JP4386870B2 - Image processing method, image processing apparatus, image forming apparatus, computer program, and recording medium - Google Patents

Description

  The present invention relates to an image processing method, an image processing apparatus, and an image processing method for converting image data to a spatial frequency to obtain a frequency component, processing the obtained frequency component, and inversely converting the processed frequency component to image data. The present invention relates to a forming apparatus, a computer program, and a recording medium.

  In the digital image processing technology, as one method for obtaining a high-definition image, there is an edge enhancement and smoothing process of an image. For example, the outline of a character or a continuous tone image (for example, a photographic paper photograph) can be clearly seen by emphasizing the edge. For a printed matter (for example, a printed photograph) formed of halftone dots, Generation of moire can be suppressed by performing smoothing. In these edge enhancement and smoothing processes, a spatial filter process using a digital filter is generally used (see, for example, Journal of Electrophotographic Society Vol. 36, No. 4 (1997), p. 343-352).

In addition, as one of the techniques for enhancing the edge component of the image, the original image in the coordinate region is transformed into the frequency region by orthogonal transformation, the frequency component is changed using a constant prepared in advance, and the changed frequency There has been proposed a method of performing inverse orthogonal transformation of components and inversely transforming into a coordinate region to obtain a sharpened image (see, for example, Patent Document 1).
JP-A-9-54825

  However, in the method of Patent Document 1, since all frequency components are changed and emphasized by applying constants prepared in advance for the entire frequency region, in a gentle curved region such as a face contour, There is a problem that unnecessary block noise stands out. In addition, the frequency component is changed by multiplying each frequency component of the orthogonally transformed frequency region by a value prepared in advance for each period of the frequency component to obtain a clearer image. Multiple values need to be set. When a plurality of values are set in this way, a large number of registers are required, and there is a problem that the circuit scale increases. Further, the method of Patent Document 1 is an edge enhancement method and does not support smoothing processing, and there is a problem that, for example, both edge enhancement and smoothing processing cannot be performed.

  The present invention has been made in view of such circumstances, and compares the frequency component with a predetermined value for each of the specific frequency components, and changes the specific frequency components based on the comparison result. By doing so, the image processing method, the image processing apparatus, and the image formation that can perform the mixed processing of the enhancement processing and the smoothing processing in the block that is the unit of conversion to the spatial frequency and can obtain a more optimal image An object is to provide an apparatus, a computer program, and a recording medium.

  In the present invention, the specific frequency component is a frequency component on the low frequency side of all frequency components and a frequency component on the high frequency side only in the horizontal direction or only in the vertical direction of the image data, and the specific frequency component When the absolute value of the coefficient is larger than the predetermined value, the changing means is configured to increase the coefficient of the frequency component, thereby providing an image processing apparatus capable of obtaining a clean edge curve. Objective.

  Further, the present invention provides a coefficient of other frequency components excluding both a frequency component on the low frequency side and a specific frequency component including the frequency component on the high frequency side only in the horizontal direction or only in the vertical direction of the image data, and the DC component. It is another object of the present invention to provide an image processing apparatus that can obtain a smoother image by configuring the changing means so as to reduce.

  Further, in the present invention, the specific frequency component is a frequency component excluding both a low-frequency side frequency component and a direct current component among all frequency components, and an absolute value of a coefficient of the specific frequency component is greater than the predetermined value. Another object of the present invention is to provide an image processing apparatus capable of obtaining a clean edge curve by configuring the changing means to increase the coefficient of the frequency component when the ratio is large.

  In the present invention, the changing means is configured to increase the specific frequency component excluding both the frequency component and the DC component on the low frequency side, and the coefficient of the other frequency component excluding both the DC component. Accordingly, another object is to provide an image processing apparatus capable of obtaining a clearer image.

  Further, in the present invention, when the absolute value of the coefficient of the specific frequency component is larger than the predetermined value, a correction coefficient that is smaller as the frequency component becomes lower and becomes higher as the frequency component becomes the coefficient of the frequency component. It is another object of the present invention to provide an image processing apparatus that can obtain an image with a beautiful edge by configuring the changing means to multiply.

  Further, in the present invention, when the absolute value of the coefficient of the specific frequency component is smaller than the predetermined value, the changing unit is configured to reduce the frequency component, so that a smoother image can be obtained. Another object is to provide an image processing apparatus.

An image processing method according to the present invention performs image processing for converting image data to a spatial frequency to obtain a plurality of frequency components, processing the obtained frequency components, and inversely converting the processed frequency components to image data In the method, the step of comparing the magnitude of the absolute value of the frequency component with a predetermined value for each of a plurality of specific frequency components of the plurality of frequency components obtained by performing conversion to a spatial frequency, and comparing based on the results and have a a step of changing the specific frequency components, the specific frequency components, the high frequency side about only the horizontal direction of the frequency component and the image data of the low frequency side of the all frequency components If the absolute value of the coefficient of the specific frequency component is larger than the predetermined value, the relationship between the frequency component and And a first process for reducing coefficients of other frequency components excluding both the specific frequency component and the direct current component, and the specific frequency component includes a frequency component on a low frequency side of all frequency components and The frequency component excluding both of the direct current components, wherein the changing unit increases the coefficient of the frequency component when the absolute value of the coefficient of the specific frequency component is larger than the predetermined value, and the specific frequency component and The second process for increasing the coefficients of other frequency components excluding both DC components is switched according to the characteristics of the image data .

An image processing apparatus according to the present invention performs image processing for converting image data to a spatial frequency to obtain a plurality of frequency components, processing the obtained frequency components, and inversely converting the processed frequency components to image data Comparing means for comparing the magnitude of the absolute value of the frequency component with a predetermined value for each of a plurality of specific frequency components of the plurality of frequency components obtained by performing conversion to a spatial frequency in the apparatus; Changing means for changing each specific frequency component based on a comparison result of the comparing means, and the specific frequency component is a low frequency side frequency component of all frequency components and a high frequency only in the horizontal direction of the image data. The frequency component on the high frequency side and the frequency component on the high frequency side only in the vertical direction, and the changing means is configured to change A first process of increasing a coefficient of a frequency component and decreasing a coefficient of another frequency component excluding both the specific frequency component and the DC component; and the specific frequency component is a lower frequency side of all frequency components The frequency component excluding both the frequency component and the DC component of the frequency component, wherein the changing means increases the coefficient of the frequency component when the absolute value of the coefficient of the specific frequency component is larger than the predetermined value. And switching means for switching the second process for increasing the coefficients of other frequency components excluding both the frequency component and the DC component according to the characteristics of the image data .

In the image processing apparatus according to the present invention, the switching means switches to the first processing when the image data is image data derived from a photographic document, and the switching means performs the switching when the image data is image data derived from a character document. It is characterized by switching to the second process .

  In the image processing apparatus according to the present invention, when the absolute value of the coefficient of the specific frequency component is larger than the predetermined value, the changing unit performs a correction that is smaller as the frequency component is lower and is larger as the frequency is higher. The coefficient is configured to be multiplied by the coefficient of the frequency component.

  The image processing apparatus according to the present invention is characterized in that the changing means is configured to decrease the coefficient of the frequency component when the absolute value of the coefficient of the specific frequency component is smaller than the predetermined value. To do.

  An image forming apparatus according to the present invention includes the image processing apparatus according to the present invention and an image forming unit that forms an image processed by the image processing apparatus on a sheet.

A computer program according to the present invention causes a computer to perform conversion of image data to a spatial frequency to obtain a plurality of frequency components, to process the obtained frequency components, and to convert the processed frequency components into image data. In the computer program to be inversely converted, the magnitude of the absolute value of the frequency component and the predetermined value for each of a plurality of specific frequency components among a plurality of frequency components obtained by performing conversion to a spatial frequency in the computer Comparing the specific frequency components on the basis of the comparison result, and causing the computer to specify the specific frequency component as a frequency component on the lower frequency side of all frequency components and image data. Frequency component on the high frequency side only in the horizontal direction and frequency component on the high frequency side only in the vertical direction, First processing for increasing the coefficient of the frequency component and decreasing the coefficient of other frequency components excluding both the specific frequency component and the DC component when the absolute value of the coefficient of the specific frequency component is larger than the predetermined value And the specific frequency component is a frequency component excluding both a low frequency side frequency component and a direct current component among all frequency components, and the changing means has an absolute value of a coefficient of the specific frequency component. When the value is larger than the predetermined value, the second process of increasing the coefficient of the frequency component and increasing the coefficient of other frequency components excluding both the specific frequency component and the DC component is switched according to the characteristics of the image data. And a procedure for making it happen .

  A recording medium according to the present invention records the computer program of the present invention.

  In the present invention, for each specific frequency component, the absolute value of the frequency component is compared with a predetermined value, and the specific frequency component is changed based on the comparison result. Different changes can be made between the portion where the value is larger than the predetermined value and the portion where the absolute value of the frequency component is equal to or smaller than the predetermined value. For example, since the portion where the absolute value of the frequency component is larger than a predetermined value includes a feature component such as an edge component, it is possible to perform enhancement processing on the edge portion and smoothing processing on the other portion. The change of the frequency component can change the coefficient of the frequency component. By switching different processing such as enhancement processing and smoothing processing according to the determination result for each frequency component, mixed processing such as enhancement processing and smoothing processing can be performed within a block that is a unit of conversion to spatial frequency. It is possible to suppress determination errors when switching between emphasis processing and smoothing processing in units of blocks, and a more optimal image can be obtained. For example, it is possible to form the more optimal image on the sheet by performing the above-described processing on the image formed on the sheet. In addition, for example, the above-described processing can be executed by a computer to obtain a more optimal image. When making a computer perform the process mentioned above, it is possible to make a computer read the recording medium on which the program which makes a computer perform the process mentioned above was read, and to perform it.

  In the present invention, among all the frequency components, the frequency component on the low frequency side and the frequency component on the high frequency side only in the horizontal direction or only in the vertical direction of the image data, the absolute value of the coefficient of the specific frequency component is the above-mentioned If it is greater than a predetermined value, the changing means increases the coefficient of the frequency component, so that all frequency components are not subjected to emphasis processing, but only frequency components determined to include feature parts such as edge components in the block The emphasis process is performed by increasing the number of lines, and the appearance of an unnecessary block pattern at the edge boundary portion on the curve is suppressed, so that a clean edge curve can be obtained. Here, the coefficient of the frequency component is, for example, a DCT coefficient of discrete cosine transform (DCT).

  In the present invention, the specific frequency component including the frequency component on the low frequency side and the frequency component on the high frequency side only in the horizontal direction or only in the vertical direction of the image data, and other frequency components excluding both DC components. Since the coefficient is reduced by the changing unit, smoothing processing can be performed by reducing other frequency components to obtain a smoother image.

  In the present invention, the specific frequency component is a frequency component excluding both the frequency component on the low frequency side and the DC component among all frequency components, and when the absolute value of the coefficient of the frequency component is larger than the predetermined value, In order for the changing means to increase the coefficient of the frequency component, the emphasis process is not performed on all frequency components, but only the frequency component determined to include a characteristic part such as an edge component is increased, and the emphasis process is performed. A beautiful edge curve can be obtained by suppressing the appearance of an unnecessary block pattern at the edge boundary portion on the curve. Here, the coefficient of the frequency component is, for example, a DCT coefficient of discrete cosine transform (DCT).

  In the present invention, the changing means increases the specific frequency component excluding both the frequency component on the low frequency side and the DC component, and the coefficient of the other frequency component excluding both the DC component. The emphasis process can be performed by increasing, and a clearer image can be obtained.

  In the present invention, when the absolute value of the coefficient of the specific frequency component is larger than the predetermined value, the changing unit sets a correction coefficient that is smaller as the frequency component is lower and is higher as the frequency component is higher. Since the coefficient of the component is multiplied, the horizontal component and the vertical component of the frequency component are corrected so as to increase toward the higher frequency side, and an image with a beautiful edge can be obtained even when reproducing a curve or the like.

  In the present invention, when the coefficient of the specific frequency component is smaller than the predetermined value, the changing unit decreases the coefficient of the frequency component, and therefore does not include features such as an edge component among the specific frequency component. Smoothing processing can be performed by reducing frequency components, and a smoother image can be obtained.

  According to the present invention, mixed processing such as emphasis processing and smoothing processing can be performed within a block, and determination errors when switching different processing such as emphasis processing and smoothing processing on a block basis are suppressed, and more optimal Can be obtained.

  According to the present invention, only a frequency component determined to include a feature such as an edge component is emphasized, and the appearance of an unnecessary block pattern at an edge boundary portion on the curve is suppressed, thereby producing a clean edge curve. Obtainable.

  According to the present invention, the specific frequency component including the frequency component on the low frequency side and the frequency component on the high frequency side only in the horizontal direction or only in the vertical direction of the image data, and other frequency components excluding both the DC component. Smoothing processing can be performed to obtain a smoother image.

  According to the present invention, a specific frequency component excluding both the frequency component on the low frequency side and the DC component, and the other frequency component excluding both the DC component can be enhanced to obtain a clearer image. it can.

  According to the present invention, by correcting the horizontal component and the vertical component of the frequency components so as to increase toward the high frequency side, an image with a beautiful edge can be obtained even in the reproduction of a curve or the like.

  According to the present invention, smoothing processing can be performed on frequency components that do not include features such as edge components, and a smoother image can be obtained.

Hereinafter, the present invention will be specifically described with reference to the drawings illustrating embodiments thereof.
(Embodiment 1)
FIG. 1 is a block diagram showing a configuration example of an image forming apparatus 40 including an image processing apparatus according to the present invention. In this description, the image forming apparatus 40 operates as a digital color copying machine. The image forming apparatus 40 includes an image input apparatus 30 capable of reading a color image, an image processing apparatus 31 capable of processing a color image, and an image output apparatus (image forming apparatus) capable of outputting a color image to a sheet or the like. Means) 32 and an operation panel 33. Although not shown, the image forming apparatus 40 includes a control unit such as a CPU (Central Processing Unit) that controls each device in the image forming apparatus 40.

  The image input device 30 includes, for example, a CCD (Charge Coupled Device), and a reflected light image from an original is read by the CCD, and RGB analog signals (R: red, G: green, B: blue) are generated. . The generated RGB analog signal is sent to the image processing device 31.

  The image processing apparatus 31 includes an A / D (analog / digital) conversion unit 311, a shading correction unit 312, an input tone correction unit 313, a region separation processing unit 314, a color correction unit 315, a black generation and under color removal unit 316, and an image quality. An adjustment unit 317, an output gradation correction unit 318, a gradation reproduction processing unit 319, and a control unit (not shown) that controls each unit are provided.

  The image processing device 31 converts the RGB analog signal received from the image input device 30 into an RGB digital signal, performs various image processing such as correction processing, and performs CMYK digital signals (C: cyan, M: magenta, Y: yellow). , K: black) and the number of gradations of the generated CMYK digital signal (hereinafter referred to as CMYK signal) is reduced to binary or quaternary. Output image data that has been binarized or binarized is temporarily stored in a storage device (not shown), and is output to the image output device 32 at a predetermined timing.

  The A / D conversion unit 311 receives the RGB analog signal from the image input device 30, converts the received RGB analog signal into an RGB digital signal, and sends the RGB digital signal to the shading correction unit 312. The shading correction unit 312 performs processing for removing various distortions generated in the illumination system, the imaging system, and the imaging system of the image input device 30 on the RGB digital signal received from the A / D conversion unit 311, and then performs input. This is sent to the gradation correction unit 313. The input tone correction unit 313 adjusts the color balance of the RGB digital signal (RGB reflectance signal) received from the shading correction unit 312 and performs processing by the image processing system employed in the image processing device 31. It is converted into an easy density signal or the like and sent to the region separation processing unit 314.

  The region separation processing unit 314 separates each pixel in the image of the RGB digital signal received from the input tone correction unit 313 into one of a character region, a halftone dot region, and a photograph region, and based on the separation result, the pixel Is output to the black generation and under color removal unit 316 and the gradation reproduction processing unit 319. The RGB digital signal received from the input tone correction unit 313 is sent to the color correction unit 315 as it is.

  The color correction unit 315 converts the RGB digital signal sent from the input tone correction unit 313 into a CMY digital signal (hereinafter referred to as a CMY signal) and includes an unnecessary absorption component in order to perform color reproduction faithfully. After performing the process of removing the color turbidity based on the spectral characteristics of the CMY color material, it is sent to the black generation and under color removal unit 316. The black generation and under color removal unit 316 performs black generation to generate a black signal (K signal) from the three color signals (C signal, M signal, and Y signal) of the CYM signal received from the color correction unit 315, A CMY signal obtained by black generation is subtracted from the CMY signal to generate a new CMY signal, and a CMYK four-color signal (CMYK signal) is sent to the image adjustment unit 317.

As a general black generation process, there is a method of generating black using skeleton black. In this method, the input / output characteristic of the skeleton curve is y = f (x), the input data is C, M, Y, the output data is C ′, M ′, Y ′, K ′, UCR (Under Color If the removal rate is α (0 <α <1),
K ′ = f {min (C, M, Y)}
C ′ = C−αK ′
M ′ = M−αK ′
Y ′ = Y−αK ′
It is represented by

  The image quality adjustment unit 317 performs frequency conversion processing on the image of the CMYK signal received from the black generation and under color removal unit 316, and whether or not an edge component is included in each frequency component included in the block that performs frequency conversion processing. Different changes are made depending on the image quality, and processing such as improving image blurring or graininess deterioration is performed. The output tone correction unit 318 converts the CMYK signal processed by the image quality adjustment unit 317 into a halftone dot area ratio that is a characteristic value of the image output device 32.

  The gradation reproduction processing unit 319 performs halftone processing using error diffusion or dither processing on the image data of the CMYK signal based on the region identification signal (processing to reduce the number of gradations to binary or quaternary). I do. The CMYK signal (image data) binarized or multi-valued by the gradation reproduction processing unit 319 is sent to the image output device 32. The image output device 32 is a device that forms an image on a recording medium such as paper based on the CMYK signal received from the image processing device 31. For example, an electrophotographic or inkjet color image output device can be used.

  The operation panel 33 is an operation device for a user to input an instruction by a key operation or the like. The user's instruction input is input from the operation panel 33 to the image input device 30, the image processing device 31, and the color image output device as control signals. 32. In response to user input, for example, an image input device 30 reads a document image, the read image data is processed by an image processing device 31, and the processed image data is formed on a sheet by an image output device 32. It can be operated as a digital color copying machine. The above processing is controlled by a control unit (CPU) (not shown).

  Next, a schematic configuration of the image quality adjustment unit 317 will be described. Here, a single color data will be described as an example for the sake of explanation, but in the actual process, the same process is performed for the four colors of CMYK. FIG. 2 is a block diagram illustrating a configuration example of the image quality adjustment unit 317. The image quality adjustment unit 317 converts the 256-gradation input image data Pi (X, Y) (CMYK signal) sent from the black generation and under color removal unit 316 into frequency components for each block, and performs frequency component determination. Based on the result, the frequency component is changed, the changed frequency component is converted into image data, and output image data Po (X, Y) of 256 gradations is output to the output gradation correction unit 318. Here, the input image data Pi (X, Y) is on the Y-th line of the image data composed of pixels arranged in a two-dimensional matrix in the X direction (right direction) and the Y direction (down direction). This is pixel data at the Xth pixel position, and a two-dimensional image is constituted by a large number of input image data Pi (X, Y).

  The image quality adjustment unit 317 converts the input image data Pi (X, Y) into frequency components (frequency component coefficients, hereinafter abbreviated as frequency components) Qj (S, T), coefficients of each frequency component Frequency component determining unit (comparing unit) 2 for determining the magnitude of the frequency component, frequency component changing unit (changing unit) 3 for changing the frequency component according to the determination result of the frequency component determining unit 2, and the changed frequency component Qk (S , T), and a control unit such as a CPU (not shown) that controls each of the units, and outputs output image data Po (X, Y) from the reverse frequency conversion unit 6. . Further, the frequency component changing unit 3 changes the specific frequency component Qj (Sa, Ta) of the frequency components Qj (S, T) so as to be enhanced, and other frequency components. It has a smoothing change unit 5 that changes Qj (Sb, Tb) to be a smoothing process.

  The input image data Pi (X, Y) sent from the black generation and under color removal unit 316 is sequentially output to the frequency conversion unit 1 with 8 × 8 pixels as one block, for example, under the control of the control unit. The frequency conversion unit 1 performs conversion (frequency conversion) to a spatial frequency (frequency domain) on the image data output in units of blocks. In this description, a discrete cosine transform (DCT) will be described as an example. The DCT can be performed using the following expression when the input image is Aij, the output image is Bij, and the row and column sizes of the input image A are M and N.

  In this description, the DCT process is performed in block units in the right direction (X direction) from the block including the upper left pixel on the image in the coordinate area (XY coordinate), and the line is changed in block units ( In the Y direction), DCT processing is finally performed up to the final block including the lowermost rightmost pixel. A frequency component (frequency component coefficient; hereinafter referred to as DCT coefficient) Qj (S, T) subjected to DCT processing by the frequency converting unit 1 is sent to the frequency component determining unit 2.

  In the frequency component determination unit (comparison means) 2, the magnitude (absolute value) of a specific AC component (AC component) other than the DC component (DC component) is a predetermined value for each DCT coefficient subjected to DCT processing in the block. Determine if greater than. FIG. 3A is a schematic diagram illustrating an example of a region in which the magnitude of the DCT coefficient in the block is determined. One block includes 8 × 8 DCT coefficients, the upper left corner of the block is the origin, the right direction is the S axis, and the lower direction is the T axis. For example, the upper left corner of the block is Qj (1, 1). Is Qj (8,8). Qj (1, 1) in the upper left is a DC component, and the others are AC components. The upper left side of the block is a low frequency region, and the lower right side of the block is a high frequency region. The first row is a frequency component related only to the horizontal direction of the image data, and the first column is a frequency component related only to the vertical direction. In the illustrated example, Qj (2,1) to Qj (7,1), Qj (1,2) to Qj (6,2), Qj (1,3) to Qj (5,3), Qj (1 , 4) to Qj (4, 4), Qj (1, 5) to Qj (3, 5), Qj (1, 6) to Qj (2, 6), Qj (1, 7) (Specific frequency component region, hereinafter referred to as a determination region).

The frequency component determination unit 2 calculates an absolute value (| Qj (S, T) |) of a DCT coefficient included in the determination region among the DCT coefficients Qj (S, T) in the block, and the calculated absolute value is predetermined. It is determined whether the value is greater than (α> 0) or less. If the absolute value is larger, it is determined that the edge is included, and the emphasis process is performed. If the absolute value is smaller, it is determined that the edge is not included, and the smoothing process is performed.
Emphasis processing when | Qj (S, T) |> α | Smoothing processing when | Qj (S, T) | ≦ α

  In other words, whether to perform enhancement processing or smoothing processing is determined not for each block but for each DCT coefficient in the block. Here, the predetermined value α can be arbitrarily set. When α is small, it is easy to determine that an edge is included, and when α is large, it is difficult to determine that an edge is included. In the present embodiment, α = 16. Note that the method for determining whether or not the DCT coefficient includes an edge is not limited to the above-described method. For example, the absolute value of a value obtained by multiplying the DCT coefficient by the magnitude of the DC component is compared with a predetermined value. It is also possible to use.

  Also, smoothing processing is performed on DCT coefficients other than the determination region in the AC component. FIG. 3B is a schematic diagram illustrating an example of a determination result. A smoothing process is performed outside the AC component determination area (Δ), and within the AC component determination area, a part (●) is emphasized according to the determination result, and the other part (◯) is smoothed. Process. In this way, by switching processing for each DCT coefficient, enhancement processing and smoothing processing can be executed in combination in a block.

  Here, DCT is executed with 8 pixels × 8 pixels as one block, and if a higher luminance is assigned to a natural image having a larger value in each block after DCT conversion, From the result, the luminance of the component corresponding to the low frequency side (upper left part of each block) is increased. That is, the low frequency component has a large amount of information. It is known that a signal having a strong correlation (correlation between a high density and a low density such as an edge) concentrates the signal power in a low frequency region when viewed in frequency. This means that there is a coefficient in which power is concentrated in each DCT coefficient and a coefficient that is not so concentrated. In other words, a large deviation occurs in the size of each DCT coefficient.

  Generally, when DCT is performed on a natural image, the DC component has a particularly large value, and the DC coefficient on the low frequency side also increases in the AC component. For this reason, in the case of an image (natural image or the like) having a feature amount such as having an edge component or having a light and shade of density, the low frequency component around the DC component has a large value. However, when DCT processing is performed on a solid image (an image having no edge component and a uniform density), only the DC component has a value, and the low frequency components around the DC component have values close to zero. Therefore, it is possible to determine whether or not the original image has a feature amount based on the value of the low frequency component. Therefore, when the DCT coefficient is equal to or greater than a predetermined value, it can be regarded as including features such as edge components.

  Further, as shown in Equation 1, when the edge component is conspicuous in the main scanning direction (X direction) or when the edge component is conspicuous in the sub scanning direction (Y direction), the AC component or DC to the right of the DC component The value of the AC component immediately below the component is affected and increases. From the above, in the determination region, when | Qj (S, T) |> α, it can be determined that an edge component is included.

  The frequency component determination unit 2 sends the DCT coefficient Qj (Sa, Ta) for performing the enhancement process among the AC components of the DCT coefficient Qj (S, T) to the enhancement processing unit 4 of the frequency component changing unit (changing unit) 3. The DCT coefficient Qj (Sb, Tb) for performing the smoothing process is sent to the smoothing processing unit 5 of the frequency component changing unit (changing unit) 3. The DCT coefficient Qj (Sa, Ta) for performing enhancement processing is a DCT coefficient that is determined to perform enhancement processing within the determination region, and the DCT coefficient Qj (Sb, Tb) for performing smoothing processing is smoothed within the determination region. The DCT coefficient determined to be processed and the DCT coefficient outside the determination region. Further, although the DC component is sent to the frequency component changing unit 3, neither the enhancement process nor the smoothing process is performed.

The enhancement processing unit 4 changes the DCT coefficient so that the DCT coefficient Qj (Sa, Ta) sent from the frequency component determination unit 2 is enhanced. For example, the DCT coefficient can be changed (increased) by multiplying the DCT coefficient Qj (Sa, Ta) by position data corresponding to the position (coordinates) in the block and a predetermined constant. is there.
DCT coefficient after change Qk (S, T) = DCT coefficient before change Qj (Sa, Ta) × position data × constant

  The position data can be a value (= S + J) obtained by adding the row number S and the column number T of the DCT coefficient Qj (S, T). FIG. 4A is a schematic diagram showing an example of DCT coefficient position data at each position (coordinate) in the determination area in the block. The position data in the case of Qj (1,2) is 1 + 2 = 3, and in the case of Qj (2,1), 2 + 1 = 3.

The constant can be set to 0.35, for example. However, this value (0.35) is just an example, and it may be a certain minute value that is emphasized. In addition, since the degree of emphasis varies depending on this value, it is preferable that the determination is performed while evaluating the image quality using an actual output image in consideration of the overall balance such as whether or not the edge is excessively emphasized. Here, since the DC component is not changed and remains Qj (1, 1), the average density of the entire block is maintained. For example, when Qj (1, 3) and Qj (2, 4) are emphasized,
Qk (1,3) = Qj (1,3) × (1 + 3) × 0.35
Qk (2,4) = Qj (2,4) × (2 + 4) × 0.35
It becomes.
As another example of adjusting the emphasis level, it is also possible to give an offset value (constant) to the position data. In this case, the offset value is 3 (constant value), the constant is 0.20,
Qk (1,3) = Qj (1,3) × (1 + 3 + offset value) × 0.20
Qk (2,4) = Qj (2,4) × (2 + 4 + offset value) × 0.20
However, offset value = 3
You can change the emphasis level.

The smoothing processing unit 5 changes the DCT coefficient so that the DCT coefficient Qj (Sb, Tb) sent from the frequency component determination unit 2 is smoothed. For example, the DCT coefficient can be changed (decreased) by dividing the DCT coefficient Qj (Sb, Tb) by position data corresponding to the position (coordinates) in the block.
Qk (S, T) = Qj (Sb, Tb) / position data

  As the position data, a two-dimensional matrix M (S, T) (S = 1 to 8, T = 1 to 8) corresponding to the block of the DCT coefficient Qj (S, T) can be used. FIG. 4B is a diagram illustrating an example of the matrix M. The matrix M is an 8 × 8 matrix corresponding to 8 × 8 DCT coefficients. In the matrix M, the position data has a minimum value at M (5,5) (where M (5,5)> 1), and the position data around M (5,5) are concentric. It will increase.

  The matrix M shown in FIG. 4B generally has a contrast sensitivity function (CSF) characteristic, that is, a characteristic that reflects human visual characteristics. In general, the sensitivity to human contrast depends on the spatial frequency, and the human perception system is considered as a kind of band-pass filter. For example, when considering a black and white striped pattern, the sensitivity to a human striped pattern varies depending on the interval between successive striped patterns. When the interval between the stripes is very small, it becomes difficult for a human to perceive the stripe pattern. For example, in FIG. 4B, the value of M (S, T) is a value that changes concentrically according to the sensitivity to human contrast, centering on the hatched frequency component (2.1). ing. When the DCT coefficient Qj (S, T) is divided by CSF, the DCT coefficient of the frequency component having high sensitivity to contrast is divided by a larger value than the frequency component having low sensitivity to contrast. Is obtained.

  According to the present embodiment, a frequency component (DCT coefficient) in a block which is a unit of DCT processing is converted so that a component having an edge portion is emphasized and the other components are smoothed. , The edge portion is emphasized, and a flat image portion having no edge portion can obtain a smoother image, and the characteristic portion of the original image can be kept good.

  The two-dimensional matrix used for smoothing is not limited to that shown in FIG. 4B, but a small value is set for a frequency component having low sensitivity to human contrast, and a large value is set for a frequency component having high sensitivity to contrast. A set two-dimensional matrix, in other words, a low-frequency component is maintained, and a two-dimensional matrix in which the outer DCT coefficient is more strongly suppressed concentrically within the 8 × 8 block may be used. For example, a two-dimensional matrix having a Gaussian distribution may be used.

  The AC component that has been subjected to the enhancement process or the smoothing process and the DC component that has not been subjected to the process are sent from the frequency component changing unit 3 to the inverse frequency converting unit 6. The inverse frequency transform unit 6 performs inverse frequency transform on the DCT coefficient Qk (S, T) sent from the frequency component change unit 3 to convert it into image data in the coordinate region. The inverse frequency transform performs the inverse transform of Equation 1.

  As described above, since it is determined whether the edge should be emphasized or smoothed in units of frequency components, not in units of blocks, and conversion is performed while switching emphasis or smoothing for each DCT coefficient. The determination error of each DCT coefficient (frequency component) can be suppressed, the edge component is emphasized, the other components are smoothed, and the image optimization process is realized. FIG. 5A is a schematic diagram showing an example of an output image when all DCT coefficients are changed in units of blocks, and FIG. 5B is an example of an output image when DCT coefficients are changed in units of DCT coefficients. It is a schematic diagram shown. As shown in FIG. 5 (a), when all DCT coefficients are changed in units of blocks, a portion that is crushed into blocks is generated in the image due to an emphasis or smoothing determination error in units of blocks. When the DCT coefficient is changed in units of DCT coefficients, such block-like crushing does not occur.

  In addition, since an image including an edge component is subjected to an emphasis process, a sharper and clearer image can be obtained. For an image not including an edge component, a smoothing process is performed in consideration of visual characteristics. . In addition, the emphasis process is not to emphasize the entire frequency range other than the DC component, but limits the emphasis region, and the emphasis coefficient is set to gradually increase as the distance from the DC component increases. The appearance of an unnecessary block pattern at a certain edge boundary can be suppressed, and a clean edge curve can be obtained.

(Embodiment 2)
FIG. 6 is a block diagram illustrating a configuration example of an image forming system according to the present invention. The image forming system includes a computer 10, an external storage device 18 connected to the computer 10, a display device 20, an operation device 22, an image input device 24, an image output device (image forming means) 26, and a communication device 28. The external storage device 18 is a device that can read data on the recording medium 19 such as a flexible disk drive or a CD-ROM drive. The display device 20 is a device capable of displaying an image such as a CRT (Cathode Ray Tube) display or a liquid crystal display. The operation device 22 is a device that can accept an operation input such as a keyboard or a mouse. The image input device 24 is a device capable of reading image data, such as an image scanner or a digital camera, and is similar to the image input device 30 of the first embodiment (FIG. 1), for example. The image output device 26 is a device capable of outputting an image to a sheet such as an ink jet printer or a laser printer, and is similar to the image output device 32 of the first embodiment (FIG. 1), for example. The communication device 28 is a fax modem or a LAN (Local Area Network) card, and is a device capable of communicating with an external network.

  The computer 10 includes a CPU 12, a RAM 14, and an HDD (hard disk drive) 16. The CPU 12 controls the RAM 14 and the HDD 16 and the devices 18 to 28 described above. Further, the CPU 12 stores data or a program received from the operation device 22 or the communication device 28 or a program or data read from the HDD 16 or the external storage device 18 in the RAM 14, and executes the program stored in the RAM 14 or calculates data. Various processing such as these are performed, and various processing results or temporary data used for various processing are stored in the RAM 14. Data such as calculation results stored in the RAM 14 is stored in the HDD 16 by the CPU 12 or output from the display device 20, the image output device 26, or the communication device 28.

  The computer 10 includes at least an image quality adjustment unit 317 (a frequency conversion unit 1, a frequency component determination unit (comparison unit) 2, a frequency component change unit (change unit) 3), and an image processing device 31 according to the first embodiment (FIG. 1). It operates as an inverse frequency component converter 6). For example, the CPU 12 also operates as the color correction unit 315 of the image processing apparatus 31. Of course, the CPU 12 can be operated as a part other than the color correction unit 315 and the image quality adjustment unit 317 of the image forming apparatus 31. For example, when outputting to the image output device 32, the CPU 12 can be operated as a gradation reproduction processing unit. Image data read by the image input device 24 is stored in the HDD 16, and the CPU 12 performs the same processing as the color correction unit 315 and the image quality adjustment unit 317 described in the first embodiment, and the processed image data is output to the image output device. It is possible to output to 32 or store in the HDD 16. Various data such as a predetermined value α to be compared with each frequency component are also stored in the HDD 16.

  A computer program recorded on a recording medium 19 such as a CD-ROM is read by the external storage device 18, stored in the HDD 16 or RAM 14, and executed by the CPU 12, thereby causing the computer 10 (CPU 12) to operate as the above-described units. Is possible. It is also possible for the communication device 28 connected to a LAN or the like to accept a computer program from another device and store it in the HDD 16 or the RAM 14. Note that the computer program for realizing the change of the frequency component according to the present invention may be included in the printer driver or may be included in the application software for image processing.

  By recording the program for changing the frequency component according to the present invention on the computer-readable recording medium 19, the recording medium 19 on which the program according to the present invention is recorded can be provided in a portable manner.

  FIG. 7 is a flowchart showing a processing procedure when the computer is operated as an image quality adjustment unit. In this example, it is assumed that frequency conversion processing is performed by DCT, frequency components (DCT coefficients) are changed, and image data of 256 gradations is output by inverse frequency conversion. Here, although the actual processing is four colors of CMYK, the processing of each color is the same, so the case of a single color will be described.

  The CPU 12 reads out one block (8 × 8 pixels) of image data from the HDD 16 to the RAM 14 (S10), performs DCT processing (S12), and stores the converted frequency component (DCT coefficient) in the RAM 14. The CPU 12 selects one of the frequency components (S14), and determines whether or not the selected frequency component is included in the determination region. When it determines with not being contained in the determination area | region (S16: NO), CPU12 performs a smoothing process (S22) and memorize | stores the frequency component after a process in RAM14. In the leveling process, the frequency component is changed (decreased) as in the first embodiment. When it is determined that it is included in the determination region (S16: YES), the CPU 12 compares the frequency component with a predetermined value (for example, α = 16) to determine whether or not the edge component is included. When the frequency component is equal to or less than the predetermined value (S18: NO), the CPU 12 performs a smoothing process (S22) and stores the processed frequency component in the RAM. When the frequency component is larger than the predetermined value (S18: YES), the CPU 12 changes the frequency component to perform enhancement processing (S20), and stores the processed frequency component in the RAM 14. In the enhancement process, the frequency component is changed (increased) as in the first embodiment.

  When selection of all the frequency components in the block has not been completed (S24: NO), the CPU 12 selects an unselected frequency component in the block (S26) and performs the same processing (S16 to S24). When selection of all frequency components in the block is completed (S24: YES), the CPU 12 performs inverse DCT processing (S28) and stores the converted image data in the RAM. When reading of all blocks has not been completed (S30: NO), the CPU 12 reads one block of image data into the RAM 14 (S10), and performs the same processing (S12 to S28). If all blocks have been read (S30: YES), the process ends.

  In the present embodiment, the recording medium 19 may be a program medium such as a memory (not shown) such as a ROM itself for processing performed by a microcomputer, or an external storage. A program reading device may be provided as a device, and a program medium that can be read by inserting a recording medium into the device may be used. In any case, the stored program may be configured to be accessed and executed by the microprocessor, or in any case, the program is read and the read program is illustrated in the microcomputer. The program may be downloaded to a non-program storage area and executed. It is assumed that this download program is stored in the main device in advance.

  Here, the program medium is a recording medium configured to be separable from the main body, such as a tape system such as a magnetic tape or a cassette tape, a magnetic disk such as a flexible disk or a hard disk, a CD-ROM / MO / MD / DVD, or the like. Disk systems such as optical disks, card systems such as IC cards (including memory cards) / optical cards, mask ROM, EPROM (Erasable Programmable Read Only Memory), EEPROM (Electrically Erasable Programmable Read Only Memory), flash ROM, etc. It may be a medium that carries a fixed program including a semiconductor memory.

  In the present embodiment, since the system configuration is such that a communication network including the Internet can be connected, a medium that fluidly carries the program so as to download the program from the communication network may be used. When the program is downloaded from the communication network in this way, the download program may be stored in the main device in advance or may be installed from another recording medium.

(Embodiment 3)
The determination area in the block for performing frequency conversion is not limited to the example shown in FIG. 3, and can be arbitrarily set. FIG. 8A is a schematic diagram illustrating another example of the determination region. In the illustrated example, Qj (4,1) to Qj (8,1), Qj (4,2) to Qj (8,2), Qj (4,3) to Qj (8,3), Qj (1 , 4) to Qj (8, 8) are set in the determination region (region of specific frequency component). The AC component in the determination region is subjected to enhancement processing or smoothing processing according to the determination result, while the AC component outside the determination region is subjected to enhancement processing. FIG. 8B is a schematic diagram illustrating an example of a determination result. Emphasis processing is performed in areas other than the AC component determination region (Δ), and in the AC component determination region, part (●) is subjected to enhancement processing according to the determination result, and other part (◯) is subjected to smoothing processing. I do. In the first and second embodiments described above, the frequency component can be changed in the determination region shown in FIG. Comparison, enhancement processing, and smoothing processing of each frequency component (DCT coefficient) and a predetermined value (where α = 32 in the example of FIG. 8) can be performed in the same manner as in the first embodiment.

  In the example of FIG. 3, the low frequency component and the high frequency component in the horizontal direction and the vertical direction are set as the determination region, and emphasis or smoothing is performed according to the determination result, and smoothing is performed except for the determination region. The dot dispersion is good and suitable for photographic images. On the other hand, in the example of FIG. 8, the determination region is set so as to exclude the low frequency component, the low frequency component is emphasized, and the enhancement or smoothing is performed according to the determination result in the other determination regions. Therefore, diagonal lines etc. are clear and suitable for character images.

(Embodiment 4)
The two types of determination areas shown in FIGS. 3 and 8 and the enhancement or smoothing process corresponding to the determination areas can be switched according to the characteristics of the image. For example, in the image forming apparatus shown in FIG. 1, the operation panel 33 accepts a document type such as a text document or a photographic document, and in the case of a photographic document, the determination region shown in FIG. 3 and the enhancement or smoothing corresponding to the determination region. In the case of a text document, control can be performed by a control unit (not shown) so as to perform the determination region shown in FIG. 8 and the enhancement or smoothing processing corresponding to the determination region. Further, for example, a region determination result such as a character region or a photograph region by the region separation processing unit 314 of the image processing apparatus 31 illustrated in FIG. 1 is sent to the image adjustment unit 317. In the case of a photograph region, the determination region and the determination region illustrated in FIG. The image adjustment unit 317 may be configured to perform enhancement or smoothing processing corresponding to the determination region, and in the case of a character region, the determination region shown in FIG. 8 and enhancement or smoothing processing corresponding to the determination region. Is possible.

  Further, the CPU 12 of the computer 10 shown in FIG. 6 is operated as the image adjustment unit 317 and the operation device 22 is operated as the operation panel 33, or the CPU 12 is operated as the region separation processing unit 314 as shown in FIG. It is also possible to switch between the determination area and the enhancement or smoothing process corresponding to the determination area and the determination area and the enhancement or smoothing process corresponding to the determination area shown in FIG.

  In the case of a character document or character region, enhancement or smoothing processing corresponding to the determination region of FIG. 8 suitable for a character image is performed, and in the case of a photographic document or photo region, it corresponds to the determination region of FIG. 3 suitable for a photographic image. By performing the enhancement or smoothing process, the image optimization process can be realized better.

  Further, in each of the above-described embodiments, the smoothing process (FIG. 3) or the enhancement process (FIG. 8) is performed on the frequency components other than the determination region. It is also possible not to change the ingredients. The determination area can be set to an arbitrary area of the AC component. For example, the entire area of the AC component can be set as the determination area.

1 is a block diagram illustrating a configuration example of an image forming apparatus including an image processing apparatus according to the present invention. It is a block diagram which shows one structural example of an image quality adjustment part. (A) is a schematic diagram which shows the example of the area | region which determines the magnitude | size of the DCT coefficient in a block, (b) is a schematic diagram which shows the example of a determination result. (A) is a schematic diagram which shows the example of the position data of the DCT coefficient of each position (coordinate) in the determination area | region in a block, (b) is a figure which shows the example of the matrix M used for a position data. (A) is a schematic diagram showing an example of an output image when all DCT coefficients are changed in units of blocks, and (b) is a schematic diagram showing an example of an output image when DCT coefficients are changed in units of DCT coefficients. is there. 1 is a block diagram illustrating a configuration example of an image forming system according to the present invention. It is a flowchart which shows the process sequence in the case of operating a computer as an image quality adjustment part. (A) is a schematic diagram which shows the other example of a determination area | region, (b) is a schematic diagram which shows the example of a determination result.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Frequency conversion part 2 Frequency component determination part 3 Frequency component change part 4 Enhancement processing part 5 Smoothing process part 6 Inverse frequency conversion part 10 Computer 12 CPU
DESCRIPTION OF SYMBOLS 18 External storage device 19 Recording medium 24, 30 Image input device 26, 32 Image output device 31 Image processing device 317 Image quality adjustment unit 33 Operation panel 40 Image forming device

Claims (8)

In an image processing method for performing conversion to spatial frequency on image data to obtain a plurality of frequency components, processing the obtained frequency components, and inversely converting the processed frequency components into image data,
Comparing the magnitude of the absolute value of the frequency component with a predetermined value for each of a plurality of specific frequency components of the plurality of frequency components obtained by performing the conversion to the spatial frequency;
And possess and changing the specific frequency components based on the comparison result,
The specific frequency component is a frequency component on the low frequency side of all frequency components, a frequency component on the high frequency side only in the horizontal direction of the image data, and a frequency component on the high frequency side only in the vertical direction. A first process of increasing the coefficient of the frequency component and decreasing the coefficient of the other frequency components excluding both the specific frequency component and the direct current component when the absolute value of the coefficient of the frequency component is greater than the predetermined value; And the specific frequency component is a frequency component excluding both a low frequency side frequency component and a direct current component among all frequency components, and the changing means has an absolute value of a coefficient of the specific frequency component as the frequency component. A second process of increasing a coefficient of the frequency component when it is larger than a predetermined value and increasing a coefficient of another frequency component excluding both the specific frequency component and the DC component; An image processing method characterized by switching in accordance with the characteristics of the image data. In an image processing apparatus that converts a spatial frequency to image data to obtain a plurality of frequency components, performs processing of the obtained frequency components, and inversely converts the processed frequency components into image data.
Comparison means for comparing the magnitude of the absolute value of the frequency component with the predetermined value for each of a plurality of specific frequency components of the plurality of frequency components obtained by performing the conversion to the spatial frequency,
Changing means for changing each of the specific frequency components based on a comparison result of the comparing means ;
The specific frequency component is a frequency component on the low frequency side of all frequency components, a frequency component on the high frequency side only in the horizontal direction of the image data, and a frequency component on the high frequency side only in the vertical direction, and the change When the absolute value of the coefficient of the specific frequency component is larger than the predetermined value, the means increases the coefficient of the frequency component and decreases the coefficient of other frequency components excluding both the specific frequency component and the DC component. The first processing to be performed, and the specific frequency component is a frequency component excluding both a frequency component on a low frequency side and a direct current component among all frequency components, and the changing means is a coefficient of the specific frequency component. If the absolute value of is greater than the predetermined value, the coefficient of the frequency component is increased, and the coefficients of other frequency components excluding both the specific frequency component and the DC component are increased. The image processing apparatus characterized by comprising second handle to, and switching means for switching in accordance with the characteristics of the image data. The switching means is configured to switch to the first process when the image data is image data derived from a photographic document, and to switch to the second process when the image data is image data derived from a text document. The image processing apparatus according to claim 2, wherein the image processing apparatus is an image processing apparatus. When the absolute value of the coefficient of the specific frequency component is larger than the predetermined value, the changing means sets a correction coefficient that is smaller as the frequency component is lower on the low frequency side and larger as the frequency component is higher side as the coefficient of the frequency component. The image processing apparatus according to claim 2, wherein the image processing apparatus is configured to perform multiplication. The changing unit, when the absolute value of the coefficient of the specific frequency component is smaller than the predetermined value, any one of claims 2 to 4, characterized in that are configured to reduce the coefficient of said frequency components The image processing apparatus according to one. An image processing device according to any one of claims 2 to 5 ,
An image forming apparatus comprising: an image forming unit that forms an image processed by the image processing apparatus on a sheet. In a computer program for causing a computer to calculate a plurality of frequency components by converting image data to a spatial frequency, to process the obtained frequency components, and to inversely convert the processed frequency components to image data.
A procedure for causing a computer to compare the magnitude of an absolute value of a frequency component with a predetermined value for each of a plurality of specific frequency components of a plurality of frequency components obtained by performing conversion to a spatial frequency,
A procedure for causing the computer to change each of the specific frequency components based on the comparison result ;
In the computer, the specific frequency component is a frequency component on the low frequency side of all frequency components, a frequency component on the high frequency side only in the horizontal direction of the image data, and a frequency component on the high frequency side only in the vertical direction. When the absolute value of the coefficient of the specific frequency component is larger than the predetermined value, the coefficient of the frequency component is increased and the coefficient of the other frequency component excluding both the specific frequency component and the DC component is decreased. 1 processing, and the specific frequency component is a frequency component excluding both a low frequency side frequency component and a direct current component among all frequency components, and the changing means is configured to calculate the absolute value of the coefficient of the specific frequency component. When the value is larger than the predetermined value, the coefficient of the frequency component is increased, and the coefficients of other frequency components excluding both the specific frequency component and the DC component are increased. Computer program the second process, characterized in that it comprises a procedure to switch in accordance with the characteristics of the image data to be. A computer-readable recording medium on which the computer program according to claim 7 is recorded.

Images