JP2003319165A - Image composition method and system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image composition method and system for applying composition processing to a composited part without losing a feature of both images in the case of compositing parts of the images, in particular, performing the composition required for processing of the joints of the images. <P>SOLUTION: An original image being an object of composition processing is received (S1). After setting an overlap amount (S2), an overlap area is prepared in a part on which two original images are overlapped (S3), and the occurrence probability of pixels of each original image is calculated on the basis of an interior division ratio of each pixel sequence in the overlap area to a distance of the border of the overlap area (S4). The composited image can be obtained by deciding (S5) which of pixels of the original images is to be used as a pixel in each pixel sequence according to the occurrence probability. <P>COPYRIGHT: (C)2004,JPO

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a technique for synthesizing a part of an image having a pattern or the like for natural synthesizing without damaging high frequency components.

[0002]

2. Description of the Related Art Conventionally, when a new image is created based on a plurality of images, image synthesizing processing is performed. This is done by determining the pixel values of the new image based on the pixel values of the multiple images. There are various rules for determining the pixel, and the most suitable rule is used according to the characteristics of the image and the purpose for which the image is used.

Further, even if it is said that the images are combined in one bit, when the whole image is combined in order to obtain a completely new image,
When processing an image, only a part may be combined as needed. As an example of the latter, there is a synthesizing process at a joint between images used in a building material printed matter. here,
The building material printed matter and the image used for the building material printed matter will be described. Building material prints are much longer (longer) than general-purpose prints such as publication prints and commercial prints. For example, considering a printed material to decorate the wall surface, the length from the floor to the ceiling of the building is usually about 2.5 m,
In addition, the horizontal length from the end to the end of the room is about 3.5 m in the case of 6 tatami mats. Therefore, the images for printed materials for building materials are also required to be endless images, that is, a continuous continuous image of at least that length, that is, an endless image.

The long image as described above is printed on the printing cylinder 1
It is impossible to fit within the circumference. Therefore, the side of the start end and the side of the end of the image for one circumference in the printing direction are connected (connected) on the plate cylinder to perform plate making. Then, each time the plate cylinder makes one rotation during printing, an image for one circumference is repeatedly printed without interruption, thereby obtaining a long printed matter of an arbitrary length.

However, if the side of the start end and the side of the end in the printing direction of the image for one circumference are simply butted and connected, the image becomes discontinuous at the connecting portion (joint), and the endless image is obtained. Can't get In order to make the plate cylinder image endless, it is necessary to make the discontinuity of both images continuous at the seam of the image and make the seam unrecognizable, that is, seamless. Becomes

Therefore, as a method of making the plate cylinder image endless by seamlessizing the image joints, which is particularly effective in the case of a wood grain pattern, as described in Japanese Patent Publication No. 43-2065, etc. As a wavy line or zigzag broken line, the joint itself is made inconspicuous and the area near the joint is manually added, decolorized, dyed, deleted, etc. by a skilled worker to correct the discontinuity of the image and seamless Has been widely used (prior art).

Further, in recent years, due to the progress of digital image processing technology applying an electronic computer, a document image is read by a scanner to be converted into digital data, and a correction portion is designated by a mouse, a tablet, a cursor or the like on a display device such as a CRT. It has been attempted to improve workability by a technique such as copying and transplanting pixels in an appropriate area from another portion of the screen (conventional technology). Furthermore, a method has also been proposed in which the work of seamlessening the joint is made efficient by automatically calculating it by a predetermined formula by averaging calculation processing of both-sided images (prior art).

[0008]

However, in the above-mentioned prior art, since the seamless operation is performed manually by a skilled worker, the work efficiency is low and it has become difficult to secure a skilled worker in recent years. Moreover, it is also difficult to reproduce a completely seamless plate when reprinting.

In the above-mentioned prior art, by adopting the digitized and mechanized processing for the work of the seamless operation, the work efficiency can be improved in the steps such as the joint of the images as compared with the above-mentioned prior art. . However, the seamlessness work itself has not been improved much in that it depends on the intuition and manual work of a skilled worker.

In the above-mentioned prior art, the efficiency has been improved by automating the work of making it seamless by digital processing. However, since the seamlessization simply adds (averages) the images on both sides at the joint, the high frequency component of the spatial frequency is lost near the joint, and the fine structure or sharpness of the image is reduced. Therefore, the problem remains that the joint is still recognized when viewed from a distance.

In view of the above points, according to the present invention, when synthesizing a part of an image, particularly, synthesizing necessary for processing a joint portion of the image, the characteristics of both images are impaired in the synthesizing part. An object of the present invention is to provide an image synthesizing method and system capable of performing such a synthesizing process.

[0012]

In order to solve the above-mentioned problems, according to the present invention, a method of combining a plurality of images by superimposing a part of the plurality of images and giving a new pixel value Inputting an image, setting an overlap amount of the input plurality of images, creating an overlap region based on the set overlap amount, and inputting the overlap region in the overlap region. Perform an image modification step that modifies the image by giving the pixel value of either image
It is characterized in that in the image correction step, the determination of which image pixel value is to be given is made based on the distance of the pixel to which a new value is given from both the images.

According to the present invention, when a new pixel value is given to each pixel of the overlap area which is an overlapping portion of a plurality of images, some correction is not performed on the pixel values of both original images, but either of them is corrected. Since the pixel values of the pixels are used as they are, it is possible to combine the images having the characteristics of the original two images without damaging the high frequency components of the images. At that time, which image pixel is used is determined according to the appearance probability based on the internal division ratio of the distance from the original image. Therefore, when the image is viewed as a whole, the image is smooth in the overlap region. Changes to. Further, when a plurality of the same images are lined up and used for synthesizing the joint portion, an endless image in which the joints are not conspicuous can be created.

[0014]

DETAILED DESCRIPTION OF THE INVENTION An embodiment of the present invention will be described in detail below with reference to the drawings. (1. System Configuration) First, the system configuration of the image composition system according to the present invention will be described. FIG. 1 is a system configuration diagram showing an embodiment of an image composition system. In FIG. 1, 1 is an original image input means, 2 is an overlap amount setting means, 3 is an arithmetic control device, 3a is an overlap area creating means, 3b is an appearance probability calculating means, 3c is a pixel determining means, 4 is an image output. It is a means.

The original image input means 1 is for inputting an original image to be a subject of the combining process, and is realized by reading the original image data with a digital data reading device. When the original image is drawn on a paper medium or the like, it is input after being digitized by using a scanner, a digital camera or the like. The overlap amount setting means 2 is for setting the extent to which a plurality of images (or the end portions facing the end portions of one image) are overlapped, that is, overlapped with each other. Realized with equipment.

The arithmetic and control unit 3 plays a central role in the present invention, and synthesizes the overlapping regions of a plurality of images by synthesizing them based on the input original image and the set overlap amount. It has the function of creating images. Actually, it is configured by a CPU of a computer and a memory, a dedicated program is read into the memory, and the CPU sequentially executes the program. The arithmetic and control unit 3 further includes an overlap area creating unit 3a, an appearance probability calculating unit 3b, and a pixel determining unit 3c. The overlap area creating means 3a has a function of calculating the size of the overlap area based on the size of the input original image and the set overlap amount, and ensuring a memory space of the calculated size. ing. Appearance probability calculation means 3
b has a function of determining how far each pixel in the overlap region is from both ends of the overlap region by the ratio. The pixel determining means 3c has a function of determining which image pixel is to be adopted as each pixel on the overlap area based on the calculated internal division ratio. The overlapping area creating means 3a, the appearance probability calculating means 3b, and the pixel determining means 3c are realized by executing their own dedicated programs by the CPU. The image output unit 4 has a function of outputting a combined image obtained by the combining process, and has an external storage such as a display device for displaying the image, a printer for printing, a hard disk for outputting as data, or the like. A device, a recording device for a portable electronic recording medium, a communication device for transmitting to a remote place via a network, and the like are applicable.

(2. Process Flow) Next, the image synthesizing method according to the present invention will be described together with the processing operation of the image synthesizing system shown in FIG. FIG. 2 is a flowchart showing an outline of the image synthesizing method according to the present invention. Here, a case will be described as an example in which, when a plurality of one images are arranged, a process of making an endless image so that the joints are not noticeable is performed. First, the original image input means 1 inputs an original image to be subjected to endless processing (step S1).
Next, the overlap amount setting means 2 is used to set the overlap amount (step S2). The overlap amount is an amount by which the end portions of the original image, which are opposed to the end portions, overlap each other, and the overlap region is determined based on this overlap amount. The overlap area is an area in which an edge portion of the original image facing the edge portion is combined. If this synthesizing process is not performed, when a plurality of original images are arranged, a step becomes conspicuous at the boundary between them.
In the present embodiment, the overlap amount is set by the number of pixels in which the end portions of the original image facing each other in one direction overlap each other.

Then, an overlap area is created based on the size of the input original image and the set overlap amount (step S3). Specifically, the overlap area creating means 3a secures a memory space required as an overlap area in the main storage device of the computer. For example, _{assume that} the size (the number of pixels) of the original image G _ORI is horizontal xp x vertical yp. At this time,
As shown in (b), when a plurality of original images G _ORI are arranged in the x-axis direction (horizontal direction in the figure), the length (number of pixels) of the side in contact with the adjacent original image G _ORI is yp. Assuming that the overlap amount is α, an overlap region O having a size of yp × α is created. Only one original image G _ORI is processed in the present embodiment, and the left and right images when the plural images are arranged as shown in FIG. 3B are the same original image G _ORI , but here, for convenience of explanation. , One is the original image G _L , and the other is the original image G _R. As shown in FIG. 3B, an overlapping area having a size of yp × α is created across the original image G _L and the original image G _R.

After the overlap area is created, next, the appearance probability of the pixel of each original image in the pixel row, which is a set of pixels in the overlap area, is calculated (step S4). The appearance probability of the pixel of each original image is a ratio indicating the degree of appearance of the pixel of which image in each pixel column. This appearance probability is calculated based on the internal division ratio of the distance between each pixel column and the boundary of the overlap region. How the appearance probability is calculated will be described with reference to FIG. FIG. 4 shows the vicinity of the overlap area O of the original image G _L and the original image G _R when the overlap amount α = 6. In the example of FIG. 4, since α = 6, x is included in the overlap area.
Six pixels are prepared in the axial direction (left and right direction in the figure). When a set of pixels in the y-axis direction is defined as a pixel column, the pixel array shown in FIG.
In the overlap region O shown, in order from the original image G _L side, first row, so that the second row, six pixel columns of ... up to the sixth column exists. In such an overlap region, the appearance probabilities P _L and P _R of the original images G _L and G _R in the pixel column of the x _{o th} column are calculated by the following [Formula 1].

[0020] [Equation 1] _{P L = (d - x 0} ) / d P R = x 0 / d

In the above [Formula 1], the occurrence probability P _L ,
P _R is the probability of adopting the pixels of the original image G _{L and} the pixels of the original image G _R , respectively. Further, x _o is 1 ≦ x _o ≦ d−1.
Takes an integer value in the range. In the example shown in FIG. 4, d (= α +
1) = 7, and x _o = 1-6. That is, in the example of FIG. 4, the appearance probability P _L is 6/7 in the first column, 5/7 in the second column, 4/7 in the third column, and 4th in the third column.
The third column is 3/7, the fifth column is 2/7, the sixth column is 1/7, and the appearance probability P _R is 1/7 in the first column and 2 in the second column.
/ 7, the third row is 3/7, the fourth row is 4/7, the fifth row is 5/7, and the sixth row is 6/7.

Then, based on the calculated appearance probability, it is determined which pixel of the two images is to be used for each pixel in the overlap area (step S5). Specifically, the appearance probability P is set for each column of the overlap region.
Each pixel is determined so as to satisfy _L and the appearance probability P _R. For example, in the case of the first column shown in FIG. 4, the occurrence probability P _L
Is 6/7 and the appearance probability P _R is 1/7, the pixels are determined so as to satisfy this probability. That is, FIG.
If there is a pixel in the yp row as shown in, in the first column,
The pixels of the original image G _L are yp × 6/7, and the pixels of the image G _R are y
p × 1/7 pieces will be selected. 2nd row to 6th
The columns up to the row are also determined based on the calculated appearance probability.

Pixels are determined for each column by determining the appearance probability P.
_L and appearance probability P _R are satisfied,
Various methods can be used as this method.
In this embodiment, a uniformly distributed random number having a value of 0 to 1 is generated for each pixel, and if it is less than P _L , the pixel value at the corresponding position is copied from the original image G _L , and if it is P _L or more. For example, the pixel value at the corresponding position is copied from the original image G _R. As a result, the pixels shown in FIG. 5 are assigned. In FIG. 5, the pixel indicated by "L" is the original image G _L.
The pixels indicated by “R” indicate that the pixels of the original image G _R are given. When the composite image is obtained by thus correcting the pixel values in the overlap area, the image output means 4 outputs the composite image (step S6).

When the synthesizing process is performed to make one original image endless as in the above example, the overlap area O becomes the same as the area O'as shown in FIG. 6 (a). . Therefore, actually, the overlap area O on one side (for example, the right side) of the original image is deleted, and the pixel value of the overlap area O is copied to the corresponding left overlap area O ′. Specifically, as shown in FIG. 6B, a combined image having a size in which one overlap area is deleted can be obtained.

(3. Application Example) In the above example, when the original images are arranged in the x-axis direction, the images are corrected by synthesizing the two images in the overlap area so that the images become endless images. However, even when a plurality of arrays are arranged in the y-axis direction, only the x-axis and the y-axis are exchanged,
Essentially the same process is performed. Also, x-axis direction, y
When arranging images in both the axial directions, the x-axis direction, y
By performing the image combining processing in both the axial directions, it is possible to obtain a combined image in which all the joints are inconspicuous even when the images are arranged in two dimensions in the vertical and horizontal directions.

In the above example, the case where a plurality of the same images are lined up has been described. However, a plurality of images having the same side length (number of pixels) to be connected are sequentially connected so that the joints are not conspicuous, and the left end is connected. It is also possible to create a seamless image by similarly performing the combining process on the left end portion of the image and the right end portion of the right end image in the overlapping area. For example, in FIG.
As shown in (c), as the original image G _R and the original image G _L , an image having a size of horizontal x _L x vertical y _{L and} an image of horizontal x _R x vertical y _L are applied to perform the combining process. . In this case, the sides to be combined have the same length y _L , but may be different in the lateral direction. In the case of the example shown in FIG. 6C, the size of the combined image is horizontal (x _L + x _R −α) × vertical y _L.

Further, the present invention can be applied to a process of replacing an inappropriate portion of an image with another small image.
In particular, a pattern image used for decorating the surface of a building material product or the like is obtained by transforming a photographed image or artificially created by computer calculation. However, at the time of initial creation of such a pattern image, an unfavorable portion may often occur. In such a case, the unfavorable portion is replaced with another small image, but since it is another image originally, the pixel values of both images are greatly different at the boundary portion around the replaced image. For this reason, the replaced portion is clearly visible, and in some cases, the replaced portion may result in an unfavorable image. The present invention is applied to prevent this.

The processing in this case will be specifically described. First, the original image G _{OR I} to be corrected and the patch image G _PAT which is a small image are input. Then, a position on the original image G _ORI to be corrected, that is, a position to be replaced with the patch image G _PAT is designated. Then, the overlap region is created as in the above embodiment. However, the difference from the above embodiment is that all four sides of the patch image G _PAT are in contact with the original image G _ORI . Therefore, the overlap area is also created corresponding to the four sides. Here, the relationship between the original image G _ORI , the patch image G _PAT , and the overlap area is shown in FIG. 7A, and the details of the overlap area are shown in FIG. 7B. If the number of pixels in the x direction of the patch image is xs, the number of pixels in the y direction is ys, and the overlap amount is α, there are two regions of size xs × α,
A total of four overlapping areas with two areas having a size of (ys-2α) × α are created. As shown in FIG. 7B, the overlap region O having a size of (ys-2α) × α
_A, O _C, overlap size of xs × alpha region O _B,
O _D will be obtained. By performing the processing of step S4 and step S5 of the above embodiment on this overlapping area, a composite image can be obtained. However, the overlap region O _A, the O _C is may be performed as it is, the overlap area O _B, the O _D, it is necessary to perform processing interchanged x-axis and y-axis directions.

[0029]

As described above, according to the present invention,
Input multiple images to be composited, set the overlap amount of the input multiple images, create an overlap area based on the set overlap amount. The image is modified by giving the pixel values of that image. When modifying this image, the decision of which image pixel value to give is
Since the decision is made based on the distance of the pixel to which a new value is given from both images, it is possible to combine the images having the characteristics of the original images without damaging the high frequency components of the images. Play. Further, when a plurality of the same images are arranged, if they are used for composition of the joint portion, it is possible to create an endless image in which the joints are inconspicuous.

[Brief description of drawings]

FIG. 1 is a configuration diagram of an image composition system according to the present invention.

FIG. 2 is a flowchart of an image synthesizing method according to the present invention.

3 is a diagram showing the relationship between the overlap region O of the case of arranging plural original images G _ORI, and the original image G _ORI.

FIG. 4 is a diagram showing the vicinity of an overlap region O.

FIG. 5 is a diagram showing the vicinity of an overlap area O after pixel determination.

FIG. 6 is a diagram showing a state of an image after a combining process.

FIG. 7 is a diagram for explaining an application example of the present invention.

[Explanation of symbols]

1. Original image input means 2 ... Overlap amount setting means 3 ... Arithmetic control device 3a ... Overlap area creating means 3b ... Appearance probability calculation means 3c ... Pixel determining means 4 ... Image output means

Claims (5)

[Claims] 1. A method of synthesizing images by superimposing a part of a plurality of images and giving new pixel values, the method comprising: inputting a plurality of images to be synthesized; Setting an overlap amount of the image, creating an overlap region based on the set overlap amount, and giving a pixel value of either of the input images in the overlap region. An image modification step of modifying the image by means of which the determination of which image pixel value to provide in said image modification step is based on the distance of the newly valued pixel from both images. An image synthesizing method characterized by: 2. The image correcting step calculates a probability of appearance of a pixel of each image for each column of the created overlap area based on an internal division ratio of distances from both ends of the overlap area. The appearance probability calculation step and the pixel determination step of allocating pixels from both images to pixels in each column in the overlap area so that the appearance probabilities of the respective images are almost the same. Characteristic image composition method. 3. A method for synthesizing a part of an image so that when the edges of the same image and the edges facing each other are connected, the edges of the image are combined, and an image to be combined is input. An image input step, a step of setting an overlap amount in a joint portion of the input image, a step of creating an overlap region based on the set overlap amount, and each of the overlap regions. For a column, an appearance probability calculating step of calculating a probability of appearance of a pixel in each image based on an internal division ratio of distances from both ends of the overlap region, and both images to pixels in each column in the overlap region. And a pixel determination step of performing the allocation from 1 to approximately the same as the appearance probability of each image, and the image combining method. 4. An original image input unit for inputting an original image to be combined, an overlap amount setting unit for setting an overlap amount of the input original image, and an overlap amount setting unit based on the set overlap amount. Overlap area creating means for creating an overlap area, and for each column in the created overlap area,
Appearance probability calculating means for calculating the appearance probability of pixels of both original images, and pixels for determining the values of the pixels in each column so as to have a ratio substantially the same as the appearance probability based on the calculated appearance probability An image synthesizing system comprising: a deciding unit; and an image outputting unit that outputs a synthetic image obtained by deciding each pixel. 5. A step of inputting a plurality of images to be combined into a computer, a step of setting an overlap amount of the input plurality of images, and an overlap area based on the set overlap amount. A step of creating, in the overlap area, an image correction step of correcting an image by giving a pixel value of either of the input images is executed, and a pixel value of either image is given in the image correction step. The program is characterized in that the determination is made on the basis of the distances of pixels to be newly given a value from the both images.