KR100462631B1 - Hallftoning method and apparatus using wavelet transformation - Google Patents

Abstract

A half-toning method and apparatus using wavelet transform is disclosed. The method includes classifying digital image data into low frequency and high frequency images using wavelet transform, halftoning a low frequency image using a first dithering matrix, and a second having a smaller size than the first dithering matrix. Halftoning only the edge portion of the high frequency image using a dithering matrix, and synthesizing the edge regions of the halftoned low frequency image and the halftoned high frequency image, and determining the synthesized result as binary image data. It features. Therefore, the digital images are classified into low frequency images and high frequency images, and half toning of low frequency images using a dithering matrix having a large number of gradations and a low resolution, and halftoning a high frequency image using a dithering matrix having a small number of gradations and a high resolution, When half-toning, the tradeoff between the number of gray levels and the resolution can be improved, and subdividing the low frequency components several times in a hierarchical manner has the effect of half-toning the original image having a very high number of gray numbers at a high resolution.

Description

Halftoning method and apparatus using wavelet transformation

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to halftoning in image processing, and more particularly, to a halftoning method and apparatus using wavelet transform.

Hereinafter, the conventional half toning method will be described as follows.

The gradation scale of digital image data of continuous gradation (or multi gradation) to be printed is adjusted. In this case, an image sharpening process is applied to the result of adjusting the gradation scale to emphasize the high frequency region such as the edge portion of the image. Using the result of the image sharpening process, a half-toning operation for obtaining binary image data of digital image data is performed. Here, half toning may be performed in various ways depending on the type of printer. For example, in the case of a laser printer, a half toning method such as intensive dot dithering is used. As such, the half-toned result is a binary image consisting of '0' and '1' and is input to a raster image scanning unit (not shown) of the laser printer. The last image scanning unit generates the laser on / off signal from the half-toned result. Here, the on / off signal exposes the photosensitive member of the printer so that an image can be printed on a member such as paper.

The conventional halftoning method described above uses a dithering matrix that always has a constant size for halftoning. If the size of the dither matrix is large, the number of gray levels that can be represented increases but the resolution decreases. On the contrary, when the dither matrix is small in size, the resolution is increased but the number of gray levels is decreased. Therefore, according to the conventional half-toning method, there is a problem in that the number of grayscales is reduced when a dithering matrix of a small size is used to increase the resolution, and the resolution is reduced when a large dithering matrix is used to increase the number of grayscales.

An object of the present invention is to provide a half-toning method and apparatus for classifying a digital image into a low frequency image and a high frequency image using wavelet transform, and halftoning by applying different dithering matrices to each of the classified images. There is.

1 is a flowchart for explaining a half-toning method using the wavelet transform according to the present invention.

FIG. 2 is a flowchart for describing an exemplary embodiment of the present invention for halftoning a low frequency image in the fourteenth step illustrated in FIG. 1.

FIG. 3 is a flowchart for describing an exemplary embodiment of the present invention for halftoning a high frequency image in the fourteenth step illustrated in FIG. 1.

FIG. 4 is a flowchart for explaining another exemplary embodiment of the present invention for halftoning a low frequency image in the fourteenth step shown in FIG. 1.

5 is a block diagram of a half-toning apparatus using wavelet transform according to the present invention.

FIG. 6 is a block diagram of a preferred embodiment of the present invention as shown in FIG.

FIG. 7 is a block diagram of a preferred embodiment of the present invention as shown in FIG. 5.

FIG. 8 is a block diagram of another preferred embodiment of the present invention as shown in FIG. 5.

9 is a block diagram of another embodiment according to the present invention of the low frequency half-toning unit shown in FIG. 5.

According to an embodiment of the present invention, a half-toning method using wavelet transform includes classifying digital image data into a low frequency image and a high frequency image using wavelet transform, and using the first dither matrix. Halftoning and halftoning only an edge portion of the high frequency image using a second dithering matrix having a size smaller than the first dithering matrix, and an edge region of the halftoned low frequency image and the halftoned high frequency image. Is synthesized, and the synthesized result is determined as binarized image data.

According to another aspect of the present invention, there is provided a half-toning apparatus using wavelet transform using an image classification unit for classifying digital image data into a low frequency image and a high frequency image using wavelet transform, and using a first dither matrix. A high frequency half toning unit halftoning the edge portion of the high frequency image using a low frequency half toning unit half-toning the low frequency image, a second dithering matrix having a size smaller than the first dithering matrix, and the half toned low frequency image It is preferable that the image synthesizer is composed of an image synthesizer which combines an edge region of the image and the half-toned high frequency image and outputs the synthesized result as binary image data.

Hereinafter, a half toning method using a wavelet transform according to the present invention will be described with reference to the accompanying drawings.

1 is a flowchart for explaining a half-toning method using wavelet transform according to the present invention, comprising: classifying a digital image into a low frequency image and a high frequency image (10th and 12th steps) and a half by classified image. Performing toning (fourteenth and sixteenth steps).

In the half-toning method using the wavelet transform according to the present invention, first, the gradation scale of the digital image (or digital image data) is adjusted (step 10). Here, gradation means a process of concentration transition from the darkest to the lightest effective concentration in the image. Further, the digital image data may have a continuous gray scale (or a large number of gray scales).

After the tenth step, the digital image data whose gradation scale is adjusted is classified into a low frequency image and a high frequency image using two-dimensional wavelet transform (step 12). Here, if the digital image data is two-dimensional wavelet transformed, four images are obtained. One of the four pictures corresponds to a low frequency picture which is an approximate picture and has a low frequency component, and the other three pictures are each a detail having high frequency components as horizontal, vertical and diagonal directions. Corresponds to a high frequency image.

The half-toning method shown in FIG. 1 may not provide a tenth step. In this case, the digital image data is classified into a low frequency image and a high frequency image using two-dimensional wavelet transform without scaling the gray scale of the digital image data (step 12).

Meanwhile, after the twelfth step, half-toning the low frequency image using the first dithering matrix and half-toning only the edge portion of the high frequency image using the second dithering matrix having a smaller size than the first dithering matrix (second Step 14). Here, the first dither matrix is a matrix having a large number of gray scales and a low resolution, and the second dither matrix is a matrix having a small gray scale and a high resolution. At this time, according to the present invention, halftoning of a low frequency image may be performed before halftoning of a high frequency image, later, or simultaneously.

FIG. 2 is a flowchart for describing an exemplary embodiment of the present invention for halftoning a low frequency image in the fourteenth step illustrated in FIG. 1. FIG. 20 and twenty-second steps).

First, the gray scale of the low frequency image is adjusted (step 20). After the twentieth step, the low frequency image having the gray scale scale adjusted is half-toned using the first dither matrix, and the process proceeds to the sixteenth step (step 22).

FIG. 3 is a flowchart illustrating an embodiment of the present invention for halftoning a high frequency image in the fourteenth step shown in FIG. 1, in which an image is sharpened and scaled as needed. (30th through 36th steps) and halftoning only the edge area of the high frequency image (38th and 40th steps).

First, inverse wavelet transform of the classified high frequency image (step 30). After the thirtieth step, it is determined whether the edge region of the high frequency image is clearly expressed using the result of the inverse wavelet transform (step 32). That is, it is determined whether the edge region is sufficiently represented.

If it is determined that the edge area is clearly expressed, the flow proceeds to step 36. However, if it is determined that the edge region is not clearly represented, the image corresponding to the inverse wavelet transformed result is sharpened (step 34). If it is determined that the edge region is clearly expressed after the thirty-fourth step, the gray scale of the sharpened result or the result of the inverse wavelet transform of the high frequency image is adjusted (step 36). After the thirty-sixth step, an edge region of the high frequency image is selected from the result of adjusting the gray scale (step 38). For example, in order to select the edge area of a high frequency image, a quadtree decomposition method may be used.

In the embodiment of the step of halftoning the high frequency image shown in FIG. 3, the 32nd, 34th and 36th steps may not be provided.

If the thirty-second and thirty-fourth steps are not provided, the gray scale of the result of the inverse wavelet transform after the thirtieth step is adjusted (step 36). However, if the 36th step is not provided, the edge region of the high frequency image is selected from the sharpened result or the result of the inverse wavelet transform of the high frequency image (step 38). In addition, when all of the thirty-second, thirty-fourth and thirty-sixth steps are not provided, after the thirtieth step, an edge region of the high frequency image is selected from the result of the inverse wavelet transform (step 38). In this manner, only the edge region is selected without selecting the whole high frequency image.

After operation 38, the selected edge region is half-toned using the second dither matrix, and the operation proceeds to operation 16 (operation 40).

Meanwhile, the edge regions of the halftoned low frequency image and the halftoned high frequency image are synthesized, and the synthesized result is determined as a binarized image (or binarized image data) (step 16).

FIG. 4 is a flowchart for explaining another exemplary embodiment of the present invention for halftoning a low frequency image in the fourteenth step shown in FIG. 1. The low frequency image is further subdivided into high frequency components and low frequency components. It consists of the steps (50th and 52nd steps).

First, the low frequency image classified in the twelfth step is classified into a sub low frequency image and a sub high frequency image using wavelet transform (step 50). At this time, the gray level of the sorted low frequency image can be adjusted before the second sorted low frequency image is secondly reclassified into a sub low frequency image and a sub high frequency image using wavelet transform. In this case, the result of adjusting the gradation is classified into a sub low frequency image and a sub high frequency image (step 50).

After the 50th step, the half-toning of the sub low frequency image is performed using a third dither matrix having a size larger than the first dither matrix, and the edge of the sub high frequency image is used using a fourth dither matrix having the same size as the first dither matrix. Half-toning only the part and proceeding to step 16 (step 52). At this time, according to the present invention, the sub low frequency image can be halftoned as shown in FIG. 2, and the sub high frequency image can be half toned as shown in FIG. In this case, the edge region of the sub-high frequency image half-toned in the 52nd step and the edge region of the high frequency image half-toned in the 52nd step and the edge region of the high frequency image half-toned in the 14th step are synthesized, and the synthesized result is binarized. Determination as image data (step 16).

According to another embodiment of the present invention, the high frequency image may be further subdivided and halftoned, similarly to the low frequency image subdivided and halftoned again as shown in FIG. 4. That is, the high frequency image classified first in the twelfth step is secondly classified into a sub low frequency image and a sub high frequency image using wavelet transform. At this time, the half-toning of the sub-low frequency image using the fifth dither matrix having a size larger than the second dither matrix, and only the edge portion of the sub-frequency image using the sixth dither matrix having the same size as the second dither matrix Halftone and proceed to step 16.

According to another embodiment of the present invention, the high-frequency or low-frequency image that is primarily subdivided in the twelfth step may not be subdivided only secondarily as described above, but may be hierarchically subdivided and half-toned several times. . That is, the first subdivided low frequency or high frequency image may be further subdivided into the low frequency component and the high frequency component, and the second subdivided sub low frequency or sub high frequency image may be subdivided into the third or subsequent quadratic hierarchically. . At this time, the more the image is subdivided, the larger the size of the dither matrix used for half toning. Here, the larger the size of the dither matrix, the lower the resolution and the higher the number of gray scales.

On the other hand, when the half-toning method according to the present invention described above is applied to a laser printer, the binarized image data resulting from the half-toning is applied to a raster image processor (not shown) to be used to drive the laser of the printer. Can be. At this time, the laser is driven so that the photosensitive drum is exposed and the toner is developed so that an image can be printed on paper.

In order to facilitate understanding of the present invention, when outputting axb pixel digital image data having a gradation of n bits (2 ⁿ ) to a laser printer having a resolution of x dot per inch (dpi), Digital image data is halftoned as binary image data as follows.

First, the gradation of digital image data is scaled (step 10). At this time, when the two-dimensional wavelet transform is applied to the gray scaled digital image data, it is classified into four a / 2 x b / 2 images (step 12). One of them is a low frequency image, and the other three are high frequency images. At this time, in order to perform the fourteenth step, the first dithering matrix used for halftoning the low frequency image is a matrix of 45 degree screen angles of c × c and has 2 × c × c + 1 gray scale numbers. The size of the binary image generated by half-toning in this manner is (c × a / 2) × (c × b / 2). Meanwhile, inverse wavelet transform is performed on the high frequency image (step 30). In this case, a thirty-second step is performed to check whether it is necessary to perform a thirty-fourth step on the inverse wavelet transformed result. In this case, after performing the 36th step selectively, the 38th step is performed to select the edge region. At this time, when performing the 40th step, the second dithering matrix used for halftoning the selected edge region is a matrix of 45 degrees screen angle of d × d, and has 2 × d × d + 1 gray scales. .

In this manner, the half-toned low frequency image and the half-toned high frequency image are synthesized, and finally, binarized image data having a size of (c × a / 2) × (c × b / 2) is obtained (step 16).

Hereinafter, the configuration and operation of a half-toning apparatus using wavelet transform according to the present invention will be described with reference to the accompanying drawings.

5 is a block diagram of a half-toning apparatus using wavelet transform according to the present invention, and includes a first scaling unit 60, an image classifying unit 62, a low frequency half toning unit 64, and a high frequency half toning unit 66. And an image synthesizing unit 68.

The half toning apparatus illustrated in FIG. 5 may perform the half toning method illustrated in FIG. 1. For example, in order to perform the tenth step, the first scaling unit 60 illustrated in FIG. 5 adjusts the gray scale of the digital image data input through the input terminal IN1, and displays the digital image data of which the gray scale is adjusted. Output to the classifier 62.

At this time, in order to perform the twelfth step, the image classifying unit 62 classifies the digital image data whose gray scale scale input from the first scaling unit 60 is adjusted into a low frequency image and a high frequency image, and classifies the low frequency image. Are output to the low frequency half toning unit 64, and the classified high frequency images are output to the high frequency half toning unit 66, respectively.

The half toning apparatus shown in FIG. 5 may not provide the first scaling unit 60. In this case, the image classifying unit 62 classifies the digital image data input through the input terminal IN1 into a low frequency image and a high frequency image using wavelet transform, and classifies the classified low frequency image and the high frequency image into the low frequency half toning unit 64. ) And high frequency half-toning unit 66, respectively.

In order to perform the fourteenth step, a low frequency half toning unit 64 and a high frequency half toning unit 66 may be provided. Here, the low frequency half-toning unit 64 half-tons the low-frequency image using the first dither matrix, and outputs the half-toned result to the image synthesizing unit 68. In addition, the high frequency half toning unit 66 halftons only the edge portion of the high frequency image using a second dithering matrix having a size smaller than that of the first dithering matrix, and outputs the half-toned result to the image combining unit 68. .

FIG. 6 is a block diagram of a preferred embodiment 64A according to the present invention of the low frequency half toning unit 64 shown in FIG. 5, which is composed of a second scaling unit 70 and a first half toning unit 72. As shown in FIG. do.

The low frequency half-toning unit 64A shown in FIG. 6 performs the twentieth and twenty-second steps shown in FIG. 2. That is, the second scaling unit 70 of the low frequency half-toning unit 64A adjusts the gray scale of the low frequency image input from the image classifying unit 62 through the input terminal IN2 to perform the twentieth step. The scaled low frequency image is output to the first half toning unit 72. In order to perform the twenty-second step, the first half-toning unit 72 half-tons the low-frequency image of which the gray scale scale input from the second scaling unit 70 is adjusted using the first dither matrix, and half-toned. Is outputted to the image synthesizing section 68 through the output terminal OUT2.

FIG. 7 is a block diagram of a preferred example 66A according to the present invention of the high frequency half-toning unit 66 shown in FIG. 5, which includes an inverse wavelet transform unit 80, an edge inspection unit 82, and image sharpening ( image sharpening) 84, an edge selector 86, and a second half-toning 88.

The high frequency half-toning unit 66A shown in FIG. 7 performs the thirtieth, thirty-second, thirty-fourth, thirty-eighth, and forty-fourth steps shown in FIG. 3. For example, in order to perform the thirtieth step, the inverse wavelet transform unit 80 performs inverse wavelet transform on the high frequency image inputted from the image classifying unit 62 through the input terminal IN3 and edges the result of the inverse wavelet transform. Output to the inspection unit 82.

In order to perform the thirty-second step, the edge inspecting unit 82 checks whether the edge region is clearly expressed from the inverse wavelet transformed result input from the inverse wavelet transforming unit 80, and uses the image as a control signal. Output to the sharpening unit 84 and the edge selection unit 86, respectively.

In order to perform the thirty-fourth step, the image sharpening unit 84 corresponds to a result of the inverse wavelet transform input from the inverse wavelet transform unit 80 in response to the control signal input from the edge inspection unit 82. The image is sharpened and the result of the sharpened image is output to the edge selector 86. For example, if it is recognized that the edge region is not clearly expressed through the control signal, the image sharpening unit 84 sharpens the image corresponding to the result of the inverse wavelet transform input from the inverse wavelet transform unit 80.

In order to perform the 38 th step, the edge selector 86 may perform an inverse wavelet transformed result from the inverse wavelet transform unit 80 or an image sharpening unit in response to a control signal input from the edge inspector 82. An edge region of the high frequency image is selected from the sharpened result input from 84, and the selected edge region is output to the second half toning unit 88. For example, if it is recognized that the edge region is not clearly expressed through the control signal, the edge selector 86 selects the edge region of the high frequency image from the sharpened result input from the image sharpening unit 84. However, if it is recognized that the edge region is clearly expressed through the control signal, the edge selector 86 selects the edge region of the high frequency image from the result of the inverse wavelet transform input from the inverse wavelet transform unit 80.

In order to perform the forty-second step, the second half toning unit 88 halftons the selected edge area input from the edge selecting unit 86 using the second dither matrix, and the half toning unit 88 outputs the half-toned result. 68) through the output terminal OUT3.

FIG. 8 is a block diagram of another preferred embodiment 66B according to the present invention of the high frequency half-toning unit 66 shown in FIG. 5, which is an inverse wavelet transform unit 100, an edge inspection unit 102, and an image sharpening unit. 104, a third scaling unit 106, an edge selecting unit 108, and a second half toning unit 110.

The high frequency half toning unit 66B illustrated in FIG. 8 performs half toning of the high frequency image illustrated in FIG. 3, that is, performs the 30 th through 40 th steps. Therefore, the inverse wavelet transform unit 100, the edge inspecting unit 102, the image sharpening unit 104, and the second half toning unit 110 illustrated in FIG. 8 are the inverse wavelet transform unit 80 illustrated in FIG. 7. ), The edge inspecting unit 82, the image sharpening unit 84, and the second half toning unit 88 play the same role, and thus description thereof will be omitted. That is, the inverse wavelet transform unit 100, the edge inspection unit 102, the image sharpening unit 104, and the second half toning unit 110 illustrated in FIG. It serves to perform steps 34 and 40 respectively.

In order to perform the thirty-sixth step, the third scaling unit 106 responds to the control signal input from the edge inspecting unit 102, and the sharpened result input from the image sharpening unit 104 or the inverse wavelet converting unit ( The gray scale of the inverse wavelet transformed result input from 100 is adjusted, and the result of adjusting the gray scale is output to the edge selector 108. For example, if it is recognized that the edge area is not clearly expressed through the control signal, the third scaling unit 106 adjusts the gray scale of the sharpened result input from the image sharpening unit 104. However, if it is recognized that the edge region is clearly expressed through the control signal, the third scaling unit 106 adjusts the gray scale of the result of the inverse wavelet transform input from the inverse wavelet transform unit 100.

In order to perform the 38 th step, the edge selector 108 selects an edge region of the high frequency image from the result of adjusting the gradation scale input from the third scaling unit 106, and selects the selected edge region as the second half toning unit. Output to (110). Like the second half toning unit 88 illustrated in FIG. 7, the second half toning unit 110 performs a second dither matrix on the selected edge region input from the edge selection unit 108 to perform the 40th step. Half-toning is performed, and the half-toned result is output to the image synthesizing unit 68 through the output terminal OUT4.

Meanwhile, in order to perform the sixteenth step, the image synthesizing unit 68 performs edges of the half-toned low frequency image input from the low frequency half toning unit 64 and the half toned high frequency image input from the high frequency half toning unit 66. The regions are synthesized and the synthesized result is output through the output terminal OUT1 as binary image data.

FIG. 9 is a block diagram of another embodiment 64B according to the present invention of the low frequency half toning unit 64 shown in FIG. 5, which includes an image classifier 120, a low frequency half toning unit 122, and a high frequency half toning unit ( 124).

The low frequency half toning unit 64B shown in FIG. 9 performs the 50th and 52nd steps shown in FIG. 4. For example, the image classifier 120 classifies the low frequency image input from the image classifying unit 62 through the input terminal IN5 into the sub low frequency image and the sub high frequency image by using the wavelet transform to perform the 50th step. The sorted sub low frequency image and the sub high frequency image are output to the low frequency half toning machine 122 and the high frequency half toning machine 124, respectively.

In order to perform the fifty-second step, the low frequency half toning unit 64B provides a low frequency half toning unit 122 and a high frequency half toning unit 124. Here, the low frequency half toning unit 122 halftons the sub low frequency image using a third dithering matrix having a size larger than that of the first dithering matrix, and outputs the half toned image through the output terminal OUT5. Will output The high frequency half toning unit 124 uses the fourth dithering matrix having the same size as the first dithering matrix to half toning only the edge portion of the sub-high frequency image, and outputs the half toned image through the output terminal OUT6. )

In this case, the image synthesizing unit 68 performs the edge region and the high frequency half toning of the half-toned sub low frequency image input from the low frequency half toning machine 122 and the half toned sub high frequency image input from the high frequency half toning machine 124. The edge areas of the high frequency image inputted from the unit 66 are synthesized, and the synthesized result is output as output image OUT1 as binary image data.

As described above, the half-toning method and apparatus using wavelet transform according to the present invention classify a digital image by a low frequency image and a high frequency image, and half-ton the low frequency image by using a dithering matrix having a large number of gray scales and a low resolution. Halftoning a high frequency image using a dithering matrix with low gradation and high resolution can improve the tradeoff between gradation and resolution when halftoning the image, and segment low frequency components several times hierarchically The effect is that half-toning of an original image having a very large number of gradations at a high resolution can be achieved.

Claims (14)

(a) classifying the digital image data into a low frequency image and a high frequency image using wavelet transform; (b) halftoning the low frequency image using a first dithering matrix, and halftoning only an edge portion of the high frequency image using a second dithering matrix having a smaller size than the first dithering matrix; And (c) synthesizing the edge regions of the halftoned low frequency image and the halftoned high frequency image, and determining the synthesized result as binarized image data. . The method of claim 1, wherein the half-toning method using the wavelet transform Adjusting the gradation scale of the digital image data and proceeding to the step (a), In the step (a), the digital image data of which the gray scale is adjusted is classified into the low frequency image and the high frequency image by using a wavelet transform. The method of claim 1, wherein the half-toning of the low frequency image in the step (b) (b11) adjusting a gray scale of the low frequency image; And (b12) half-toning the low-frequency image of which the gray scale is adjusted by using the first dither matrix. The method of claim 1, wherein the half-toning of the high frequency image in the step (b) (b21) inverse wavelet transforming the high frequency image; (b22) selecting the edge region of the high frequency image from the inverse wavelet transformed result; And (b23) halftoning the selected edge region by using the second dither matrix. The method of claim 4, wherein the half-toning of the high frequency image in the step (b) (b24) after step (b21), determining whether the edge area is clearly expressed, and if it is determined that the edge area is clearly expressed, proceeding to step (b22); And (b25) if it is determined that the edge region is not clearly represented, further comprising sharpening an image corresponding to the result of the inverse wavelet transform and proceeding to step (b22), (B22) is a half-toning method using the wavelet transform, characterized in that for selecting the edge region of the high frequency image in the sharpened result. The method of claim 5, wherein the half-toning of the high frequency image in the step (b) After the step (b25), further comprising adjusting the gray scale of the sharpened result and proceeding to the step (b22), In the step (b22), the edge toning of the high frequency image is selected based on the sharpened result of adjusting the gray scale. The method of claim 1, wherein the half-toning of the low frequency image in the step (b) Classifying the low frequency image into a sub low frequency image and a sub high frequency image using the wavelet transform; And Half toning the sub low frequency image using a third dither matrix having a size larger than the first dither matrix, and using the fourth dither matrix having the same size as the first dither matrix, an edge portion of the sub high frequency image. Halftoning the bay, In the step (c), the edge region of the half-toned sub-frequency image, the edge region of the half-toned sub-frequency image, and the edge region of the high frequency image are synthesized, and the synthesized result is determined as the binarized image data. Halftoning method using wavelet transform. An image classification unit for classifying digital image data into low frequency images and high frequency images using wavelet transform; A low frequency half toning unit half toning the low frequency image using a first dither matrix; A high frequency half toning unit half-toning only an edge portion of the high frequency image by using a second dither matrix having a smaller size than the first dither matrix; And And an image synthesizer for synthesizing the edge regions of the halftoned low frequency image and the halftoned high frequency image and outputting the synthesized result as binarized image data. 10. The apparatus of claim 8, wherein the half toning device using the wavelet transform A first scaling unit which adjusts the gradation scale of the digital image data and outputs the digital image data of which the gradation scale is adjusted to the image classification unit, And the image classifying unit classifies the digital image data of which the gradation scale is input from the first scaling unit into the low frequency image and the high frequency image. The method of claim 8, wherein the low frequency half toning unit A second scaling unit which adjusts a gray scale of the low frequency image and outputs a low frequency image in which the gray scale is adjusted; And And a first half-toning unit for half-toning the low frequency image of which the gray scale scale inputted from the second scaling unit is adjusted using the first dithering matrix. The method of claim 8, wherein the high frequency half-toning unit An inverse wavelet transform unit for inverse wavelet transforming the high frequency image; An edge selector for selecting the edge region of the high frequency image from the inverse wavelet transformed result input from the inverse wavelet transformer; And And a second half toning unit half-toning the selected edge region by using the second dither matrix. The method of claim 11, wherein the high frequency half-toning unit An edge checker which checks whether the edge region is clearly expressed from the inverse wavelet transformed result input from the inverse wavelet transformer, and outputs the inspected result as a control signal; And And an image sharpening unit configured to sharpen an image corresponding to the inverse wavelet transformed result in response to the control signal. The edge selector selects the edge region of the high frequency image from the inverse wavelet transformed result input from the inverse wavelet transform unit or the sharpened result input from the image sharpening unit in response to the control signal. Halftoning apparatus using a wavelet transform, characterized in that. The method of claim 12, wherein the high frequency half-toning unit In response to the control signal, adjust a gray scale of the sharpened result input from the image sharpening unit or the inverse wavelet transformed result input from the inverse wavelet transform unit, and adjust a gray scale scale result. Further comprising a third scaling unit for outputting, And the edge selector selects the edge region of the high frequency image from a result of adjusting the gradation scale input from the third scaling unit. The method of claim 8, wherein the low frequency half toning unit An image classifier for classifying the low frequency image into a sub low frequency image and a sub high frequency image using the wavelet transform; A low frequency half toning device for half-toning the sub low frequency image using a third dithering matrix having a size larger than the first dithering matrix; And A high frequency half toning device for halftoning only an edge portion of the sub-frequency image by using a fourth dithering matrix having the same size as the first dithering matrix, The image synthesizing unit synthesizes the edge region of the half-toned sub-frequency image, the edge region of the half-toned sub-high frequency image, and the edge region of the high frequency image, and outputs the synthesized result as the binarized image data. Halftoning device using Brit transform.