frame”. And inside this frame (on the line of the frame
Pixel is also present)
The pixels that make up the image will be included. This (vs.
If the pixel data of the elephant image is within the rectangle
Is called “storage”.In other words, each pixel of the aggregate
The task of putting the file into the frame is the memory of the computer
Color information corresponding to the position coordinates of the image pixels
Or shading information). Next, as shown in FIG.
The maximum value XS of the x-axis at the coordinates (xy coordinates) 13 is 1
And the maximum value YS of the y-axis is set to 1
Is converted to the v axis, and is formed by the u axis and the v axis.
Is the parameter coordinate 14 (uv coordinate),
The coordinate value P of the pixel P at the xy coordinate 13 is P = (P
x, Py) is a parameter locus on the uv coordinates 14.
It can be represented by a standard value P = (u, v) (where 0 ≦
u ≦ 1, 0 ≦ v ≦ 1). That is,According to the coordinates of FIG.
The parameter coordinate values of each pixel indicated by
4 to the deformation parameter coordinates indicated by the d coordinate.
The transformation is performed to transform parameter coordinate values.
To explain further,The origin of the reference outer frame 11 shown in FIG.
The coordinate value (0,0) of the pixel located at the point A is
Point A which is the origin in the parameter coordinates 14 shown in 4d
Becomes the coordinate value (0, 0), and the reference shown in FIG.
The coordinates (XS, XS) of the pixel located at point B of the outer frame 11
0) is a point at the parameter coordinates 14 shown in FIG.
In FIG. 4B, the coordinate value becomes (1, 0).
The coordinates (X
(S, YS) corresponds to the parameter coordinates 14 shown in FIG.
At point C, the coordinate value (1, 1) is obtained.
Coordinates of the pixel located at point D of the reference outline 11 shown
(0, YS) corresponds to the parameter coordinates 14 shown in FIG.
At the point D in the coordinate value (0, 1). Note that A
Parameters of pixels located other than point, point B, point C and point D
The data coordinate value P = (u, v) is, as is known,
Each one of the x-axis and y-axis of mark 13 that is furthest away from the origin
It can be easily calculated by proportional calculation with these coordinate values as 1.
Can be. Next, the coordinates are expanded to the coordinates shown in FIG.Tapa
Each pixel based on the parameter coordinate values is shown in FIG.
FIG. 4 (g) is developed in order as shown in FIG.
By obtaining the Tomesh coordinates, each pixel is converted to the real coordinates
Expand to integer float mesh coordinates on axis
To That is, the rectangular reference outer frame 11 (A-B
-CD) and the line segment AB and the line segment D-
C is in the desired shape that can be described by a cubic Bezier function
A'-B'-C'-D 'shown in FIG.
In the same manner as above, the point A 'is defined as the origin.
And the direction of the line segment A'-B 'is defined as a coordinate axis x (x-axis),
The maximum value of the x-axis is set to XS ', and the line segment A'-D'
The direction is coordinate axis y (y-axis), and the maximum value of the y-axis is Y
After setting the actual coordinates 15 as S ',
When the maximum value XS 'of the x-axis is converted to a u-axis which is 1,
4 is converted to a v-axis in which the maximum value YS 'of the y-axis is 1.
F, the deformation formed by the u-axis and the v-axis
Parameter coordinates (uv coordinates) are assumed to be 16. Also, the standard
The deformation of the outer frame 11 into the deformed outer frame 12 is out of the standard.
Four line segments forming the form 11 (A-B-C-D)
Replacing with four line segments constituting the deformed outline frame 12
Execute by That is, the reference outer frame 11 (ABC
-D) is calculated by using the equation
It is read by the frame shape reading means 4 or given in advance.
Deformation from the shape data of multiple obtained desired deformation outline frames
For each of the equations of the four line segments that make up the outer frame 12,
Can be easily implemented by replacing them to be equal
It is self-evident. In addition, the deformation direction of the deformed outer frame 12
One of the line segments (upper or left) is described by F (t)
fcurve, the other (bottom or right) is described by G (t)
The linear direction of the deformed outer frame 12
One of the line segments (left or top) is `` aline '' and the other (right
Or below) is called a line (FIG. 4e). Here, the u axis of the parameter coordinates 14 and
It is assumed that the parameter coordinate value does not change even if the v axis changes.
Then, when the deformation parameter coordinate value of the pixel P is calculated,
As shown in FIG.
In the process of changing from A'-B 'to curve D'-C'
Of the curve that occurs, Represents the j-th line in the v-axis direction
A curve F (Lj0-Lj1) is calculated, and the calculated curve is calculated.
F is Lj (u)(= (Ljx (u), Lji (u)))
WhenWhen expressed, On the curve F, i-th in the u-axis direction
Parametersui determined by the point P '(Ljx (u
i), Lji (ui)) are the deformation parameter coordinate values of the point P
BecomesTherefore, each pixel on the deformed outline 12
P ′ is the mutual of each pixel P on the reference outline frame 11.
To the reference outer frame 11
Maintain the relative positional relationship of the original image data
In this way, the original image data can be transformed into the desired shape.
Can be done. Then, the coordinate values of the deformation parameters are
Convert as integer float mesh coordinates on target 15
Thereby, the deformation ends.That is, each picture of the original image
The relative positions of the xel to the reference outer frame 11 and
While maintaining the same positional relationship,
The transformation is completed by positioning each pixel of the original image.
You. ThisOf the integer as the real coordinates 15 after the transformation of
The mesh mesh coordinates are shown in FIG. Next, vertical conversion of the original image data and
The method of calculating the intermediate curve in the horizontal transformation will be described.
You. FIG. 5 is a diagram for explaining a method of calculating an intermediate curve.
FIG. As shown in FIG. 5, the curves A0-A3 are A (t) = (1-t)³A0 + (1-t)²・ T ・ A
1+ (1-t) · t²・ A2 + t³・ A3 Curve B0-B3 B (t) = (1-t)³・ B0 + (1-t)²・ T ・ B
1+ (1-t) · t^{2.B2 + t 3}・ B3 It is assumed that Then, from the curves A0-A3 to B0-B
3 is represented by F (t) (where 0 ≦
t ≦ 1), F (t) = (1-t)³・ F0 + (1-t)²・ T ・ F
1+ (1-t) · t²・ F2 + t³・ F3 Can be described as Here, the intermediate curve F (t) is described.
Generally describe the description parameters F0, F1, F2, and F3.
To state: F0 = (1-u) · A0 + u · B0 F1 = (1-u) · A1 + u · B1 F2 = (1-u) · A2 + u · B2 F3 = (1-u) · A3 + u · B3 (Where 0 ≦ u ≦ 1). In this equation, u isVerticalIn case of directional deformation
Value of v (coordinates in the v-axis direction in parameter coordinates)
Value)sideIn the case of directional deformation, the value of u
Using the coordinate values in the u-axis direction in the parameter coordinates).
The intermediate curve F at each deformation parameter coordinate 16 and
And deformation parameter coordinate value P ′ (u, v) (= (Ljx
(Ui), Lji (ui))) can be calculated.
You. This transforms the parameter coordinates of each pixel
Convert to transformation parameter coordinate values in parameter coordinates
Float mesh of each pixel on real coordinates 15
It can be developed as coordinates. Next, FIG.
6 to FIG. FIG. 6 is a schematic diagram for explaining column division.
FIG. 7 is a schematic diagram illustrating line division, and FIG.
FIG. 9 is a schematic diagram for explaining, and FIG. 9 illustrates band setting.
FIG. 10 is an explanatory diagram, and FIG.
FIG. 11 is an explanatory diagram illustrating a block. The image deformation drawing unit setting includes a deformation outer frame.
Integer float mesh coordinates on 12 real coordinates 15
Each pixel that has been expanded is temporarily stored as shown in FIG.
Column processing, and then each band is divided into
By dividing the lines as shown in FIG.
2 (real coordinate float mesh coordinates 16)
As shown in FIG. 8, the desired four pixels P (shown in FIG.
)), The minimum unit 17 for deformation processing (block)
This is done by dividing into The setting of the band is as shown in FIG.
For example, the start points P00 and P01 have already been set.
Settings, P10 and P11 are set on the picurve
Number of cells (flpxl) and pixels on gcurve
Both numbers (glpxl) are above a predetermined threshold
Done by gradually expanding the processing bandwidth up to
You. This specific threshold value depends on image quality and processing speed.
Can be determined from the balance of
It is good to be about. When setting up a band for the first time
The positions of P00 and P01 as starting points are
Each will be located on aline. The setting of the blocks is as shown in FIG.
For example, the start points B00 and B10 are already set.
Is set, B01 and B11 are set in the linear direction (B00
-B01) pixel number (cpxl) and linear direction
Both (B10-B11) pixel count (blpxl)
Increases the processing bandwidth until the threshold is exceeded
It is done by going. The value of this specific threshold
Can be determined based on the balance between image quality and processing speed
However, for example, it may be set to about three pixels. In addition,
When setting the lock for the first time, B00 as the starting point
And B10 are located on fcurve
Will do. As described above, the band and the block
Is set, and as shown in FIG.
Consisting of B00-B01-B11-B10 formed
The square is defined as the minimum unit 17 (block) of the deformation processing. That is, as shown in FIG.
The lock has the deformation curves j and j + i, the straight lines i and i +
1. Next, the acquisition of the original image density value will be described.
You. The original image density value is obtained by
In the present embodiment, as a density value for filling
Considering the processing speed of filling each block,
The four pixels (B00, B01, B11, B1)
0) corresponding to the upper left pixel (B00)
By acquiring the image density value (density value: G00) of the image
Is performed. The original image density value (G
00) is stored in the memory before the block is deformed.
The original image density value (G00)
It has become so. That is, the minimum unit 17 of the deformation process
(Block: B00-B01-B11-B10) determined
At that point, the position coordinates of the corresponding original image are uniquely determined,
Easily obtain the density value of the original image corresponding to pixel (B00).
Can be obtained.Specifically, as an image
It is assumed that N pixels and M pixels are arranged vertically. number
From the 0th to (N-1) th in the horizontal direction,
In the vertical direction, counting from 0th to (M-1) th
, The n-th pixel P in the horizontal direction and the m-th pixel P
Is (n-1, m-1) (see c in FIG. 4).
See). Then, obtain the density value of the original image from the position coordinates of the original image
If so, the image density value corresponds to the pixel coordinate value.
State and already stored (generally,
At this point, such a response can already be taken.
ing). Examples of image data with 256 gradations
And take it up, the density value of the above position coordinates (0,0)
Assume that G (0,0) is stored at address a in the memory. This
Here, G (0,0) is the density corresponding to the position coordinates (0,0).
Indicates a degree value. Dark of position coordinates (1, 0)
If the degree value is stored at address a + 1, the position coordinates
The density value G (n, m) of (n, m) is the density of the address Nm + n.
Can be extracted as a degree value. Further, the deformation process
Minimum unit 17 (block: B00-B01-B11-B
10) is, for example, the same image density value (or color
-Image value) to specify the rectangle to be drawn
You. Next, FIG. 12 and FIG.
13 will be described. FIG. 12 illustrates a rectangular rectangle surrounding a block.
FIG. 13 is a drawing for explaining the deformation outline frame.
FIG. The image drawing is performed in each of the blocks 17 described above.
With the corresponding original image density values,
This is performed by outputting to a desired output device. The filling is performed as shown in FIG.
A square (B00-B01-B1) forming the block 17
1-B10), instead of painting this square (B
00-B01-B11-B10) on the real coordinates 15
Fill the rectangle (Q0-Q1-Q2-Q3)
This is done by: Block 17 is a rectangular rectangle
To paint the block 17 directly
The diagonal lines are processed at the same time.
When using a computer, the processing speed of the computer
This is to prevent the processing speed from becoming too slow.
Computer (a very expensive and specialized computer for scientific calculations)
However, it is not always necessary when using a computer. The rectangular quadrangle has its minimum value.
Is (xmin, ymin), and the maximum value is (xmax, ymin).
y0), Q0 = (xmin, ymi
n), Q1 = (xmin, ymax), Q2 = (xma
(x, ymax), Q3 = (xmax, ymin)
Can easily calculate the maximum / minimum value of coordinates.
Can be. Further, as shown in FIG.
Filling of the deformed outline frame by the shape
Fill with a rectangle of each rectangle as they overlap
Overwrite the overlapping part when
Image by filling the inside of the deformed outline
The processing speed of depiction can be increased. Next, such an image data editing method will be described.
Algorithm (Processing Structure) FIGS. 14 and 15
According to the flowchart shown in FIG.
Further explanation will be given. FIGS. 14 and 15 show an image data editing method.
FIG. 1 is a flowchart illustrating the algorithm of FIG.
4 shows one of the divided flowcharts, and FIG.
The flowchart is divided into two parts, and FIG.
It is a schematic diagram showing a Bezier function and its parameters,
FIG. 17 is an explanatory diagram for explaining the cubic Bezier function.
FIG. 8 is an explanatory diagram for explaining a method of calculating the distance of the curve.
19 is an explanatory diagram for explaining an example of a method for calculating the distance of a curve.
FIG. 20 illustrates another example of a method for calculating the distance of a curve.
FIG. 21 illustrates a linear cubic Bezier function.
FIG. Here, the deformation of the cubic Bezier function is
This will be briefly described with reference to FIG. The cubic Bezier function is shown in FIG.
As shown, two end points P0, P3 and two
There is a point P2. The positions of these two curve control points P1 and P2
By changing the position, the cubic Bezier function is transformed. Ingredient
Physically, when changing the positions of the two points P1 and P2,
First, at the user interface level,
Perform an operation such as clicking the mouse near the point
The change mode of the point is entered by the neighborhood judgment. Press mouse
With the mouse down, change its position by moving the mouse
When the mouse is released, the position changes anew
The position is registered and the point change mode is released.
To generate a new cubic Bezier curve
become.Further, as an example of the modification of the reference outline frame, the reference outline
Transform one linear part of the frame into a cubic Bezier function
The operation will be described with reference to FIG. Bezier parame
Data P0 (P0x, P1y), P1 (P1x, P1
y), P2 (P2x, P2y), P3 (P3x, P3
y), the curve L (Lx, Ly) is 0 ≦ t ≦
L = P0 (1-t) using t being 1 ^Three + P1t (1-
t) ^Two + P2t ^Two (1−t) + P3t ^Three Can be written as
Wear. Then, one line of image elements is plotted on the curve L shown in FIG.
By changing this curve, the corresponding pixel
Will change. The actual operation is as follows
P2 that is not normally needed only when performing the deformation below and
The coordinates of P1 are displayed.
Click near 2 (about 5 pixels) to specify the operating point
Then move the mouse to change its position
You. Click again after confirming the changed point,
Once the position is determined A Bezier curve according to these coordinates is generated again
Pixels that correspond to successive points on this curve (pixels
Is completed by arranging the data of (1). As shown in FIG. 14 and FIG.
When the processing structure of the data editing method is started,
In step ST20, initial settings are performed first, and the next
Proceed to step ST21. The initial setting in step ST20 is as follows.
This is accomplished by sequentially performing the contents of the following eight items. (1) pxlcnt of original image
Setting of the number of cells) and base (number of pixels in the linear direction)
(Set according to the deformation direction). That is, vertical deformation and horizontal deformation
Select the direction deformation. This gives the integer mesh coordinates
The pixels of the original image are stored in the reference outline frame 11 consisting of
can do. 2) Deformation pair (fcurve, gcurv)
e) Coefficient of curve parameter (cubic Bezier parameter)
Calculation of the value (for calculating the intermediate curve). Here, the coefficient value is "shape
The operation coefficient of the parameter is referred to as "
Let the parameters of the cubic Bezier function B (t) be P0, P
1, P2 and P3, B (t) = (-P0 + 3P1-3P2 + P3) t³+3
(P0-2P1 + P2) t²+3 (-P0 + P1) t +
P0 = M3t³+ M2t²+ M1t + M0 M3, M2, M1, and M0 when expressed as follows. Toes
The line segment in the deformation direction of the deformed outer frame 12 is
It will be described as parameter. (3) The cubic Bezier curve (fcurve)
Calculation of the number of continuous contour pixels (fccpxl) (e, FIG. 4
(FIG. 9). Here, the number of continuous contour pixels (fccpxl)
The calculation is to calculate the distance of the curve. The calculation of the distance of the curve will be described.
Generally, it is difficult to calculate the distance of a curve programmatically.
Despite the difficulty, the third-order Bezier function is as shown in FIG.
Determined by four parameters P0, P1, P2, and P3
And the function B (t) can be defined as
make use of, B (t) = (-P0 + 3P1-3P2 + P3) t³+3
(P0-2P1 + P2) t²+3 (-P0 + P1) t +
P0 It can be expressed as. Then, the first derivative by t is B ′
(T) is the curve distance DH Can be represented by However, the actual calculation is performed by a program.
Has processing time and other issues.
Considering the complexity, as shown in FIG.
Decompose into straight lines, and treat the sum of the straight line distances as curve distances
Is desirable. That is, to calculate the curve distance,
It is good to choose from the two methods shown in the following. The first method for calculating the curve distance is shown in FIG.
A representative point on the curve in advance to form a polygonal line
The number corresponding to the sum of the straight-line distances is defined as the number of division points.
Things. To improve the accuracy in this case, a straight line
The degree (number of divisions) may be increased. The second method of calculating the curve distance is shown in FIG.
In this way, a division point is formed according to the curvature of the curve,
The sum of the line distances is defined as a curve distance. (4) The cubic Bezier curve (gcurve)
Calculation of the number of continuous contour pixels (gcpxl) (same as (3))
Mr). (5) Continuous contour pic of straight line
Calculation of the number of cells (alpxl) (e in FIG. 4, FIG. 9). here
The calculation of the number of continuous contour pixels (alpxl) is directly
It is to calculate the distance of the line. The calculation of the distance of a straight line will be described.
As shown in FIG. 21, the start point / end point of the straight line is P0
(P0x, P0y) and P1 (P1x, P1y)
In this case, the distance between the two points is Can be represented by This is easily calculated programmatically.
Can be issued. (6) Continuous contour picture of a straight line (bline)
Calculation of the number of cells (blpxl) (similar to (5)). (7) Processing accuracy (rounding error recovery value: X
(ERR, YERR) setting. The setting of the processing accuracy means the appropriate processing described above.
An integer forming the reference outer frame 11 by the accuracy setting unit 10
Transform each pixel developed on mesh coordinates 13 of
Integer float mesh coordinates 15 forming the outer frame 12
It is developed continuously without any gaps above. The above (2) to (7) are integer flow numbers.
Deformable outline frame 1 consisting of auto mesh coordinates 15 (real coordinates)
Second, the pixels of the original image can be developed. (8) Initial setting for loop processing. Toes
The band processing for dividing the deformed outer frame 12 into columns.
The left end point of the first band (P00 (left
The coordinate values of (upper), P01 (lower left) are set. Next, in step ST21, all
It is determined whether or not all the processing has been completed (processing completed?).
In the case of ES (processing end), the processing ends, and NO (processing
Is not completed), the next step ST22
Proceed to. Next, in step ST22, the process
Initialize the band and proceed to the next step ST23
You. Initial processing band in step ST22
The setting is accomplished by performing the following three items in order.
Is performed. (1) Right end point of processing band (shown in FIG. 9)
Set coordinate values of P10 (upper right), P11 (lower right)
You. To set the right end point of this processing band
Is (A) Terminate processing units in the upper and lower directions of deformation, and
Calculate value (t0) (B) Corresponding coordinate values (P10, P10) on the cubic Bezier curve
11) Ask This is done by: Specifically, the known pixel number
P00(Coordinates of the upper left point P00 on fcurve in FIG. 9)
value)From processing bandwidth(Set the distance between pixels to 1 or more)To
Accordingly, the pixel number P10 at the right end point(The upper right point P in FIG. 9
00 coordinate value on fcurve)As well as
The curve distance from xell number P00 to P10Fig. 17
According to the method shown in FIG.4e.And
Figure 9By dividing by the number of continuous contour pixels shown in
Ask for. Subsequently, the A'-B 'cubic Bezier curve relation shown in FIG.
Let t = t0 be the number t parameterThe cubic Bezier function
Substituting for B (t)The coordinate value of P10 is obtained. Likewise
The t parameter of the D'-C 'cubic Bezier function shown in FIG.
Assuming that t = t0, the coordinate value of P11 is obtained. (2) Straight line in the deformation direction (P00-P10)
Of the number of continuous contour pixels (flpxl) of FIG.
Here, the calculation of the number of continuous contour pixels (flpxl) means
This is to calculate the distance of a straight line. How to calculate the distance of a straight line
The method is the same as described above. (3) Straight line in the deformation direction (P01-P11)
Of the number of continuous contour pixels (glpxl) of FIG. 9 (FIG. 9).
Here, the calculation of the number of continuous contour pixels (glpxl) means
This is to calculate the distance of a straight line. Next, at step ST23, fl
Whether the distance of pxl is longer than the distance of glpxl (fl
pxl> glpxl? ) Is determined, and NO (flpxl
If the distance is shorter than the glpxl distance),
Proceed to step ST24a, and proceed to step ST24a.
And the longer glpxl distance is predetermined.
Whether or not it is smaller than the threshold of
Is smaller and the processing is unsuitable)
Is larger than the maximum value), the next step ST26
(FIG. 15), and the determination in step ST24a is YE
If S (less than the threshold), the next step
Proceeds to ST25a, and ends in step ST25a.
It is determined whether it is a processing unit and not one point.
In the case of ES (final processing unit), the next step ST2
6 (FIG. 15), and NO (not the final processing unit)
In this case, the process returns to step ST21 and proceeds to step ST24a.
Is NO, or the determination in step ST25a is Y
Repeat until ES is reached. If the determination in step ST23 is YES (f
lpxl distance is greater than glpxl distance)
Proceeds to the next step ST24b and proceeds to step S24b.
At T24b, the longer flpxl distance is preset.
Whether it is smaller than a predetermined threshold (preset
Is smaller than the width of the band that has been processed and is not processing)
If NO (greater than the threshold), the next step
Proceeding to step ST26 (FIG. 15), step ST24b
If the determination is YES (smaller than the threshold),
Proceeds to step ST25b
Is the final processing unit and whether it is not 1 point
And if YES (final processing unit), the next step
Proceeds to step ST26, NO (not final processing unit)
In the case of, the process returns to step ST21 and proceeds to step ST24.
b is NO, or the determination in step ST25b is
Repeat until YES. That is, from step ST23 to step ST23
By ST25 (25a, 25b), the distance of flpxl
And the distance of glpxl are both greater than the threshold.
Until the band width is determined as the processing unit.
I have decided. Next, in step ST26, the process
Whether all blocks in the symmetric band have been processed
(End of processing?), And if YES (End of processing)
Proceeds to step ST32, which will be described later, and returns NO (the processing ends).
If not, the process proceeds to the next step ST27.
Run. Next, in step ST27, the processing block
Performs lock initialization and proceeds to the next step ST28.
You. The first processing block in step ST27
The period is set by performing the following four items in order.
Will be performed. (1) Initial setting for loop processing. Toes
And the band obtained by dividing the deformed outer frame 12 into columns is further divided into rows.
End point of the first block in the block processing
B00 (upper left), B10 (upper right) shown in FIG.
Is set. (2) End point coordinate value of minimum unit 17 for deformation processing
(Real numbers: B00, B01, B11, B10).
That is, the deformed outer frame 12 (float mesh coordinates 15)
Area within the desired four pixels (B00, B01, B
11, B10) (B00-B0)
1-B11-B10) The minimum unit 17 of the transformation process
Set the coordinate values of four pixels to be (block)
You. (3) Straight line in the straight line direction (B00-B01)
Calculation of the number of continuous contour pixels (cpxl) (FIG. 10). This
Here, the calculation of the number of continuous contour pixels (cpxl) refers to a straight line
Is calculated. How to calculate the distance of a straight line
This is the same as described above. (4) Straight line in the straight line direction (B10-B11)
Calculation of the number of continuous contour pixels (dpxl) (FIG. 10). This
Here, the calculation of the number of continuous contour pixels (dpxl) means a straight line
Is calculated. How to calculate the distance of a straight line
This is the same as described above. Next, in step ST28, cp
xl is longer than the distance of dpxl (cpxl
> Dpxl? ) And NO (the distance of cpxl is dp
xl), the next step ST
29a, and in step ST29a, the longer one
Is greater than a predetermined threshold value.
Whether it is small (smaller than the preset width of block 17)
To determine if the processing is inconsistent) and determine NO (from the threshold
In the case of (large), the process proceeds to the next step ST31,
If the determination in step ST29a is YES (less than the threshold value)
I), the process proceeds to the next step ST30a,
Is the final processing unit in step ST30a, and
Judge whether it is not 1 point, YES (final processing unit)
In the case of, the process proceeds to the next step ST31, and
If it is not the final processing unit), return to step ST26.
The determination in step ST29a is NO, or
It repeats until the judgment of step ST30a becomes YES. If the determination in step ST28 is YES (c
pxl distance is greater than dpxl distance)
Proceeds to the next step ST29b, and proceeds to step ST29b.
29b, the longer cpxl distance is preset.
Whether the threshold value is smaller than a predetermined threshold value.
Is smaller than the width of the lock 17 and the processing is incompatible)
If NO (greater than the threshold), the next step
Proceeds to step ST31, and the determination in step ST29b is Y
In the case of ES (less than the threshold), the next step
The process proceeds to step ST30b, and the
Judge whether it is the final processing unit and whether it is not 1 point,
If YES (final processing unit), the next step ST
31 and if NO (not the final processing unit)
Returning to step ST26, the judgment of step ST29b is
No, or the determination in step ST30b is YES.
Repeat until That is, from step ST28 to step
By ST30 (30a, 30b), the distance of cpxl
Until both the distances dpxl and dpxl are greater than the threshold.
The processing is repeated and the minimum unit of deformation processing 17 (block)
Width (size) is determined. Next, in step ST31, drawing
/ Print output processing is performed and the process proceeds to the next step ST32.
You. Drawing / Printing Output in Step ST31
The processing is accomplished by sequentially performing the following five items.
Is performed. (1) Obtaining an image density value. In other words, the transformation
Pixels that form the logical minimum unit 17 (block)
(B00, B01, B11, B10)
Original image density value (image density) corresponding to pixel (B00)
Value: G00) is defined as the density value of the entire block. (2) A rectangle surrounding the minimum unit 17 for deformation processing
Calculation of (fmax, fmin). Here, the transformation processing minimum
Calculation of rectangle (fmax, fmin) surrounding unit 17
Is a square (B00-B01-
B11-B10) to form a rectangular square.
is there. (3) Addition of rounding error recovery value and painting
Determine the fixed integer rectangular area. This forms block 17
Surround the square (B00-B01-B11-B10)
The four end points of the rectangular quadrangle form the deformed outline frame 12
Coordinates on float mesh coordinates 15 (real coordinates)
And a square (B00-B0) as shown in FIG.
1-B11-B10).
-Q2-Q3) to form integer float mesh coordinates
15 and are continuously developed without gaps. (4) Integer rectangular area is painted with density value G00
Crush This is the integer flow that forms the deformed outer frame 12.
4 expanded to coordinates on the port mesh coordinates 15 (real coordinates)
A rectangle surrounding the square (B00-B01-B11-B10)
The density value of the original image is inside the square (Q0-Q1-Q2-Q3)
(Image density value: G00). (5) Updating of vertical coordinate values (B00 <-B0)
1, B10 <-B11). This is the next block 17
Block 17 in order to perform the processing of
Block B01 and B11 to be processed next.
To be replaced with the upper end points B00 and B10 of step 17
It is. Next, in step ST32, the transformation
Updating the coordinate values in the upper direction (P00 <-P10, P01 <-
P11), and return to step ST21 for all the processing.
From step ST21 to step ST32 until
Is to be repeated. The upper part of the deformation direction in step ST32
Update coordinate value (P00 <-P10, P01 <-P11)
Will be filled in for the next band.
The right end points P10 and P11 of the selected band are processed next.
To the left end points P00 and P01 of the band
Things. As described above,Each on the deformed outline frame 12
Each pixel has the phase of each pixel on the reference outline 11.
Because we can easily reflect each other's positional relationship,Base
The relative position of the original image data to the quasi-outer frame 11 is maintained.
The original image data into the desired shape.
Can be formed by pixels such as photos
Original image data can be circular, elliptical, sector, square
Freely transform into shapes, strips, cylinders, arches, etc.
Possible and excellent image quality after deformation
It can be. A modification which is a modification of the reference outer frame 11 is shown.
As shown in FIG. 22, the shape of the external shape frame 12 is trapezoidal, circular,
Default deformation of shape, cylinder or arch, fan, etc.
Shape, and change this default deformed shape and trapezoid
Shape, deform ellipses from circles, cylindrical or arched
Change the shape of the arch to the depth of the arch, or
Default deformation such as arbitrary deformation by int operation
It is preferable to use a modified deformed shape whose shape has been modified. In addition, color image editing
FIG. 23 shows a modified example. The original picture used in the present invention is as described above.
Image information (density value) of depth as image data
Floor consisting of a set of points (pixels)
Not only black and white or color photos with tones
Image data of original images that can be binarized such as characters and figures
(Vector data or dot data)
Can be. That is, from the binary image image,
Support a very wide range of original images up to the color image
be able to. FIG. 2 shows a modification of the binary image image.
It is shown in FIG. The present invention is limited to the above embodiment.
It is not a thing and can be changed as needed. [0121] As described above, the image data of the present invention
EditionGatheringAccording to the method, the reference outer frame is deformed into a desired shape.
The original image data for the reference outline frame.Ofrelative
Deform the original image data while maintaining the positional relationship.
From the binary image,
Original image data consisting of color images
Fan shape, square shape, strip shape, cylindrical shape, arch shape
It can be freely deformed as well as the image of the deformed image
Extremely good that the quality can be excellent
It works.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a block diagram showing a configuration of a main part of an image data editing apparatus according to the present invention. FIG. 2 is a flowchart showing image data editing according to the present invention. is an explanatory view illustrating a combination of a line segment of time variations, a is shown to monochrome photographs of an example of an original image
True, b is shown to diagram the reference outline frame, c is the reference longitudinal deformation combining indicate to diagrams of a line segment when the outline frame, d is shown to diagram the deformation outline frame after longitudinal deformation, e is a diagram showing an intermediate image example of displaying an image longitudinal deformation on indicates to display,
f indicates a combination of line segments when the reference outer frame is deformed in the horizontal direction.
To the diagram, g is shown to diagram the deformation outline frame after lateral deformation, h
Is displayed on the display showing the image deformed in the horizontal direction .
FIG . 4 is a diagram illustrating an example of an intermediate image. FIG . 4 is an explanatory diagram illustrating vertical deformation.
Denotes a pixel stored in the reference outline frame, b denotes a pixel on the reference outline frame, c denotes a pixel position on the reference outline frame, d denotes a pixel position on the parameter coordinates, and e denotes a deformation. The outline frame is shown, f indicates the position of the pixel on the deformation parameter coordinates, and g indicates the position of the pixel converted on the deformation outline frame. FIG. 5 is an explanatory diagram illustrating a calculation method of an intermediate curve. FIG. 6 is a schematic diagram illustrating column division. FIG. 7 is a schematic diagram illustrating row division. FIG. 8 is a schematic diagram illustrating blocks. FIG. 10 is an explanatory diagram for explaining setting of a block. FIG. 11 is an explanatory diagram for explaining a block. FIG. 12 is an explanatory diagram for explaining a rectangular rectangle surrounding a block. FIG. 14 is an explanatory diagram illustrating the filling of the outer frame. FIG. 14 is a flowchart illustrating the algorithm of the image data editing method according to the present invention divided into two parts. FIG. 15 is a flowchart illustrating the algorithm of the image data editing method according to the present invention. FIG. 16 is a schematic diagram illustrating a cubic Bezier function and its parameters. FIG. 17 is an explanatory diagram illustrating a cubic Bezier function. FIG. 18 is a diagram illustrating a method of calculating a distance of a curve. FIG. 19 is an explanatory diagram illustrating an example of a method of calculating a distance of a curve. FIG. 20 is an explanatory diagram illustrating another example of a method of calculating a distance of a curve. FIG. 21 is a diagram illustrating a linear cubic Bezier function. FIG. 22 is a diagram showing examples of shapes of a reference outer frame and a deformed outer frame. FIG. 23 is a diagram showing an example of an original image according to the method of the present invention.
Displayed on the display showing the black-and-white photo and its modification
Description Figure [a code indicating the image of FIG. FIG. 25 after deformation using a conventional projection deformation showing a modified example of FIG. 24 shows images of the binary according to the present invention a method of indicating an intermediate image example 1 image data editing device 2 original image data reading means 3 parameter coordinate forming means 4 deformed outline frame shape reading means 5 deformation parameter coordinate forming means 6 real coordinate calculating means 7 image deformation drawing unit setting means 8 original image density value obtaining means 9 Image rendering means 10 Processing accuracy setting unit 11 Reference outer frame 12 Deformed outer frame 13 Integer mesh coordinates (xy coordinates) 14 Parameter coordinates 15 Integer float mesh coordinates (real coordinates) 16 Deformation parameter coordinates 17 Deformation processing minimum unit ( block)

────────────────────────────────────────────────── ─── Continuation of the front page (56) References JP-A-62-206554 (JP, A) JP-A-2-78368 (JP, A) JP-A-3-196187 (JP, A) JP-A-4-206 373084 (JP, A) (58) Fields investigated (Int. Cl. ⁷ , DB name) H04N 1/387 G06T 3/00

Claims (1)

(57) [Claims] [Claim 1] Original image data as image information of an original image
Is a set of pixels that can be processed by a computer
These pixels are referred to as coordinate axes with a substantially rectangular reference outline frame.
Stored in the reference outer frame as integer mesh coordinates
And the coordinate values of each pixel stored in this reference
The maximum distance for each coordinate axis from the origin of the coordinate axes of the reference outer frame
Parameter where the coordinate value of the pixel at the position
And convert each pixel on the mesh coordinates
Expanded into a substantially rectangular parameter coordinate where the maximum value of the coordinate axis is 1.
Open, the reference outline frame can be described by at least a function
Deformed into a deformed outer frame of a desired shape
Is the actual coordinate with the coordinate axis, and this actual coordinate is
Coordinate value of the position farthest away from the origin of each coordinate axis for each coordinate axis
Are the deformation parameter coordinates each of which is 1, and then
The parameter coordinate value of each pixel is transformed parameter
The coordinates are transformed into deformation parameter coordinates on
Xel on the real coordinates with integer float mesh coordinates
And expand the area within the real coordinates
Divided into blocks as the minimum unit for shape processing, and
One of the pixels located in that block
The image density value of the original image of the pixel of
Area for each rectangle on the real coordinates surrounding the block
Is painted with the image density value of the obtained original image.
An image data editing method, characterized in that: