KR20110075110A - Image size converting apparatus - Google Patents

Abstract

PURPOSE: A device for converting an image size is provided to provide an image converted according to the taste of each user. CONSTITUTION: A source image is inputted(S1210). The device is combined of an image cropping operator, an image scaling operator, and/or modified seam carving operator. The inputted source image is transformed through the combination. At least one temporary image is generated through the conversion(S1220). In a temporal image, a target image is selected(S1230).

Description

Image size converter {IMAGE SIZE CONVERTING APPARATUS}

The disclosed technique relates to an image size converting apparatus, and more particularly, to an apparatus for converting an image size.

In addition to large display devices such as TVs and PCs, personal terminals such as PDAs and mobile phones can display images on the screen. Various display devices display images on the screen in different aspect ratios or resolutions, depending on the application. Therefore, image size conversion methods for generating an image suitable for a display device have been studied.

Among the embodiments, the image size converting apparatus includes a display device, a user interface unit, a memory for storing a source image, and a control unit for controlling the image converting unit for converting the source image. The image converting unit receives a source image, converts the received source image by combining an image cropping operator, an image scaling operator, and / or a modified seam carving operator to convert at least one input image. Generate a temporary image of the modified seam operator that determines the direction in which the change in pixel value is greatest at the boundaries of objects contained in the image, and at a predetermined distance in the determined direction. Corresponding to an operator removing or adding a seam based on the rate of change between pixels-selecting a target image from the at least one temporary image.

Among the embodiments, the image size conversion method comprises (a) receiving a source image, (b) an image cropping operator, an image scaling operator and / or a modified seam carving operator Combining the input source images to generate at least one temporary image, wherein the modified shim carving operator determines a direction in which a change in pixel value is greatest at a boundary of objects included in the image. Determine and correspond to an operator that removes or adds a seam based on a rate of change between pixels at a predetermined distance in the determined direction— and (c) selecting a target image from the at least one temporary image. It includes.

The description of the disclosed technique is merely an example for structural or functional explanation and the scope of the disclosed technology should not be construed as being limited by the embodiments described in the text. That is, the embodiments may be variously modified and may have various forms, and thus the scope of the disclosed technology should be understood to include equivalents capable of realizing the technical idea.

On the other hand, the meaning of the terms described in the present application should be understood as follows.

The terms " first ", " second ", and the like are used to distinguish one element from another and should not be limited by these terms. For example, the first component may be named a second component, and similarly, the second component may also be named a first component.

The term “and / or” should be understood to include all combinations that can be suggested from one or more related items. For example, the meaning of “first item, second item and / or third item” may be given from two or more of the first, second or third items as well as the first, second or third items. Any combination of the possible items.

When a component is referred to as being "connected" to another component, it should be understood that there may be other components in between, although it may be directly connected to the other component. On the other hand, when a component is said to be "directly connected" to another component, it should be understood that there is no other component in between. On the other hand, other expressions describing the relationship between the components, such as "between" and "immediately between" or "neighboring to" and "directly neighboring to", should be interpreted as well.

A singular expression should be understood to include a plurality of expressions unless the context clearly indicates otherwise, and the terms "comprise" or "having" include features, numbers, steps, actions, components, parts, or combinations thereof. It is to be understood that this does not preclude the existence or addition of one or more other features or numbers, steps, operations, components, parts or combinations thereof.

For each step, the identifiers (e.g., a, b, c, ...) are used for convenience of description, and the identifiers do not describe the order of the steps, and each step is clearly contextual. Unless stated in a specific order, it may occur differently from the stated order. That is, each step may occur in the same order as specified, may be performed substantially simultaneously, or may be performed in the reverse order.

All terms used herein have the same meaning as commonly understood by one of ordinary skill in the art unless otherwise defined. Terms defined in commonly used dictionaries should be interpreted to be consistent with meaning in the context of the relevant art and can not be construed as having ideal or overly formal meaning unless expressly defined in the present application.

1 is a block diagram illustrating an image size converting apparatus according to an embodiment of the disclosed technology.

Referring to FIG. 1, the image size converting apparatus 100 may include a memory 110, a display 120, a user interface 130, a controller 140, and an image converter 150.

The memory 110 stores a source image, the display 120 displays an image, and the user interface 130 receives a command from a user. The controller 140 controls the memory 110, the display 120, and the user interface 130. The controller 140 loads the image converter 150 and converts the image through the image converter 150.

FIG. 2 is a diagram for describing an image conversion control screen of the image conversion unit of FIG. 1.

Referring to FIG. 2, the image conversion control screen 210 may include a source image opening input unit 220, an image conversion input unit 230, a target image size input unit 240, and a source image display unit 250.

When the controller 140 loads the image converter 150, the image converter 150 displays the image conversion control screen 210 on the display 120. When the open source image input unit 220 is selected, the image converter 150 loads a source image for converting the size from the memory 110 and displays the loaded source image on the source image display unit 250. The source image for converting the size may be selected by the user.

The image converter 150 receives the size of the target image through the target image size inputter 240. In one embodiment, the size of the target image may correspond to the number of pixels of the target image. In one embodiment, the image converter 150 may directly receive the size of the target image, that is, the number of pixels, and in another embodiment, may select one of the preset example sizes.

If the image conversion input unit 230 is selected after the sizes of the source image and the target image are selected, the image conversion unit 150 performs an image conversion process.

Hereinafter, a process of converting the source image through the image converter 150 will be described.

Since the characteristics are different for each image, it is difficult to obtain a natural reduced or expanded image when the same image converter is applied to each image. Accordingly, the image converter 150 arbitrarily combines the operators to generate the first operator combination, and generates the first target image based on the generated first operator combination. The image converter 150 displays the generated first target image on the display 120 and receives evaluation information about the first target image.

The image converter 150 converts the source image by combining an image cropping operator, an image scaling operator, and / or a modified seam carving operator to generate at least one temporary image. Select the target image from at least one temporary image. The modified seam carving operator determines the direction in which the change in the pixel value is greatest at the boundaries of the objects contained in the source image, and based on the rate of change between the pixels at a predetermined distance in the determined direction. Convert the image by removing or adding it.

In one embodiment, the image converter 150 may randomly combine the types, order, number, etc. of the operator. The combination of operators may be represented by (k1O1, k2O2, k3O3, ...). Oi represents the type of operator and ki represents the number of pixels reduced or expanded by the operator. Thus, ∑ki is equal to the difference between the size of the source image and the size of the target image.

3 is a diagram for describing an image cropping operator and an image scaling operator.

Referring to FIG. 3, FIG. 3A illustrates an original image, FIG. 3B illustrates an image cropped through an image cropping operator, and FIG. 3C illustrates an image scaling operator. Represents the scaled image of the original image.

The image cropping operator removes the outer areas of the image to highlight some areas in the image or to convert the aspect ratio of the image. FIG. 3B is an image in which an external region is removed from the original image of FIG. 3A through an image cropping operator. The image scaling operator enlarges or reduces the size of the image. For example, an image scaling operator can magnify an image through upsampling or interpolating and can downscale an image through downsampling. FIG. 3C is a reduced image of the original image of FIG. 3A through image scaling.

The modified seam carving operator is one of the methods of content-aware image size conversion. The modified shim carving operator determines the direction in which the change in the pixel value is greatest at the boundary of objects included in the source image, and based on the rate of change between the pixels at a predetermined distance in the determined direction. Convert the image by removing or adding seam). For example, the modified shim carving operator may correspond to a flow guided shim carving operator.

4 is a view for explaining a modified seam carving operator.

Referring to FIG. 4, (a) of FIG. 4 shows a drawing for obtaining a deflection value based on a direction axis from a feature point without considering the direction of an image, and FIG. 4 (b) shows an edge tangent flow (ETF). Fig. 4 (c) shows a diagram for calculating a deflection value in the direction of a normal vector based on (b) of Fig. 4, by performing an algorithm. Indicates.

The modified seam carving operator performs an edge tangent flow (ETF) algorithm to determine the direction in which the change in pixel value is greatest at the boundaries of the objects contained in the image, and to determine the pixels at a predetermined distance in the determined direction. Find the rate of change of the liver. The edge tangent flow algorithm finds a tangent vector representing the direction in which the change in pixel value is the smallest at the boundary in the image and normalizes the tangent vector. Equation 1 below calculates a normalized tangent vector t (x) in a pixel x on an image.

t _x (x) represents the x-axis component of the tangential vector and t _y (x) represents the y-axis component of the tangential vector. FIG. 4B is a diagram illustrating a normalized tangent vector calculated through Equation 1. FIG.

Since the vertical direction of the tangential vector has the largest change in pixel value, the edge tangent flow algorithm calculates the direction derivative of the image in the vertical direction of the tangential vector to determine the rate of change (or direction) between the pixels at a given distance in that direction. According to the deflection value). Equation 2 below calculates a deflection value according to a direction.

는 방향에 따른 편향치를 나타내고 I(x)는 이미지를 나타낸다. 도 4의 (c)는 수학식 2를 통해 접선 벡터의 수직 방향으로 방향에 따른 편향치를 구하는 방법을 나타내는 도면이다.

Denotes a deflection value along the direction and I (x) denotes an image. FIG. 4C is a diagram illustrating a method of obtaining a deflection value according to a direction in a vertical direction of a tangential vector through Equation (2).

The modified shim carving operator calculates the rate of change (or energy) of the pixel based on the normalized tangent vector and the deflection value along the direction, and detects the shim connecting the pixels with the lowest rate of change. Equation 3 below calculates the rate of change (or energy) in the vertical direction in a specific pixel on the image.

는 수학식 2에서 구한 방향에 따른 편향치를 나타낸다. 아래 수학식 4는 w_v(x)를 구하는 식이다.w _v (x) represents the vertical weight according to the structure of the image,

Denotes a deflection value according to the direction obtained from Equation 2. Equation 4 below calculates w _v (x).

e _y represents the y-axis direction unit vector, i.e., (0, 1), and t (x) represents the normalized tangent vector.

The modified shim carving operator may reduce the width of the source image by detecting a vertical seam connecting the pixels having the smallest change rate based on the vertical change rate of Equation 3 and removing the seam.

Equation 5 below calculates a rate of change (or energy) in a horizontal direction in a specific pixel on an image.

는 수학식 2에서 구한 방향에 따른 편향치를 나타낸다. 아래 수학식 6은 w_h(x)를 구하는 식이다.w _h (x) represents the horizontal weight along the structure of the image,

Denotes a deflection value according to the direction obtained from Equation 2. Equation 6 below calculates w _h (x).

e _x represents the x-axis direction unit vector, that is, (1,0), and t (x) represents the normalized tangent vector.

The modified shim carving operator may reduce the height of the source image by detecting a horizontal seam connecting the pixels having the smallest change rate based on the horizontal change rate of Equation 5 and removing the seam.

FIG. 5 is a diagram for describing an image conversion control screen including a temporary image generated by the image converter of FIG. 1.

Referring to FIG. 5, the image conversion control screen 210 may include a source image open input unit 220, an image conversion input unit 230, a target image size input unit 240, a source image display unit 250, and a temporary image display unit ( 510 and the temporary image evaluation information input unit 520. 5 is a diagram when six temporary images are generated.

The source image opening input unit 220, the image conversion input unit 230, the target image size input unit 240, and the source image display unit 250 are the same as the descriptions of FIG. 2 listed above.

The temporary image display unit 510 displays the temporary images generated by the image converter 150, respectively. When six temporary images are generated, the temporary images may be displayed through the six temporary image display units 510a, 510b, 510c, 510d, 510e, and 510f.

The temporary image evaluation information input unit 520 receives evaluation information on each of the at least one temporary image displayed on the temporary image display unit 510 and determines at least one candidate image. In one embodiment, the evaluation information may correspond to an evaluation score for the temporary image displayed on the temporary image display unit 510. In another embodiment, the evaluation information may correspond to the selection / exclusion of the temporary image displayed on the temporary image display unit 510.

FIG. 6 is a diagram for describing a method of receiving evaluation information on the image conversion control screen of FIG. 5.

Referring to FIG. 6, the temporary image evaluation information input unit 520 may receive evaluation information on each of at least one temporary image displayed on the temporary image display unit 510.

After receiving evaluation information about each of the temporary images, the image converter 150 selects images having high evaluation information among the temporary images and determines the selected images as candidate images. In one embodiment, the image converter 150 may determine a predetermined number of candidate images in the order of the high evaluation score.

After selecting the candidate images, the image converter 150 obtains image conversion information from each of the at least one candidate image. The image transformation information may correspond to information about at least one operator (ie, image cropping operator, image scaling operator and / or modified seam carving operator) performed on the candidate image.

The image converter 150 generates an image by reconverting the source image based on the image conversion information. For example, the image converter 150 may generate new image conversion information by performing GA (Genetic Algorithm) operation on the image conversion information, and may reconvert the source image based on the generated new image conversion information. .

The image conversion unit 150 receives evaluation information about each of the at least one image and determines at least one candidate image until the final target image is determined, and obtains image conversion information from each of the at least one candidate image. And reconverting the source image based on the image conversion information to generate an image.

According to an embodiment, the image converter 150 may perform a GA operation using a bit string including image conversion information as genetic information.

7 is a diagram for describing a bit string including image conversion information.

Referring to FIG. 7, the bit string includes an operator number information field (NumofOp) 710, an image conversion type information field (MixorReg) 720, a size information field (reqWidth) 730 of a pixel to be reduced or expanded, an operator type. An information field (WhichOp) 740, an operator order information field (order) 750, and an operator repetition count information field (opMap) 760 are included.

The operator number information field (NumofOp) 710 includes information about the number of operators included in the operator combination, and the size of the field is 2 bytes. The image conversion type information field (MixorReg) 720 is a type (Mix) that does not know the order and number of iterations of the operator combination before the conversion, or whether the operator order of the operator combination before the conversion is a fixed type (Regular). This field contains information about the size of a field.

The size information field (reqWidth) 730 of the pixel to be reduced or expanded includes information on the size of the pixel to be reduced or expanded, that is, the difference between the size of the source image and the size of the target image, and the size of the field is 3 bytes. The operator type information field (WhichOp) 740 includes information on the type of the operator included in the operator combination, and the size of the field is 3 bytes. The operator order information field 750 includes information on the order of operators included in the operator combination, and the size of the field is 3 bytes. The operator repeat count information field (opMap) 760 includes repeat count information of an operator included in a corresponding operator combination, and the size of the field is 3 bytes.

The image converter 150 performs a GA operation using a bit string including image conversion information of the candidate image as genetic information.

8 is a diagram for describing a bit string including image transformation information about a candidate image.

FIG. 8 assumes that the image converter 150 selects three images having high evaluation information as candidate images. Referring to FIG. 8, the image converter 150 uses bit strings 810, 820, and 830 including image transformation information of three candidate images as genetic information. For example, the first bit string 810 includes image conversion information for generating the first image, the second bit string 820 includes image conversion information for generating the second image, and the third bit string. 830 includes image conversion information that generates a third image.

The image converter 150 performs a GA operation based on the first bit string 810, the second bit string 820, and the third bit string 830. The GA generates new genetic information by performing a crossover operation or a mutation operation based on the genetic information. The crossover operation is an operation for swapping information included in the genetic information based on a crossover point, and the variation operation is an operation for inverting a specific bit among the information included in the genetic information. In one embodiment, the crossover probability and the variation probability of the genetic algorithm operation may be preset.

FIG. 9 is a diagram for describing a process of performing a crossover operation based on the bit string of FIG. 8.

Referring to FIG. 9, the genetic algorithm uses information included in the genetic information as a crossover point between an image transformation type information field (MixorReg) 720 and a size information field (reqWidth) 730 of a pixel to be reduced or expanded. Swap. Genetic algorithms may exchange information located behind the crossover point.

For example, the genetic algorithm may include a size information field (reqWidth) 730 and an operator type information field (WhichOp) of a pixel to be reduced or expanded in the first bit string 810 and the second bit string 820 through a crossover operation. 740, an operator order information field 750, and an operator repeat count information field (opMap) 760 may be exchanged. The genetic algorithm performs a crossover operation based on the first bit string 810 and the second bit string 820 to generate new bit strings (or genetic information) 910 and 920.

FIG. 10 is a diagram for describing a process of performing a shift operation based on the bit string of FIG. 8.

FIG. 10 illustrates a case in which the genetic algorithm exchanges information included in genetic information using a crossover point between a size information field (reqWidth) 730 and an operator type information field (WhichOp) 740 of a pixel to be reduced or expanded. Assumed

Referring to FIG. 10, the genetic algorithm generates a new bit string (or genetic information) 1010 and 1020 by performing a crossover operation based on the first bit string 810 and the third bit string 830. . The transition operation inverts specific bits of the bit strings 1010 and 1020 generated through the crossover operation. For example, the shift operation may invert the value of the bit included in the operator order information field 750 of the newly generated bit string 1010 from 0 to 1. In one embodiment, the number of bits to perform the shift operation may vary according to the shift probability. In one embodiment, the position of the bit to perform the shift operation can be arbitrarily determined.

The genetic algorithm generates a predetermined number of new bit strings (or genetic information). In one embodiment, the genetic algorithm may generate six new bit strings. The image converter 150 generates image conversion information corresponding to the corresponding bit string based on the information included in the new bit string, and reconverts the source image based on the image transform information to generate the converted images.

FIG. 11 is a diagram for describing an image conversion control screen including an image converted through new image conversion information.

Image conversion control screen 210, source image open input unit 220, image conversion input unit 230, target image size input unit 240, source image display unit 250, target image display unit 510 and target image evaluation The information input unit 520 is the same as the description of FIG. 5 listed above.

The target image display unit 510 displays the images generated by the new image conversion information, respectively. When six images are generated through six image conversion information, the images are displayed on the six target image display units 510a, 510b, 510c, 510d, 510e, and 510f. The target image evaluation information input unit 520 receives evaluation information about the image displayed on the target image display unit 510.

The image conversion unit 150 receives evaluation information about each of the at least one image and determines at least one candidate image until the final target image is determined, and obtains image conversion information from each of the at least one candidate image. And a process of generating an image by reconverting the source image based on the image conversion information.

12 is a flowchart illustrating a process of converting an image.

The image size conversion device receives the source image (S1210), converts the source image by combining an image cropping operator, an image scaling operator and / or a modified seam carving operator to at least One temporary image is generated (S1220). The modified seam carving operator determines the direction in which the change in the pixel value is greatest at the boundaries of the objects contained in the image, and based on the rate of change between the pixels at a predetermined distance in the determined direction. Remove or add.

The image size converting apparatus selects a target image from at least one temporary image (S1230). In one embodiment, the image size converting apparatus may receive evaluation information about each of the at least one temporary image and determine at least one candidate image. The image size converting apparatus may obtain image conversion information from each of the at least one candidate image, and reconvert the source image based on the image conversion information.

The image conversion unit receives evaluation information about each of the at least one image until the final target image is determined, determining at least one candidate image, obtaining image conversion information from each of the at least one candidate image, and image conversion information. Based on the reconversion of the source image to generate an image can be repeated.

The disclosed technique may have the following effects. It is to be understood, however, that the scope of the disclosed technology is not to be construed as limited thereby, as it is not meant to imply that a particular embodiment should include all of the following effects or only the following effects.

The image size converting apparatus according to an exemplary embodiment may naturally convert an image in consideration of characteristics of the image. When converting the size of an image, an operator combination suitable for each image may be generated in consideration of the structural characteristics of the image, and a naturally resized image may be obtained through the operator combination.

The image size converting apparatus according to an embodiment may provide an image whose size is converted according to a preference of each user by reflecting a user's evaluation in an image converting process. The user's evaluation is input to the images generated by the plurality of operator combinations, and a new operator combination is generated based on the operator combination with high evaluation to convert the image, thereby providing a converted image according to each user's preference. have.

In addition, the image size conversion apparatus according to an embodiment converts the image based on the direction of the image, so that the distortion of the image can be reduced and a more natural image can be obtained when the image having the orientation is converted.

Although described above with reference to embodiments of the disclosed technology, those skilled in the art can variously modify and change the disclosed technology without departing from the spirit and scope of the technology described in the claims below. I can understand that.

1 is a block diagram illustrating an image size converting apparatus according to an embodiment of the disclosed technology.

FIG. 2 is a diagram for describing an image conversion control screen of the image conversion unit of FIG. 1.

3 is a diagram for describing an image cropping operator and an image scaling operator.

4 is a view for explaining a modified seam carving operator.

FIG. 5 is a diagram for describing an image conversion control screen including a temporary image generated by the image converter of FIG. 1.

FIG. 6 is a diagram for describing a method of receiving evaluation information on the image conversion control screen of FIG. 5.

7 is a diagram for describing a bit string including image conversion information.

8 is a diagram for describing a bit string including image transformation information about a candidate image.

FIG. 9 is a diagram for describing a process of performing a crossover operation based on the bit string of FIG. 8.

FIG. 10 is a diagram for describing a process of performing a shift operation based on the bit string of FIG. 8.

FIG. 11 is a diagram for describing an image conversion control screen including an image converted through new image conversion information.

12 is a flowchart illustrating a process of converting an image.

Claims (12)

(a) receiving a source image; (b) transforming the input source image to produce at least one temporary image by combining an image cropping operator, an image scaling operator and / or a modified seam carving operator The modified shim carving operator determines the direction in which the change in the pixel value is greatest at the boundary of objects contained in the image, and based on the rate of change between the pixels at a predetermined distance in the determined direction. corresponds to an operator removing or adding seam; And (c) selecting a target image from the at least one temporary image. The method of claim 1, wherein selecting a target image from the at least one temporary image (c1) receiving evaluation information on each of the at least one temporary image and determining the at least one candidate image; (c2) obtaining image conversion information from each of the determined at least one candidate images, wherein the image conversion information is assigned to at least one operator (ie, image cropping operator, image scaling operator and / or modification) performed on the candidate image. Corresponding to information on the seam carving operator); And and (c3) reconverting the source image based on the obtained image conversion information to determine the target image. The method of claim 2, wherein the reconverting the source image based on the obtained image conversion information to determine the target image comprises: (c31) generating new image conversion information by performing a GA (Genetic Algorithm) operation on the obtained image conversion information; And (c32) reconverting the source image based on the generated new image conversion information. The method of claim 3, wherein the image conversion information is And information on the number of operators, information on the type of image conversion, information on the size of pixels to be reduced or expanded, information on the type of the operator, information on the order of the operators, and / or information on the number of repetitions of the operators. The method of claim 2, wherein the reconverting the source image based on the obtained image conversion information to determine the target image comprises: Repeating the steps (c1) to (c3) to determine the final target image. The method of claim 1, wherein the direction of change of pixel value at the boundary of the objects included in the source image is greatest. And the change in pixel value at the boundary is a vertical direction of a tangent vector representing the smallest direction. The method of claim 1, wherein the modified shim carving operator And a rate of change between pixels at a predetermined distance in the determined direction through Equation 1 below. [Equation 1]

From here, g ° (x): rate of change or deflection along the direction I (x): source image t: tangent vector The method of claim 1, wherein the modified shim carving operator The method of claim 2, wherein the vertical change rate in the pixel is calculated through Equation 2 below, and a seam connecting the pixels having the smallest vertical change rate is detected. [Equation 2]

From here, E _v (x): Vertical change rate (or energy) at a specific pixel w _v (x): vertical weighting according to the structure, t (x): Normalized tangent vector (t (x) = [t _x (x), t _y (x)]) e _y : y-axis unit vector g ° (x): rate of change or deflection along the direction I (x): source image The method of claim 1, wherein the modified shim carving operator An image size converting method according to Equation 3 below, obtaining a horizontal rate of change in pixels and detecting a seam connecting pixels having the smallest horizontal rate of change. &Quot; (3) "

From here, E _h (x): rate of change of horizontal direction (or energy) at a specific pixel w _h (x): horizontal weighting according to the structure, t (x): Normalized tangent vector (t (x) = [t _x (x), t _y (x)]) e _x : x-axis unit vector g ° (x): rate of change or deflection along the direction I (x): source image Display devices; A user interface unit; A memory for storing a source image; And A control unit for controlling the image conversion unit for converting the source image, The image conversion unit The source image is input, and a combination of an image cropping operator, an image scaling operator, and / or a modified seam carving operator is used to transform the input source image to produce at least one temporary image. The modified shim carving operator determines the direction in which the change in pixel value is greatest at the boundary of objects included in the image, and based on the rate of change between pixels at a predetermined distance in the determined direction. Corresponding to an operator removing or adding a seam, wherein the target image is selected from the at least one temporary image. The method of claim 10, wherein the image conversion unit Receive evaluation information about each of the at least one temporary image, determine the at least one candidate image, obtain image conversion information from each of the determined at least one candidate image, and the image conversion information is performed on the candidate image. Corresponding to information about at least one operator (i.e., image cropping operator, image scaling operator, and / or modified seam carving operator)-reconverting the source image based on the obtained image conversion information to convert the target image. Determining an image size converting apparatus. The method of claim 11, wherein the image conversion unit And generating new image conversion information by performing GA (Genetic Algorithm) operation on the obtained image conversion information, and reconverting the source image based on the generated new image conversion information.

Images