JP4067538B2 - 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 for performing image processing by changing a coefficient of a spatial frequency component of image data, an image processing apparatus for executing the image processing method, an image forming apparatus provided with the image processing apparatus, and a computer as the image processing apparatus. The present invention relates to a computer program that functions as a computer program, and a computer-readable recording medium that records the computer program.

  As a method of reducing the number of gradations of an image including a halftone, for example, a method of binarizing a 256 gradation image into a two gradation image, the gradation value of the image is compared with a predetermined value, that is, a threshold value. There are a binarization method, a dither method, an error diffusion method, and the like (Patent Document 1, Patent Document 2, and Patent Document 3).

  FIG. 26 is a diagram illustrating an example of a 4 × 4 dither matrix used in a dither method for binarizing a 256-gradation image. In the dither matrix, a predetermined value from 0 to 240 is set according to the position of the pixel. In the dither method, the gradation value of the input 256 gradation image data is compared with a predetermined value set in the dither matrix for each pixel, and if the gradation value is equal to or larger than the predetermined value, the gradation value is determined. When the gradation value is less than a predetermined value, the image is binarized by setting the gradation value to 0.

  In the error diffusion method, a quantization error generated when binarizing each pixel of input image data, that is, a difference between a binarized gradation value and a gradation value of a pixel before binarization is not yet determined. This is a method of binarizing by distributing to pixels that have not been binarized. When the pixel to be binarized is the target pixel, the quantization error of the target pixel is weighted according to the relative position from the target pixel and then before binarization processing positioned around the target pixel. Is added to the gradation value of the pixel.

  FIG. 27 is a diagram illustrating an example of a weighting coefficient matrix used in the error diffusion method. In the example of FIG. 27, the horizontal direction, that is, the horizontal direction in the figure is the X direction and the vertical direction, that is, the vertical direction in the figure is the Y direction, and a 3 × 2 weighting coefficient matrix including the pixel of interest (IX, IY) is shown. ing. The weighting coefficient matrix has weighting coefficients on the lower left side, the lower side, the lower right side, and the right side on the basis of the target pixel (IX, IY). For example, the gradation value of the pixel of interest (IX, IY) is compared with a predetermined value, and when the gradation value is equal to or greater than the predetermined value, the gradation value of the pixel of interest (IX, IY) is set to 255, and less than the predetermined value. In this case, the gradation value of the target pixel (IX, IY) is set to 0. Next, the difference between the binarized 255 or 0 gradation value and the gradation value of the pixel of interest (IX, IY) before binarization, that is, the quantization error is calculated based on the weighting coefficient matrix. Are distributed to pixels before binarization processing. However, since the pixel (IX-1, IY) adjacent to the left of the target pixel (IX, IY) is quantized before the target pixel (IX, IY), the quantization error is not distributed.

  When the quantization error is Err, the four neighboring (IX + 1, IY), (IX + 1, IY + 1), (IX, IY + 1), and (IX-1, IY + 1) of the pixel of interest (IX, IY) have Err × (7/16), Err × (1/16), Err × (5/16), Err × (3/16) are allocated.

  Since the quantization error is distributed to surrounding unprocessed pixels based on the weight coefficient matrix, the error diffusion method has an advantage that a moire pattern is less likely to appear in the binarized image.

In addition, according to the method described in Japanese Patent Laid-Open No. 2002-10085, image data is converted into image data having a spatial frequency component, and a halftone determined in advance for the coefficient of the converted spatial frequency component. The halftone processing of the image is performed using the data converted into the spatial frequency domain.
Japanese Patent Laid-Open No. 2000-299783 Japanese Patent Laid-Open No. 06-189119 Japanese Patent Laid-Open No. 2002-10085

  However, according to the dither method, since binarization is performed using a dither matrix having the same pattern, there is a problem that a dither-specific texture, that is, a periodic pattern image is generated in the binarized image.

  In addition, according to the error diffusion method, the error is diffused based on the same matrix for each pixel, so that a large quantization error is diffused in a chained manner in a highlight portion having a large gradation value, and the pixels are connected. was there. That is, there is a problem that pixels having different gradation values have the same gradation value, and an image in which the pixels are partially connected is generated.

  Furthermore, in the invention described in Japanese Patent Application Laid-Open No. 2002-10085, since the halftone data determined in advance is used, the occurrence of texture or the like similar to the error diffusion method or the dither method occurs. There's a problem. That is, since the halftone process is merely performed in the frequency domain using the same method as the above-described conventional method, the same problem as described above occurs.

  The present invention has been made in view of such circumstances, and after changing the coefficient of the AC component according to the characteristics of the image, by adding a specific value to the coefficient of the AC component, that is, by adding noise to the image, An image processing method capable of reducing the number of gradations of an image while solving the problem of connection between pixels of the error diffusion method and maintaining a good characteristic portion of the original image, and an image for executing the image processing method It is an object of the present invention to provide a processing apparatus, an image forming apparatus including the image processing apparatus, a computer program that causes a computer to function as the image processing apparatus, and a computer-readable recording medium that records the computer program.

  Also, the presence / absence of the edge portion of the image is determined based on the coefficient of the predetermined frequency component of the image data and the average gradation value of the image. In particular, the presence / absence of the edge portion is determined by changing the determination criterion of the edge portion according to the average gradation value of the image After changing the coefficient of the AC component according to the determined result, adding a specific value to the coefficient of the AC component, that is, adding noise to the image, the problem of connection between pixels that the error diffusion method has An image processing method capable of reducing the gradation value of an image while maintaining the characteristic portion of the original image satisfactorily by image processing according to the presence or absence of an edge portion of the image, and an image for executing the image processing method A processing apparatus, an image forming apparatus including the image processing apparatus, a computer program that causes a computer to function as the image processing apparatus, and the computer program are recorded The is another object to provide a computer readable recording medium.

  Furthermore, the configuration is such that the presence / absence of an edge portion is determined based on the magnitude of the product obtained by multiplying the coefficient of the AC component on the low frequency side by the coefficient of the DC component, thereby allowing the coefficient of the AC component on the high frequency side to be determined. It is possible to determine whether or not the image to be processed has an edge part more accurately than when using the image, and to maintain the characteristic part of the original image better, while maintaining the gradation value of the image. Another object of the present invention is to provide an image processing apparatus capable of reducing the image quality and an image forming apparatus including the image processing apparatus.

  Furthermore, when it is determined that the coefficient of the predetermined frequency component or the absolute value of the product is greater than or equal to the predetermined value, that is, when it is determined that the image includes an edge portion, the image processing device is configured to increase the coefficient of the AC component. Image processing apparatus capable of selectively emphasizing an image having an edge portion and reducing the number of gradations of the image while maintaining a good characteristic portion of the original image, and the image processing apparatus Another object is to provide an image forming apparatus including the above.

  Furthermore, when it is determined that the coefficient of the predetermined frequency component or the absolute value of the product is greater than or equal to the predetermined value, that is, when it is determined that the image includes an edge portion, an AC component having only a horizontal spatial frequency component and By increasing the coefficient of each alternating current component having only the spatial frequency component in the vertical direction, the image having the edge portion is selectively emphasized, and the number of gradations of the image is reduced while keeping the characteristic portion of the original image good. An image processing method capable of reducing the occurrence of a block pattern in a curved portion of an image with a reduced number of gradations, an image processing apparatus for executing the image processing method, and the image processing apparatus An image forming apparatus, a computer program that causes a computer to function as the image processing apparatus, and a computer that records the computer program To provide a readable recording medium and other purposes.

  Furthermore, an image processing method capable of more effectively enhancing an edge portion without losing a characteristic portion of the image by multiplying a coefficient of the AC component by a larger real number as the frequency of the AC component is higher, the image An image processing apparatus that executes the processing method, an image forming apparatus including the image processing apparatus, a computer program that causes a computer to function as the image processing apparatus, and a computer-readable recording medium that records the computer program are provided. The purpose.

  Furthermore, when it is determined that the coefficient of the predetermined frequency component or the absolute value of the product is not equal to or greater than the predetermined value, that is, when it is determined that the image is a flat image that does not include an edge portion, the image of the AC component coefficient is reduced. By configuring the processing device, an image processing device capable of selectively smoothing an image that does not include an edge portion and maintaining a good characteristic portion of the original image while reducing the number of gradations of the image, and the image processing device Another object is to provide an image forming apparatus including an image processing apparatus.

  Furthermore, when it is determined that the image to be processed is an image having no edge portion, when the absolute value of the coefficient of the predetermined frequency component is less than the predetermined value, that is, when the image is a flat image, An image having an edge portion by reducing the gradation coefficient without changing the coefficient of the AC component when the coefficient of the component is reduced and the absolute value of the coefficient of the predetermined frequency component is greater than or equal to the predetermined value. For the image, the edge part is emphasized and the image gradation value is maintained while smoothing the image especially for the flat image among the images having no edge part, while maintaining the characteristic part of the original image in good condition. Another object of the present invention is to provide an image processing apparatus and an image forming apparatus including the image processing apparatus.

An image processing method according to the present invention converts image data into image data having a spatial frequency component, and changes the coefficient of the spatial frequency component to perform image processing. In the image processing method, the predetermined frequency component of the converted image data It is determined whether or not the absolute value of the coefficient is greater than or equal to a first predetermined value, and when it is determined that the absolute value is greater than or equal to the first predetermined value, the coefficient of the AC component of the image data is greater than 1. By multiplying a large real number or dividing the AC component coefficient of the image data by a positive real number smaller than 1, the coefficient is changed, and the coefficient of the AC component of the image data changed by calculation is changed to the spatial frequency of blue noise. by adding the coefficient value of the component, the image data to which the coefficient value has been added, and converted back to image data having a spatial coordinate components, the gradation value of image data obtained by inverse transformation compared to a second predetermined value, And converting the gradation value corresponding to the compare result.

An image processing method according to the present invention converts image data into image data having a spatial frequency component, and changes the coefficient of the spatial frequency component to perform image processing. In the image processing method, the predetermined frequency component of the converted image data When the absolute value of the product obtained by multiplying the coefficient by the coefficient of the DC component is greater than or equal to a first predetermined value, and it is determined that the absolute value is greater than or equal to the first predetermined value An image obtained by multiplying the coefficient of the alternating current component of the image data by a real number larger than 1 or dividing the coefficient of the alternating current component of the image data by a positive real number smaller than 1 and changing the coefficient by calculation the coefficient of the AC component of the data by adding the coefficient value of the spatial frequency components of the blue noise image data image data to which the coefficient value has been added, which is converted back to image data having a spatial coordinate components, and inverse transform The gradation value is compared with a second predetermined value, and converting the gradation value corresponding to the comparison result.

In the image processing method according to the present invention, when it is determined that the absolute value is equal to or greater than the first predetermined value, an AC component having only a horizontal spatial frequency component and an AC component having only a vertical spatial frequency component By multiplying each coefficient by a real number larger than 1, the coefficient is changed by calculation.

In the image processing method according to the present invention, when the absolute value is determined to be greater than or equal to the first predetermined value, the coefficient is multiplied by a larger (smaller) real number as the frequency of the AC component is higher (lower). The coefficient is changed by calculation.

An image processing apparatus according to the present invention includes a conversion unit that converts image data into image data having a spatial frequency component, and performs image processing by changing a coefficient of the spatial frequency component. Determining means for determining whether an absolute value of a coefficient of a predetermined frequency component of the image data converted by the converting means is equal to or greater than a first predetermined value; and, if the absolute value is equal to or greater than the first predetermined value, If the determination means makes a determination, the coefficient is changed by multiplying the coefficient of the AC component of the image data by a real number larger than 1 or dividing the coefficient of the AC component of the image data by a positive real number smaller than 1. arithmetic means, adding means for adding coefficient values of the spatial frequency components of the blue noise to the coefficient of the AC component of the image data changed by said computing means, by said adding means, the coefficient value is pressurized The image data, compares the inverse conversion means for inverse converting the image data having a spatial coordinate components, the gradation value of image data obtained by inversely transformed by inverse transformation means and the second predetermined value, according to the comparison result And a means for converting to a gradation value.

An image processing apparatus according to the present invention includes a conversion unit that converts image data into image data having a spatial frequency component, and performs image processing by changing a coefficient of the spatial frequency component. Determining means for determining whether or not an absolute value of a product obtained by multiplying a coefficient of a direct current component by a coefficient of a predetermined frequency component of image data converted by the converting means is equal to or greater than a first predetermined value; Is determined to be greater than or equal to the first predetermined value, the AC component coefficient of the image data is multiplied by a real number greater than 1, or the AC component coefficient of the image data is a positive value less than 1. by dividing a real number, and calculating means for changing the said coefficients, and adding means for adding the coefficient value of the spatial frequency components of the blue noise to the coefficient of the AC component of the image data changed by said computing means, The addition means, the image data to which the coefficient value has been added, and inverse conversion means for inverse converting the image data having a spatial coordinate component, a predetermined gradation value of the inverse transformed image data of the second by inverse transformation means Means for comparing with a value and converting it to a gradation value according to the comparison result.

In the image processing apparatus according to the present invention, the conversion unit converts the image data into image data having a spatial frequency component in a predetermined frequency range, and the determination unit is a coefficient of an AC component on the low frequency side. It is characterized in that it is determined whether or not an absolute value of a product obtained by multiplying a coefficient of a direct current component is equal to or greater than the first predetermined value.

In the image processing apparatus according to the present invention, when the determination unit determines that the calculation unit determines that the absolute value is equal to or greater than the first predetermined value, an AC component having only a horizontal spatial frequency component and a vertical direction The coefficient of each of the AC components having only the spatial frequency component is multiplied by a real number greater than 1.

  The image processing apparatus according to the present invention is characterized in that the computing means multiplies the coefficient by a larger (smaller) real number as the frequency of the AC component is higher (lower).

In the image processing apparatus according to the present invention, when the determination unit determines that the absolute value is not equal to or greater than the first predetermined value, the arithmetic unit is a positive real number smaller than 1 in the coefficient of the AC component of the image data. Or the coefficient of the AC component of the image data is divided by a real number larger than 1.

In the image processing apparatus according to the present invention, the determination unit includes a unit that determines whether an absolute value of a coefficient of a predetermined frequency component of the image data converted by the conversion unit is equal to or greater than the first predetermined value. The adding means determines that the absolute value of the product is not equal to or greater than the first predetermined value, and the absolute value of the coefficient is equal to or greater than the first predetermined value; The coefficient value of the spatial frequency component of blue noise is added to the coefficient of the AC component of the image data converted by the conversion means, and the calculation means is characterized in that the determination means is such that the absolute value of the product is the first value. If it is determined that it is not greater than or equal to a predetermined value and the absolute value of the coefficient is not greater than or equal to the first predetermined value, the coefficient of the alternating current component of the image data converted by the conversion means is multiplied by a positive real number smaller than 1, Or the exchange of image data Wherein the coefficients of the are to be divided by a real number larger than 1.

  An image forming apparatus according to the present invention includes the image processing apparatus, and is configured to form an image based on image data processed by the image processing apparatus.

A computer program according to the present invention is a computer program for causing a computer to convert image data into image data having a spatial frequency component and changing the coefficient of the spatial frequency component to perform image processing. Determining whether the absolute value of the coefficient of the predetermined frequency component is greater than or equal to a first predetermined value, and if the computer determines that the absolute value is greater than or equal to the first predetermined value, Multiplying the coefficient of the alternating current component of the image data by a real number larger than 1 or dividing the coefficient of the alternating current component of the image data by a positive real number smaller than 1; step of adding the coefficient value of the spatial frequency components of the blue noise to the coefficient of the AC component of the modified image data was , Compared to a computer, the image data to which the coefficient value has been added, a step of inverse transformation on the image data having a spatial coordinate components, the computer, the inverse transformed image data tone value and a second predetermined value And a step of converting to a gradation value corresponding to the compared result.

A computer program according to the present invention is a computer program for causing a computer to convert image data into image data having a spatial frequency component and changing the coefficient of the spatial frequency component to perform image processing. Determining whether the absolute value of the product obtained by multiplying the coefficient of the predetermined frequency component by the coefficient of the DC component is equal to or greater than a first predetermined value, and causing the computer to determine whether the absolute value is the first value If it is determined that the value is greater than or equal to a predetermined value, the AC component coefficient of the image data is multiplied by a real number greater than 1, or the AC component coefficient of the image data is divided by a positive real number less than 1. a step of changing the coefficients, the computer, the spatial frequency of the blue noise to the coefficient of the AC component of the image data is changed by the operation A step of adding the partial coefficient values, the computer, the image data to which the coefficient value has been added, a step of inverse transformation on the image data having a spatial coordinate components, the computer, the gradation of the inverse transformed image data Comparing the value with a second predetermined value and converting the value to a gradation value corresponding to the result of the comparison.

The computer program according to the present invention may cause the computer to multiply the AC component coefficient of the image data by a real number greater than 1 or divide the AC component coefficient of the image data by a positive real number less than 1. The step of changing a coefficient has only an AC component having only a horizontal spatial frequency component and a vertical spatial frequency component in the computer when the absolute value is determined to be equal to or greater than the first predetermined value. The step of multiplying the coefficient of each alternating current component by a real number larger than 1 is characterized.

The computer program according to the present invention may cause the computer to multiply the AC component coefficient of the image data by a real number greater than 1 or divide the AC component coefficient of the image data by a positive real number less than 1. The step of changing the coefficient includes a step of causing the computer to multiply the coefficient by a larger (smaller) real number as the frequency of the AC component is higher (lower).

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

  In the present invention, by determining whether or not the absolute value of the coefficient of the predetermined frequency component among the coefficients of the spatial frequency component included in the image data is greater than or equal to the predetermined value, the target image area has an edge portion. Whether or not the image is an image can be determined. Then, by changing the coefficient of the AC component according to the result of determining the magnitude of the absolute value, the image is enhanced or smoothed according to the characteristics of the image processing target, and the image after the image processing is processed. The gradation value is decreased based on a predetermined value. Therefore, it is possible to reduce the number of gradations of the image while keeping the characteristic portion of the image good. Note that the method of reducing the number of gradations compared to the predetermined value is not limited to the threshold method, but includes a dither method, an error diffusion method, or the like.

In addition, by adding the coefficient value of the spatial frequency component of blue noise having frequency characteristics that are difficult for the human eye to perceive to the AC component coefficient, the gradation value of each pixel in the highlight portion where the gradation value is large It is possible to prevent the pixels from being connected to each other. Further, it is possible to improve the texture in the intermediate density region. It is not necessary to add the coefficient values for all coefficients of the AC components may be added to the coefficient values in a part of the coefficients of the AC components. Moreover, there is no need for adding the coefficient value of the same value for all coefficients of the AC component.

In the present invention, it is determined whether or not the absolute value of the product obtained by multiplying the coefficient of the predetermined frequency component by the coefficient of the direct current component among the coefficients of the spatial frequency component of the image data is equal to or greater than the predetermined value. Thus, it is determined whether or not the image to be processed is an image having an edge portion.
The coefficient of the predetermined frequency component is larger as the gradation difference of the edge portion of the image is larger, and the absolute value of the product is larger.
The coefficient of the direct current component corresponds to the average gradation value of the image. The larger the average gradation value of the image, the larger the coefficient of the direct current component. Accordingly, the larger the average gradation value of the image, that is, the brighter the image, the larger the product, so that it is easy to determine that the image to be processed is an image having an edge portion.
On the other hand, since it becomes easier for a person to recognize an edge portion when the image is brighter than when the image is dark, even if the image has the same coefficient of the predetermined frequency component, the larger the average gradation value of the image, Strongly recognize the edge part.
Accordingly, since the strength of the edge portion recognized by the person corresponds to the magnitude of the absolute value, the presence / absence of the edge portion according to the human recognition characteristic is determined by determining the presence / absence of the edge portion using the product. Is possible.
For example, when a white image with a large DC component coefficient contains light gray characters, the absolute value of the AC component coefficient is small because the gradation difference in the edge portion is small, but the DC component coefficient is large. The absolute value of the product increases, and the image is determined to be an image having an edge portion.
Then, by changing the coefficient of the AC component according to the result of determining the magnitude of the absolute value, the feature of the image processing target, that is, the enhancement or smoothing of the image according to the presence or absence of the edge portion, The gradation value of the image after image processing is decreased based on a predetermined value.
Therefore, it is possible to reduce the gradation value of the image while keeping the characteristic portion of the image better.

  In the present invention, by determining whether or not the absolute value of the product obtained by multiplying the coefficient of the alternating current component on the low frequency side in the predetermined frequency range by the coefficient of the direct current component is greater than or equal to the predetermined value, It is determined whether or not the image to be processed is an image having an edge portion. The coefficient of the AC component on the low frequency side is related to the strength of the edge portion compared to the AC component on the high frequency side. Therefore, when the edge portion is determined using the coefficient of the AC component on the low frequency side, it can be determined more accurately whether the image has the edge portion.

  In the present invention, when the determining means determines that the absolute value of the product of the predetermined frequency component or the product of the predetermined frequency component multiplied by the DC component coefficient is greater than or equal to the predetermined value, By multiplying the coefficient by a real number greater than 1 or dividing the AC component coefficient by a positive real number less than 1, the AC component coefficient can be increased to enhance the edge portion of the image including the edge portion. . That is, it is possible to perform processing for selecting an image including an edge portion from images to be processed and emphasizing the edge portion of the image. Therefore, since the characteristic part of the image is relatively emphasized, the number of gradations of the image can be reduced while keeping the characteristic part of the original image good.

  In the present invention, when it is determined that the coefficient of the predetermined frequency component or the absolute value of the product is equal to or greater than the predetermined value, that is, when it is determined that the image includes an edge portion, the spatial frequency component in the horizontal direction The coefficient value is changed by multiplying the coefficient of each of the alternating current component having only the spatial component and the alternating current component having only the spatial frequency component in the vertical direction by a real number larger than 1. In this case, since the coefficient of the AC component is larger than the coefficient before the change, the edge portion of the image is emphasized. The present invention includes a case where the value of the coefficient is changed by dividing the coefficient of the AC component by a real number smaller than 1.

  Also, instead of changing the coefficients of all AC components, the coefficients of the AC component having only the horizontal spatial frequency component and the AC component having only the vertical spatial frequency component are changed. Even after the decrease is made, it is possible to suppress the occurrence of an unnecessary block pattern in the curved portion of the image of the density region, that is, the edge portion related to the curve, and an image having a cleaner curved portion can be obtained. The same effect can be obtained when the coefficient is divided by a positive real number smaller than 1.

  In the present invention, the higher (lower) the frequency of the AC component, the larger (smaller) real number is multiplied by the coefficient of the AC component. Compared with the case where the coefficient of the AC component having a low frequency is changed, the case where the coefficient of the AC component having a high frequency is changed can effectively enhance the edge portion while retaining the characteristic portion of the image. Therefore, it is possible to obtain an image with more effective edges after threshold processing.

  In the present invention, when the determining means determines that the coefficient of the predetermined frequency component or the absolute value of the product is not equal to or greater than the predetermined value, the AC component coefficient is multiplied by a positive real number smaller than 1, or the AC component coefficient Is divided by a real number larger than 1, the coefficient of the AC component can be reduced, and an image that does not include an edge portion can be smoothed. That is, it is possible to perform processing for selecting a flat image from the images to be processed and smoothing the image. Therefore, since the characteristic part of the image is relatively emphasized, the number of gradations of the image can be reduced while keeping the characteristic part of the original image good.

In the present invention, by determining whether or not the absolute value of the coefficient of the predetermined frequency component is greater than or equal to a predetermined value for an image whose absolute value of the product obtained by the multiplication is less than a predetermined value, It is determined whether or not the image to be processed is a flat image, that is, an image having a small gradation difference in the entire image.
For an image having an edge portion, the image is enhanced by increasing the coefficient of the AC component, and for a flat image, the image is smoothed by decreasing the coefficient of the AC component. For an image that does not have an edge portion and is not flat, neither the image enhancement nor the smoothing is performed, and the processing subsequent to the addition of coefficient values related to blue noise is executed. For the image, since it is difficult to determine which processing of enhancement or smoothing is appropriate for enhancing the feature of the image, the enhancement and smoothing of the image are not performed.

  Therefore, by emphasizing the edge portion and further smoothing the flat image, it is possible to reduce the gradation value of the image while keeping the feature portion of the image good by enhancing the feature of the image.

  In the image processing method, the image processing apparatus, the image forming apparatus, the computer program, and the recording medium according to the present invention, the image processing is performed according to the characteristics of the image, and the pixel in the highlight portion where the gradation value is large. It is possible to prevent the images from being connected to each other, and it is possible to reduce the number of gradations of the image while maintaining a good characteristic portion of the original image.

In the image processing method, the image processing apparatus, the image forming apparatus, the computer program, and the recording medium according to the present invention, the image processing is performed according to the characteristics of the image, and the pixel in the highlight portion where the gradation value is large. It is possible to prevent the images from being connected to each other, and it is possible to reduce the number of gradations of the image while maintaining a good characteristic portion of the original image.
In particular, since the presence or absence of the edge portion is determined according to the average gradation value of the image, it is possible to perform image processing according to the brightness of the image and the gradation difference of the edge portion.

  In the image processing apparatus and the image forming apparatus according to the present invention, whether or not the image to be processed has an edge portion more accurately than in the case where the determination is made using the coefficient of the AC component on the high frequency side. It is possible to reduce the gradation value of the image while keeping the characteristic portion of the original image better.

  In the image processing apparatus and the image forming apparatus according to the present invention, an image including an edge portion is emphasized, and a feature portion of the image is relatively emphasized, thereby maintaining a good feature portion of the original image. The number of gradations of the image can be reduced.

  In the image processing method, the image processing apparatus, the image forming apparatus, the computer program, and the recording medium according to the present invention, a block pattern is generated in the curved portion of the image as compared with the case where all the AC component coefficients are changed. This can be effectively prevented, and the number of gradations of the image can be reduced while keeping the characteristic portion of the original image better.

  In the image processing method, the image processing apparatus, the image forming apparatus, the computer program, and the recording medium according to the present invention, the edge portion can be effectively emphasized without losing the characteristic portion of the image, and the original image The number of gradations of the image can be reduced while maintaining the characteristic portion of the image.

  In the image processing apparatus and the image forming apparatus according to the present invention, a flat image that does not include an edge portion is smoothed, and a characteristic portion of the original image is improved by relatively enhancing the characteristic portion of the image. The number of gradations of the image can be reduced while maintaining the above.

  Further, in the method of reducing the number of gradations of an image including a halftone, for example, the number of gradations is reduced from an image of 256 gradations to an image of 4 gradations. Therefore, when the digital copying machine performs spatial filtering in the preceding stage and performs processing to reduce the number of gradations of the image after filtering, or the computer performs filtering using image editing software, and the image level after filtering is processed. When performing the process of reducing the logarithm, there is a risk that the effect of the filter process may be weakened due to the decrease in the number of gradations. However, the image processing apparatus and the image forming apparatus according to the present invention include an edge portion. By emphasizing the image or smoothing a flat image that does not include an edge portion and relatively enhancing the feature portion of the image, it is possible to suppress a reduction in the effect of the filtering process. It is possible to reduce the number of gradations of the image while maintaining good.

  In the image processing apparatus and the image forming apparatus according to the present invention, an edge portion is emphasized for an image having an edge portion, and particularly an image having no edge portion is particularly flat. By smoothing the image, it is possible to reduce the gradation value of the image while maintaining a good characteristic portion of the original image.

(Embodiment 1)
Hereinafter, an image processing method and an image processing apparatus according to Embodiment 1 of the present invention will be described in detail with reference to the drawings illustrating the embodiments. FIG. 1 is a block diagram showing an image processing apparatus according to the present invention. The image processing apparatus is an apparatus that generates and outputs image data Po (X, Y) in which the number of gradations of image data Pi (X, Y) having a spatial coordinate component input to the image processing apparatus is reduced. . For example, image data Po (X, Y) in which 256 gradation image data Pi (X, Y) is reduced to 4 gradations is generated and output. Here, the image data Pi (X, Y) is a gradation value of an image formed by pixels arranged in a two-dimensional matrix of X and Y directions perpendicular to each other, that is, horizontal lines and vertical lines. X indicates the position on the horizontal line in the image, and Y indicates the position on the vertical line in the image.

  The image processing apparatus includes an image data storage unit 11 for temporary storage that stores input image data Pi (X, Y). The image data Pi (X, Y) stored in the image data storage unit 11 is converted by the frequency conversion unit 12 into image data having a spatial frequency component, that is, DCT coefficient Qj (S, T), discrete cosine transform (DCT: Discrete Cosine). And is output to the frequency component determination unit 13 and the change unit 14. The frequency component determination unit 13 determines the magnitude of the coefficient of the predetermined spatial frequency component, and outputs the determined result to the change unit 14. Based on the determination result of the frequency component determination unit 13, the changing unit 14 partially changes the value of the DCT coefficient Qj (S, T) subjected to discrete cosine transform by the frequency conversion unit 12, and changes the changed DCT coefficient Qk (S, T) is output to the noise adding unit 15. The noise adding unit 15 adds the DCT coefficient of the image having the blue noise characteristic to the DCT coefficient Qk (S, T), and outputs the DCT coefficient Ql (S, T) obtained by adding the DCT coefficient to the inverse frequency converting unit 16. To do. The DCT coefficient of an image having blue noise characteristics is an example of a specific value. The inverse frequency transform unit 16 performs inverse frequency transform on the noise-added DCT coefficient Ql (S, T) to generate image data Pm (X, Y), and performs threshold processing on the generated image data Pm (X, Y). To the unit 17. The threshold processing unit 17 compares the gradation value of the image data Pm (X, Y) with a predetermined value, and sets the gradation value of the image data Pm (X, Y) to four values, for example, 0, according to the comparison result. 85, 171 and 255, and the changed image data Po (X, Y) is output to the outside. The image data storage unit 11, the frequency conversion unit 12, the frequency component determination unit 13, the change unit 14, the noise addition unit 15, the inverse frequency conversion unit 16, and the threshold processing unit 17 are controlled by the control unit 10 of the microcomputer. .

  Hereinafter, the function of each component will be described more specifically. Image data Pi (X, Y) input to the image processing apparatus is sequentially stored in the image data storage unit 11. The image data Pi (X, Y) stored in the image data storage unit 11 is sequentially output to the frequency conversion unit 12 under the control of the control unit 10 with the pixel groups in the X direction and the 8 × 8 direction as unit blocks. The

  The frequency converter 12 converts the image data Pi (X, Y) output for each unit block from the image data storage unit 11 into image data Qj (S, T) having a spatial frequency component. More specifically, the frequency conversion unit 12 receives image data Pi (X, Y) having 8 × 8 pixels as a unit block, performs discrete cosine transform on the image data Pi (X, Y), and performs discrete cosine transform. The DCT coefficient Qj (S, T) is output to the frequency component determination unit 13 and the change unit 14. The discrete cosine transform is expressed by Equation (1). Here, S represents the frequency in the X direction, T represents the frequency in the Y direction, M represents the number of pixels in the X direction in the unit block, and N represents the number of pixels in the Y direction in the unit block. In the present embodiment, M = N = 8.

  The frequency conversion unit 12 performs a discrete cosine transform for each unit block in the X direction from the block including the uppermost pixel on the two-dimensional image to be image-processed, and finally changes the line in the Y direction. In particular, the discrete cosine transform is performed up to the unit block including the lower right pixel.

  FIG. 2 is a conceptual diagram showing the DCT coefficients converted into spatial frequency components by the frequency converter 12 using 8 × 8 sections. A section located at the upper left corner and including a black circle mark indicates a DC component of a DCT coefficient, and other sections excluding the DC component indicate an AC component. The S axis indicates the magnitude of the frequency in the X axis direction in the spatial image, and the T axis indicates the magnitude of the frequency in the Y axis direction in the spatial image. The frequency component determination unit 13 calculates the absolute value of the DCT coefficient of the predetermined frequency component among the spatial frequency components, and determines the determination data F to the change unit 14 depending on whether the calculated value is equal to or greater than the predetermined value. Output. The determination data F is data for the changing unit 14 to determine the magnitude of the absolute value of the DCT coefficient of the predetermined frequency component.

  More specifically, the frequency component determination unit 13 determines the DCT coefficients, Qj (1, 0), Qj (0, 1), and Qj (1, 1) of the spatial frequency component corresponding to the section including the triangle mark. Absolute values q10, q01, and q11 are calculated. The DCT coefficient, Qj (1, 0), Qj (0, 1), and Qj (1, 1) are examples of predetermined frequency components. Then, it is determined whether at least one of the absolute values q10, q01, or q11 is a positive predetermined value α, for example, 64 or more. When it is determined that any one of the absolute values q10, q01, or q11 is greater than or equal to the predetermined value α, 1 is set to the determination data F, and the determination data F is output to the changing unit 14. When it is determined that all of the absolute values q10, q01, and q11 are less than the predetermined value α, 0 is set to the determination data F, and the determination data F is output to the changing unit 14.

  By comparing the absolute values q10, q01, and q11 with the predetermined value α, it is possible to determine whether or not the unit block to be processed includes an edge portion in the density region. The predetermined value α can be arbitrarily set by the designer. When the predetermined value α is smaller than 64, it becomes easier to determine that the absolute values q10, q01, and q11 are equal to or larger than the predetermined value α compared to when the predetermined value α is larger than 64, and an edge is added to the unit block to be image-processed. It becomes easy to judge that it contains.

  The changing unit 14 changes the DCT coefficient Qj (S, T) by calculation in accordance with the value of the determination data F output from the frequency component determination unit 13. The DCT coefficient Qj (S, T) is changed for each unit block.

  FIG. 3 is a conceptual diagram showing frequency components for changing the DCT coefficient. When determination data F having a value of 1 is output from the frequency component determination unit 13, the changing unit 14 is a real number greater than 1, for example, 1.3, with respect to the DCT coefficients Qj (S, T) of all AC components. And the DCT coefficient Qk (S, T) obtained by multiplying by 1.3 is output to the noise adding unit 15. In FIG. 3, the DCT coefficient Qj (S, T) of the AC component excluding the DC component indicated by the hatched block is changed. When the determination data F having the value 0 is output, the changing unit 14 sets the DCT coefficient Qk (S, T) = Qj (S, T) without changing the value of the DCT coefficient Qj (S, T). The data is output to the noise adding unit 15 as it is. The DCT coefficient Qk (S, T) processed by the changing unit 14 is expressed by the following equation (2).

Qk (S, T) = Qj (S, T) (S = T = 0)
= Qj (S, T) × 1.3 (S ≠ 0 or T ≠ 0) Expression (2)

  Next, the noise adding unit 15 adds the DCT coefficient of the image data having noise with blue noise characteristics to the DCT coefficient Qk (S, T) processed by the changing unit 14, and the DCT coefficient Ql after the addition (S, T) is output to the inverse frequency converter 16.

  FIG. 4 is a diagram illustrating an example of a blue noise mask. Blue noise is given as 256 × 256 matrix data, and the matrix data is called a blue noise mask. The noise adding unit 15 performs discrete cosine transform on the blue noise mask shown in FIG. 4 and holds the normalized DCT coefficient value. FIG. 5 is a diagram illustrating an example of DCT coefficients obtained by performing discrete cosine transform on a blue noise mask. In this embodiment, the blue noise mask is subjected to discrete cosine transform for each 8 × 8 unit block, and DCT coefficients of 32 × 32 blocks are held. Then, any DCT coefficient of the 32 × 32 block is added to the DCT coefficient Qk (S, T) of the unit block input to the noise adding unit 15. Similarly, any DCT coefficient of the 32 × 32 block is sequentially added to other unit blocks that are sequentially input.

  Blue noise is pattern data having spatial frequency components that are difficult to perceive by the human eye. The human eye cannot perceive a pattern image of a certain spatial frequency or higher, and the MTF (Modulation Transfer Function) of the visual system is known as a kind of low-pass filter (Takeshi Hamada, “ Image Quality Technology ", Journal of the Imaging Society of Japan, 2001, Vol. 40, No. 3, p.239-243). Blue noise can be obtained by manipulating a pseudo random pattern and generating a pattern in which the main component of the spatial frequency is distributed in a band equal to or higher than the cutoff frequency of the visual system MTF.

  By adding the DCT coefficient of blue noise to the DCT coefficient Qk (S, T), it is possible to prevent the pixels from being connected to each other in the highlight portion having a large gradation value. Moreover, the texture in the intermediate density portion can be improved.

  The inverse frequency conversion unit 16 performs inverse frequency conversion on the DCT coefficient Ql (S, T) output from the noise addition unit 15 to image data Pm (X, Y) having a spatial coordinate component, and the inverse frequency converted image data Pm. (X, Y) is output to the threshold processing unit 17. Specifically, inverse discrete cosine transformation of Formula (1) is performed.

  The threshold processing unit 17 converts the density region image data Pm (X, Y) output from the inverse frequency conversion unit 16 into multi-valued image data Po (X, Y) using a plurality of predetermined values. . For example, the image data Pm (X, Y) is converted into 4-valued image data Po (X, Y) by the following equation (3) using three predetermined values, 42, 127, and 212.

If 0 <Pm (X, Y) ≦ 42, Po (X, Y) = 0
If 42 <Pm (X, Y) ≦ 127, Po (X, Y) = 85
If 127 <Pm (X, Y) ≦ 212, Po (X, Y) = 171
If 212 <Pm (X, Y) ≦ 255, then Po (X, Y) = 255 (3)

  Next, the process procedure of the control part 10 is demonstrated using a flowchart. FIG. 6 is a flowchart illustrating a processing procedure related to image processing of the control unit 10. First, the control unit 10 sets the number of unit blocks when the image data Pi (X, Y) that is input and stored in the image data storage unit 11 is divided into 8 × 8 pixels as unit blocks, as a variable n. (Step S1). For example, in the case of image data with 256 × 256 pixels, 32 × 32 is set as the variable n. Then, the control unit 10 reads the image data Pi (X, Y) in a unit block of 8 × 8 pixels, and outputs the read image data Pi (X, Y) to the frequency conversion unit 12 (step S2).

  Next, the control unit 10 performs discrete cosine transform on Pi (X, Y) in the frequency transform unit 12 and outputs the DCT coefficient Qj (S, T) obtained by discrete cosine transform to the frequency component determination unit 13 and the change unit 14. (Step S3). Then, the control unit 10 calculates the absolute values q10, q01, and q11 of the DCT coefficients of the predetermined frequency component in the frequency component determination unit 13 (step S4), and any one of the calculated absolute values q10, q01, or q11 Is greater than or equal to a predetermined value α (step S5). If any of the absolute values q10, q01, or q11 is greater than or equal to the predetermined value α (step S5: YES), 1 is set to the determination data F, and the determination data F set to 1 is output to the changing unit 14 ( Step S6). When all of the absolute values q10, q01, and q11 are less than the predetermined value α (step S5: NO), the determination data F is set to 0, and the determination data F set to 0 is output to the changing unit 14 (step S5). S7).

  Next, the control unit 10 changes the value of the DCT coefficient Qj (S, T) to Qk (S, T) by the equation (2) in the changing unit 14 and outputs the value to the noise adding unit 15 (step S8). Then, the control unit 10 adds the DCT coefficient related to the blue noise to the DCT coefficient Qk (S, T), and outputs the DCT coefficient Ql (S, T) obtained by adding the noise to the inverse frequency conversion unit 16 by the noise addition unit 15. (Step S9).

  Next, the control unit 10 performs inverse discrete cosine transform on the DCT coefficient Ql (S, T) to Pm (X, Y) in the inverse frequency transform unit 16, and sets Pm (X, Y) obtained by inverse discrete cosine transform as a threshold value. The data is output to the processing unit 17 (step S10). Then, in the threshold processing unit 17, the control unit 10 converts Pm (X, Y) into image data Po (X, Y) that is quaternized by the equation (3), and outputs the image data to the outside (step S11). Next, the control unit 10 subtracts 1 from the variable n (step S12), and determines whether or not the subtracted variable n is 0 (step S13). That is, it is determined whether or not image processing has been completed for all unit blocks. When it is determined that the variable n is 0 (step S13: YES), the control unit 10 ends the image processing. If it is determined that the variable n is not 0 (step S13: NO), the control unit 10 returns the process to step S2, and performs the image processing of steps S2 to S11 for the remaining unit blocks.

  The operation and effect of the above-described processing will be described. First, the processes in step S4 and step S5 make it possible to determine whether or not the unit block has a characteristic part such as an edge part. Since the DCT coefficient of the low frequency component excluding the direct current component has more information about the image than the high frequency component, the absolute values q10, q01, and q11, that is, the DCT coefficient of the predetermined frequency component are set as the magnitudes. This is because it can be determined whether or not the unit block has image information by comparing with the predetermined value α. In general, the DCT coefficient of an image having an edge portion, not an image having a flat uniform density, is larger in a direct current component and a low frequency component than in a high frequency component, and image information is mainly concentrated in a low frequency component. This is well known (Hishiji Kishi, “Digital Image Processing Understandable”, CQ Publisher, p. 121-128). Since the DCT coefficient of the direct current component is proportional to the average gradation value in the density region of the unit block, it cannot be determined whether or not the image has an edge portion of the image.

  For example, when discrete cosine transform is performed on a solid image, that is, an image that does not include an edge portion, the DCT coefficient of the AC component is 0, and only the DC component is 0 or a value other than 0 corresponding to the average density of the image. In contrast, when an image having an edge portion is subjected to discrete cosine transform, the DCT coefficient of the low frequency component takes a value other than 0 corresponding to the feature of the edge portion.

  Next, in steps S6 to S8, image processing for enhancing the edge portion only for the image having the edge portion by changing the DCT coefficient depending on whether the image of the unit block has the edge portion or not. I do. The image enhancement process is performed by increasing the value of the DCT coefficient of the AC component. As described above, since the DCT coefficient of the AC component has information related to the edge portion of the image, the value of the DCT coefficient of the AC component is made larger than the value of the DCT coefficient of the original image, thereby obtaining the edge portion. Can be emphasized. On the other hand, when the edge portion is not included in the image portion, the image is not enhanced. Therefore, only the edge portion can be emphasized in the entire image, and the number of gradations can be reduced while keeping the characteristic portion of the original image good.

  In step S9, by adding the DCT coefficient of blue noise that is difficult for humans to perceive to the DCT coefficient Qk (S, T), the dispersibility of pixels in the highlight area of the image processed image, or the intermediate density area The texture can be improved. Also, since blue noise is difficult for humans to perceive, image degradation due to noise addition can be minimized.

  In the present invention, it is determined whether or not an edge part is included in each unit block. When the edge part is included, the spatial frequency component is changed so that the edge is emphasized. The number of gradations of the image can be reduced while maintaining the characteristic portion of the original image in good condition. In particular, it is effective for an image containing a lot of text or graphic art.

  Further, by adding the DCT coefficient of blue noise, it is possible to prevent pixels from being connected in a highlight portion having a large gradation value while minimizing image degradation.

  Note that the real value 1.3 multiplied by the DCT coefficient in step S8 is an example, and any value that can increase the value of the DCT coefficient within a range not exceeding the maximum value of the DCT coefficient may be used.

  In step S5, whether or not the unit block has an edge portion is determined based on the absolute values of the low frequency components, q10, q01, and q11. However, the present invention is not limited to this. The same determination may be performed using the absolute value of the DCT coefficient of the high spatial frequency component. For example, the reference area is increased to DCT coefficients Qj (0, 2), Qj (1, 2), Qj (2, 2), Qj (2, 1), Qj (2, 0), and these absolute values are set. The determination may be made by using. When the absolute value of the DCT coefficient obtained by enlarging the reference area is used, it can be more accurately determined whether or not the image includes an edge portion. Further, when determining the magnitude of each DCT coefficient, it may be determined using a predetermined value β different from the predetermined value α.

(Embodiment 2)
FIG. 7 is a block diagram illustrating an image processing apparatus according to the second embodiment. The image processing apparatus according to the second embodiment is similar to the image processing apparatus according to the first embodiment illustrated in FIG. 1. The control unit 10, the image data storage unit 11, the frequency conversion unit 12, the frequency component determination unit 13, and the change Unit 14, noise addition unit 15, inverse frequency conversion unit 16, and threshold processing unit 17. The image processing apparatus according to the second embodiment further includes a DC component determination unit 18. The configuration, operation, and effect of the image processing apparatus according to the second embodiment other than the DC component determination unit 18 are the same as the configuration, operation, and effect of the image processing apparatus according to the first embodiment. The same reference numerals are assigned and detailed description is omitted.

  The DC component determination unit 18 receives the DCT coefficient Qj (S, T) from the frequency conversion unit 12, and the DCT coefficient Qj (0, 0) of the DC component among the input DCT coefficients Qj (S, T). It is determined whether or not the value is within a predetermined range greater than the lower limit N1 and smaller than the upper limit N2, and the determined result is output to the noise adding unit 15. Note that the minimum value of the DCT coefficient Qj (0, 0) of the DC component is 0, and the maximum value is 8 × 255 = 2040.

  For example, if the DCT coefficient Qj (0, 0) of the DC component is a value greater than N1 = 0 and less than N2 = 2040, the DC component determination unit 18 sets 0 in the DC component determination data, and DCT DCT When the coefficient Qj (0, 0) is either N1 = 0 or N2 = 2040, 1 is set in the DC component determination data and output to the noise adding unit 15.

  When DC component determination data having a value of 1 is output from the DC component determination unit 18, the noise adding unit 15 does not perform noise addition, and when DC component determination data having a value of 0 is output, Embodiment 1 Add the noise shown in.

  Since the magnitude of the DCT coefficient Qj (0, 0) of the direct current component indicates the average gradation value in the density region of the unit block, the unit of the DCT coefficient Qj (0, 0) is determined by determining the magnitude of the DCT coefficient Qj (0, 0). It can be determined whether the block is entirely black or white. When the processing block is entirely black or white, the noise adding unit 15 does not perform noise addition processing. Therefore, a uniform black or white block can be obtained, and a good image can be obtained.

(Embodiment 3)
FIG. 8 is a block diagram illustrating an image processing apparatus according to the third embodiment. The image processing apparatus according to the third embodiment is similar to the image processing apparatus according to the second embodiment illustrated in FIG. 7. The control unit 10, the image data storage unit 11, the frequency conversion unit 12, the frequency component determination unit 13, and the change Unit 14, noise addition unit 15, inverse frequency conversion unit 16, threshold processing unit 17, and DC component determination unit 18. The image processing apparatus further includes a LUT (LUT: Look Up Table) unit 19 for storing a blue noise DCT coefficient used for noise addition in the noise addition unit 15. The configuration, operation, and effect of the image processing apparatus according to the third embodiment other than the LUT unit 19 are the same as the configuration, operation, and effect of the image processing apparatus according to the second embodiment. Reference numerals are assigned and detailed description is omitted.

  The LUT unit 19 performs discrete cosine conversion on the image data having the blue noise characteristic in the density region in advance for each unit block, and stores the DCT coefficient obtained by the discrete cosine conversion for each unit block. Further, a unit block whose DCT coefficient values are all 0 is stored.

  The DC component determination data output from the DC component determination unit 18 is output to the LUT unit 19. When DC component determination data having a value of 1 is output to the LUT unit 19, the noise adding unit 15 reads out a unit block having all DCT coefficients of 0 from the LUT unit 19 and outputs DC component determination data having a value of 0. If so, a unit block having a DCT coefficient of blue noise is read from the LUT unit 19 and noise addition is performed.

  More specifically, the blue noise mask shown in FIG. 4 is subjected to discrete cosine transform using 8 × 8 as a unit block, and normalized DCT coefficients as shown in FIG. 5 are calculated. The LUT unit 19 stores 8 blocks × 8 blocks of 8 × 8 unit blocks obtained by performing discrete cosine transform on blue noise, for a total of 64 blocks. The LUT unit 19 stores blocks whose DCT coefficients are all zero. The noise adding unit 15 sequentially reads out the unit blocks of the DCT coefficients related to the blue noise from the LUT unit 19, and outputs the read DCT coefficients for each unit block from the changing unit 14. Add to.

  FIG. 9 is a flowchart showing a processing procedure related to noise addition. In Embodiment 3, the control part 10 performs the process shown in FIG. 9 instead of the process of step S9 shown in FIG. That is, after finishing the process of step S8, before performing the process of step S10, the process of step S100 thru | or step S105 is performed. When the process of step S8 is completed, the control unit 10 determines whether the DC component determination unit 18 determines whether the DCT coefficient Qj (0, 0) of the DC component is N1 or less or N2 or more ( Step S100). When it is determined that the value of the DCT coefficient Qj (0, 0) is N1 or less or N2 or more (step S100: YES), the DC component determination data is set to 1 and the DC component determination is set to 1. Data is output to the LUT unit 19 (step S101). When the DC component determination data having a value of 1 is output to the LUT unit 19, the noise adding unit 15 reads out DCT coefficients having all values of 0 from the LUT unit 19 as a unit block (step S102). Then, the read DCT coefficient is added to Qk (S, T), and Ql (S, T) obtained by adding the DCT coefficient is output to the inverse frequency converter 16 (step S103). When the process of step S103 is completed, the process of step S10 shown in FIG. 6 is executed.

  When it is determined that the DCT coefficient Qj (0, 0) is a value larger than N1 and smaller than N2 (step S100: NO), the direct current component determination unit 18 sets 0 to the direct current component determination data. The DC component determination data is output to the LUT unit 19 (step S104). When DC component determination data having a value of 0 is output to the LUT unit 19, the noise adding unit 15 reads out DCT coefficients related to blue noise from the LUT unit 19 in a unit block (step S105). Then, the read DCT coefficient is added to Qk (S, T), and Ql (S, T) obtained by adding the DCT coefficient is output to the inverse frequency converter 16 (step S103). When the process of step S103 is completed, the process of step S10 shown in FIG. 6 is executed.

  According to the third embodiment, since the LUT unit stores the DCT coefficient of the image having the blue noise characteristic calculated in advance, the noise addition process can be performed at a higher speed than when the LUT unit is not provided. it can. Therefore, image processing can be performed at high speed.

(Embodiment 4)
FIG. 10 is a block diagram illustrating an image processing apparatus according to the fourth embodiment. The image processing apparatus according to the fourth embodiment is similar to the image processing apparatus according to the first embodiment shown in FIG. 1 in that the control unit 10, the image data storage unit 11, the frequency conversion unit 12, the frequency component determination unit 13, 2 change part 14a, noise addition part 15, reverse frequency conversion part 16, and threshold processing part 17 are provided. However, the second changing unit 14a in the image processing apparatus according to the fourth embodiment has a function different from that of the changing unit 14 of the image processing apparatus according to the first to third embodiments. Since the configuration, operation, and effect of the image processing apparatus according to the fourth embodiment other than the second changing unit 14a are the same as the configuration, operation, and effect of the image processing apparatus according to the first embodiment, The same reference numerals are assigned and detailed description is omitted.

  The function which the 2nd change part 14a has is demonstrated. When the frequency component determination unit 13 determines that any one of the absolute values q10, q01, or q11 is equal to or greater than the predetermined value α and outputs determination data F having a value of 1 to the second change unit 14a, the second change unit As described in Embodiment 1, 14a multiplies the DCT coefficient Qj (S, T) of the AC component by a real number larger than 1 in order to emphasize the edge portion. When the frequency component determination unit 13 determines that all of the absolute values q10, q01, and q11 are less than the predetermined value α and outputs determination data having a value of 0 to the second change unit 14a, the second change unit 14a performs a process of reducing the value of the DCT coefficient Qj (S, T) other than the DC component and the low-frequency DCT coefficient.

  FIG. 11 is a conceptual diagram showing the frequency region of the DCT coefficient that is changed by the second changing unit 14 a when the determination data F having a value of 0 is output from the frequency component determining unit 13. Of the DCT coefficients Qj (S, T), S and T DCT coefficients Qk (S, T) satisfying 3 <S + T ≦ 14 are to be changed. This change area is preset in the second change unit 14a or the control unit 10, for example.

  A specific changing procedure of the second changing unit 14a will be described. FIG. 12 is a conceptual diagram showing a two-dimensional matrix used when the second changing unit 14a performs the changing process. The two-dimensional matrix shown in FIG. 12 has 8 × 8 matrix data M (S, T) corresponding to 8 × 8 spatial frequency components. However, S = 0 to 7 and T = 0 to 7 (S and T are integers). When 0 ≦ S + T ≦ 3, the value of the matrix data M (S, T) is 1, and when 3 <S + T ≦ 14, the matrix data M (S, T) is a real number of 1 or more.

  The two-dimensional matrix shown in FIG. 12 generally has a characteristic reflecting a contrast sensitivity function (CSF), that is, a characteristic reflecting 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. The value of M (S, T) is a value that changes concentrically depending on the sensitivity to human contrast, for example, with the hatched frequency component as the center in FIG.

When the determination data F having a value of 0 is output from the frequency component determination unit 13, the second change unit 14 a sets the DCT coefficient Qj (S, T) to M ( Change by dividing by S, T). The calculation of equation (4) may be obtained up to the decimal point.
Qk (S, T) = Qj (S, T) / M (S, T) (4)

  Next, the second changing unit 14 a outputs the changed DCT coefficient Qk (S, T) to the noise adding unit 15. When the determination data F having a value of 1 is output from the frequency component determination unit 13, the DCT coefficient Qj (S, T) of the AC component is converted into the DCT coefficient Qk (S, T) as described in the first embodiment. ) = Qj (S, T) × 1.3, and the changed DCT coefficient Qk (S, T) is output to the noise adding unit 15. The processing after the noise adding unit 15 is sequentially processed in the same processing procedure as in the first embodiment.

  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, when a unit block does not include an edge portion, the image of the unit block is smoothed using matrix data of a two-dimensional matrix, and when the unit block includes an edge portion, the edge portion of the image Is emphasized. Therefore, in the entire image, the edge portion is emphasized and the flat portion not including the edge portion is smoothed, so that the graininess of the image can be effectively suppressed, and the characteristic portion of the original image is improved. The number of gradations can be reduced while maintaining.

  The matrix data of the two-dimensional matrix used for smoothing is not limited to that shown in FIG. 12, 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. The set two-dimensional matrix data, that is, the coefficients of the low frequency components are maintained, and the two-dimensional matrix in which the outer DCT coefficients are more strongly suppressed concentrically around a predetermined frequency component in the 8 × 8 block. Should be used. For example, a two-dimensional matrix having a Gaussian distribution may be used.

  In the embodiment, the frequency component of 3 <S + T ≦ 14 is changed. However, the present invention is not limited to this, and the frequency component of 1 <S + T ≦ 14 may be changed.

  Further, in the fourth embodiment, the image processing apparatus may be configured by including the DC component determination unit 18 and the LUT unit 19. In this case, the same effect as in the second and third embodiments can be obtained.

(Embodiment 5)
FIG. 13 is a block diagram illustrating an image processing apparatus according to the fifth embodiment. As in the image processing apparatus of the first embodiment shown in FIG. 1, the image processing apparatus is a control unit 10, an image data storage unit 11, a frequency conversion unit 12, a frequency component determination unit 23, a change unit 24, and a noise addition unit. 15, an inverse frequency conversion unit 16 and a threshold processing unit 17 are provided. However, the frequency component determination unit 23 and the change unit 24 in the image processing apparatus according to the fifth embodiment have different functions from the frequency component determination unit 13 and the change unit 14 of the image processing apparatus according to the first embodiment. .

  The function which the frequency component determination part 23 has is demonstrated. FIG. 14 shows the spatial frequency component of the DCT coefficient Qj (S, T), the magnitude of which is determined by the frequency component determination unit 23 among the spatial frequency components, using 8 × 8 sections. It is a conceptual diagram. The section located at the upper left corner and including the black circle mark indicates the DC component of the DCT coefficient Qj (S, T), and the other sections excluding the DC component indicate the AC component. The S axis indicates the magnitude of the frequency in the X axis direction in the spatial image, and the T axis indicates the magnitude of the frequency in the Y axis direction in the spatial image. That is, the S axis and the T axis indicate the magnitudes of the frequencies in the X axis direction and the Y axis direction in the concentration region.

  The frequency component determination unit 23 calculates the absolute value of the DCT coefficient of the spatial frequency component corresponding to the section including the triangle mark. That is, the absolute value q10 of the DCT coefficient Qj (1, 0), the absolute value q01 of the DCT coefficient Qj (0, 1), the absolute value q11 of the DCT coefficient Qj (1, 1), and the DCT coefficient Qj (2, 0). Absolute value q20, absolute value q02 of DCT coefficient Qj (0,2), absolute value q21 of DCT coefficient Qj (2,1), absolute value q12 of DCT coefficient Qj (1,2), and DCT coefficient Qj (2, The absolute value q22 of 2) is calculated respectively.

  Then, the frequency component determination unit 23 determines whether any of q10, q01, q11, q20, q02, q21, q12, or q22 is a positive predetermined value α or more. That is, whether q10 ≧ α or q01 ≧ α or q11 ≧ α or q20 ≧ α or q02 ≧ α or q21 ≧ α or q12 ≧ α or q22 ≧ α is satisfied. judge. It is assumed that the predetermined value α can be arbitrarily set, and the smaller (larger) the set value of the predetermined value α is, the easier (and less likely) to determine that the unit block image includes an edge. In the fifth embodiment, a predetermined value α = 64 is set as an example. The set predetermined value α is used as a determination material for determining whether or not an image in the density region of each unit block includes an edge portion.

  When it is determined that the conditional expression is satisfied, that is, when any of q10, q01, q11, q20, q02, q21, q12, or q22 is determined to be greater than or equal to the predetermined value α, the frequency component determination unit 23 determines 1 is set in the data F, and the determination data F in which 1 is set is output to the changing unit 24. When it is determined that the conditional expression is not satisfied, that is, when it is determined that any of q10, q01, q11, q20, q02, q21, q12, and q22 is less than the predetermined value α, the frequency component determination unit 23 The determination data F is set to 0, and the determination data F set to 0 is output to the changing unit 24. The determination data F is sequentially output for each unit block.

  The changing unit 24 changes the DCT coefficient Qj (S, T) by calculation according to the value of the determination data F output from the frequency component determination unit 23. The DCT coefficient Qj (S, T) is changed for each unit block.

  FIG. 15 is a conceptual diagram illustrating frequency components at which the changing unit 24 according to the fifth embodiment changes the DCT coefficient. In FIG. 15, the hatched section that is inclined downward to the right is the DCT coefficient Qj (S, T) that is changed by the changing unit 24 when the determination data F having a value of 1 is output from the frequency component determining unit 23. The change area | region in frequency space is shown. The DCT coefficients (S, T) corresponding to the hatched sections are AC component coefficients Qj (S, T) (0 <S ≦ 7, T = 0) having only the spatial frequency component in the horizontal direction, AC component coefficient Qj (S, T) having only a spatial frequency component (S = 0, 0 <T ≦ 7), and AC component coefficient Qj (1, 1) having fundamental frequency components in the horizontal and vertical directions, respectively. ). The frequency component region of the DCT coefficient Qj (S, T) that is changed by the changing unit 24 is set in advance in the changing unit 24 or the control unit 10, for example.

  FIG. 16 is a conceptual diagram showing frequency information data for changing DCT coefficients. As shown in FIG. 16, the frequency information data for the frequency (S, T) is data for weighting when changing the DCT coefficient Qj (S, T), and is set as a value of S + T + 2.

  For example, the frequency information data for the frequency (0, 1) is 0 + 1 + 2 = 3, the frequency information data for the frequency (0, 7) is 0 + 7 + 2 = 9, the frequency information data for the frequency (1, 0) is 1 + 0 + 2 = 3, and the frequency (7 , 0) is 7 + 0 + 2 = 9.

  When the determination data F = 1 is output from the frequency component determination unit 23, the change unit 24 has a DCT coefficient Qj (S, T) in the change region shown in FIG. 15 larger than 1 determined for each frequency. Multiply the real number to change the DCT coefficient Qj (S, T) to the DCT coefficient Qk (S, T). A real number larger than 1 is a number obtained by multiplying frequency information data related to the frequency (S, T) by a constant of 0.35. Therefore, the relationship between the DCT coefficient Qj (S, T) in the change area before being changed by the changing unit 24 and the DCT coefficient Qk (S, T) after the change is expressed by the following equation (5). .

Qk (S, T) = Qj (S, T) × (frequency information data) × 0.35
= Qj (S, T) × (S + T + 2) × 0.35 Expression (5)
(0 <S ≦ 7 and T = 0 or S = 0 and 0 <T ≦ 7 or S = T = 1)

  Further, the relationship between the DCT coefficient Qj (S, T) other than the changed region output to the changing unit 24 and the DCT coefficient Qk (S, T) output from the changing unit 24 to the noise adding unit 15 is expressed by the following equation ( 6).

Qk (S, T) = Qj (S, T) (6)
(S = T = 0, or S ≠ 0 and T ≠ 0, except S = T = 1)

  Note that the real number greater than 1 is a value that enhances the image by multiplication with the DCT coefficient Qj (S, T), so that the actual printing is performed in consideration of the overall balance such as whether or not the edge portion is excessively emphasized. It is better to decide while evaluating the image quality using a sample.

For example, when the DCT coefficient Qj (0, 2) is output, the changing unit 24
Qk (0,2) = Qj (0,2) × (0 + 2 + 2) × 0.35
Is output to the noise adding unit 15 and the DCT coefficient Qj (2, 3) is output,
Qk (2,3) = Qj (2,3)
Is output to the noise adding unit 15.

  Even when the process of changing the DCT coefficient Qj (S, T) to the DCT coefficient Qk (S, T) is performed according to the expressions (5) and (6), the DCT coefficient Qj (0, 0) of the DC component is changed. Since the value is not changed, the average density of the entire unit block image is maintained.

  When the determination data F = 0 is output from the frequency component determination unit 23, the changing unit 24 determines the value of the DCT coefficient Qj (S, T) other than the DC component and the low-frequency DCT coefficient as in the fourth embodiment. The process which decreases is performed. That is, the changing unit 24 changes the DCT coefficient Qj (S, T) to the DCT coefficient Qk (S, T) according to the equation (4), and outputs the changed DCT coefficient Qk (S, T) to the noise adding unit 15. To do.

  The DCT coefficient Qk (S, T) output from the changing unit 24 is processed in the same manner as in the first embodiment by the noise adding unit 15, the inverse frequency converting unit 16, and the threshold processing unit 17.

  FIGS. 17 and 18 are flowcharts illustrating a processing procedure related to image processing of the control unit 10 according to the fifth embodiment. Here, a case where an image with four gradations is obtained will be described. When the input image data Pi (X, Y) is stored in the image data storage unit 11, the control unit 10 firstly stores the image data Pi (X, Y) stored in the image data storage unit 11. The number of unit blocks when 8 × 8 pixels are divided into unit blocks is set as a variable n (step S201). Then, the control unit 10 reads the image data Pi (X, Y) in a unit block of 8 × 8 pixels, and outputs the read image data Pi (X, Y) to the frequency conversion unit 12 (step S202).

  Next, the control unit 10 performs discrete cosine transform of Pi (X, Y) in the frequency transform unit 12 and outputs the DCT coefficient Qj (S, T) obtained by discrete cosine transform to the frequency component determination unit 23 and the change unit 24. (Step S203). Then, the control unit 10 calculates the absolute values q10, q01, q11, q20, q02, q21, q12, and q22 of the DCT coefficient of the predetermined frequency component in the frequency component determination unit 23 (step S204), and the calculated absolute It is determined whether any of the values q10, q01, q11, q20, q02, q21, q12, or q22 is greater than or equal to a predetermined value α (step S205). When it is determined that any one of the absolute values q10, q01, q11, q20, q02, q21, q12, or q22 is greater than or equal to the predetermined value α (step S205: YES), 1 is set in the determination data F, and 1 is set. The set determination data F is output to the changing unit 24 (step S206). When it is determined that all of the absolute values q10, q01, q11, q20, q02, q21, q12, and q22 are less than the predetermined value α (step S205: NO), 0 is set to the determination data F and 0 is set. The determined determination data F is output to the changing unit 24 (step S207).

  Next, the control unit 10 determines whether or not the determination data F = 1 is output from the frequency component determination unit 23 to the change unit 24 (step S208). When it is determined that the determination data F = 1 is output to the changing unit 24 (step S208: YES), the control unit 10 uses the changing unit 24 to calculate the DCT coefficient Qj (S, T) using the equations (5) and (5). According to (6), the DCT coefficient is changed to Qk (S, T), and the changed DCT coefficient Qk (S, T) is output to the noise adding unit 15 (step S209). That is, in step S209, the DCT coefficient Qj (S, T) is changed so as to enhance the edge portion of the image.

  When it is determined that the determination data F = 1 is not output to the changing unit 24, that is, when the determination data F = 0 is output (step S208: NO), the control unit 10 uses the DCT coefficient in the changing unit 24. Qj (S, T) is changed to DCT coefficient Qk (S, T) by equation (4), and the changed DCT coefficient Qk (S, T) is output to noise adding unit 15 (step S210). That is, in step S210, the DCT coefficient Qj (S, T) is changed so as to smooth the image.

  When the process of step S209 or step S210 is finished, the control unit 10 adds the DCT coefficient related to blue noise to the DCT coefficient Qk (S, T) in the noise addition unit 15, and adds the noise to the DCT coefficient Ql (S , T) is output to the inverse frequency converter 16 (step S211). That is, the DCT coefficient of noise that has blue noise characteristics in the density region and is converted into image data having a spatial frequency component is sequentially added.

  Next, the control unit 10 performs inverse discrete cosine transform on the DCT coefficient Ql (S, T) to Pm (X, Y) in the inverse frequency transform unit 16, and sets Pm (X, Y) obtained by inverse discrete cosine transform as a threshold value. The data is output to the processing unit 17 (step S212). That is, the frequency domain data is converted into the density domain data. The inverse discrete cosine transform is performed by the inverse transform of Equation (1). Then, the control unit 10 performs simple quaternary threshold processing on Pm (X, Y) by the equation (3) in the threshold processing unit 17 and outputs the quaternized image data Po (X, Y) to the outside ( Step S213). Next, the control unit 10 subtracts 1 from the variable n (step S214), and determines whether or not the subtracted variable n is 0 (step S215). That is, it is determined whether or not image processing has been completed for all unit blocks. When it is determined that the variable n is 0 (step S215: YES), the control unit 10 ends the image processing. When it is determined that the variable n is not 0 (step S215: NO), the control unit 10 returns the process to step S202, and performs the image processing of steps S202 to S213 for the remaining unit blocks.

  Since configurations, operations, and effects other than the frequency component determination unit 23 and the change unit 24 included in the image processing apparatus according to the fifth embodiment are the same as those in the first embodiment, the same portions are denoted by the same reference numerals. Detailed description is omitted.

  In the image processing apparatus configured in this way, the DCT coefficient Qj (S, T) in the entire frequency range other than the DC component is not changed in the process of enhancing the edge portion of the image performed in step S209. Since the frequency region of the DCT coefficient Qj (S, T) to be changed is limited to the change region shown in FIG. 15, it is possible to suppress the occurrence of a block pattern at the edge portion related to the curve of the image. Therefore, compared with the case where the DCT coefficients Qj (S, T) of all the AC components are changed, an image after threshold processing having a cleaner curved portion can be obtained, and the characteristic portion of the original image can be maintained better. However, the number of gradations can be reduced.

  FIG. 19 is an explanatory diagram showing the difference in image quality depending on the frequency domain of the DCT coefficient changed by the changing unit. FIG. 19A is an example of an image obtained when image processing is performed by changing the DCT coefficient Qj (S, T) only in the change region shown in FIG. 15, and FIG. This is an example of an image obtained when image processing is performed on the same original image as FIG. 19A by changing the DCT coefficients Qj (S, T) of all frequency components other than the components, that is, all AC components.

  As can be seen from FIG. 19, the block pattern generated in the face outline portion, that is, the edge portion related to the curve, appearing in the image of FIG. 19A is suppressed well compared to the image of FIG. 19B. Yes.

  In the image processing apparatus according to the fifth embodiment, the higher the frequency of the DCT coefficient, the larger the real number is multiplied. Therefore, the edge is more effective than the case where all the DCT coefficients are multiplied by the same real number. An image can be obtained. Therefore, it is possible to reduce the number of gradations of the image while keeping the characteristic portion of the original image better.

  Further, in addition to the DCT coefficients of the AC component having only the horizontal spatial frequency component and the AC component having only the vertical spatial frequency component, the DCT coefficients having the horizontal and vertical fundamental frequency components are changed. Therefore, the edge portion can be emphasized more effectively.

  In the fifth embodiment, the DCT coefficients each having a fundamental frequency component are also changed. However, the DCT coefficients each having a fundamental frequency component may be changed.

  In the fifth embodiment, the process of smoothing the image is performed in step S210. However, the process of step S210 may not be performed. That is, in step S208, when it is determined that the determination data F = 0 is output to the changing unit, the noise addition and output processing in step 211 may be performed without changing the DCT coefficient. .

  Further, the DCT coefficient is changed by multiplying the DCT coefficient Qj (S, T) by the number obtained by multiplying the frequency information data related to the frequency (S, T) by a constant of 0.35. The number multiplied by the DCT coefficient Is not limited to this, and may be another real number larger than 1. For example, the constant 0.35 may be a real number such as 0.34, 0.36, or 0.4. Further, although the frequency information data is S + T + 2, it may be another real number having a larger (smaller) value as the frequency of the DCT coefficient is larger (smaller).

  Furthermore, the frequency region of the DCT coefficient determined by the frequency component determination unit may be further expanded in the low frequency region if the hardware condition permits.

  Further, in the fifth embodiment, the image processing apparatus may be configured by including the DC component determination unit 18 and the LUT unit 19. In this case, the same effect as in the second and third embodiments can be obtained.

(Embodiment 6)
Hereinafter, an image processing method and an image processing apparatus according to the sixth embodiment will be described in detail based on the drawings illustrating the embodiments. The image processing apparatus according to the sixth embodiment is similar to the image processing apparatus according to the first embodiment illustrated in FIG. 1 in that the control unit 10, the image data storage unit 11, the frequency conversion unit 12, the frequency component determination unit 13, and the change Unit 14, noise addition unit 15, inverse frequency conversion unit 16, and threshold processing unit 17.
The frequency component determination unit 13 according to the sixth embodiment executes a determination process different from that of the first embodiment. The configuration, operation, and effect of the image processing apparatus according to the sixth embodiment other than the frequency component determination unit 13 are the same as the configuration, operation, and effect of the image processing apparatus according to the first embodiment. The same reference numerals are assigned and detailed description is omitted.

  FIG. 20 is a conceptual diagram showing a block of image data after being converted into a spatial frequency component by the frequency converter 12. The S axis indicates the magnitude of the frequency in the X axis direction in the spatial image, and the T axis indicates the magnitude of the frequency in the Y axis direction in the spatial image. Of the 8 × 8 components, the component corresponding to the upper left position coordinate in the frequency space coordinates of FIG. 20 is the DC component, and the component corresponding to the position coordinate other than the position coordinate is the AC component.

  The DCT coefficient Qj (S, T) of the edge block determination region having the 3 × 3 matrix size shown in FIG. 20, that is, the DCT coefficient Qj (S, T) of the predetermined frequency component is the position coordinate of the DC component (s0, t0), the DCT coefficient Qj (S, T) of the spatial frequency component at the position coordinate is Qj (s0, t0), and the position coordinate on the right side of the position coordinate (s0, t0) in FIG. 20 is (s0 + 1, t0). ), The DCT coefficient Qj (S, T) of the spatial frequency component at the position coordinate is Qj (s0 + 1, t0), the position coordinate on the right side of the position coordinate (s0 + 1, t0) is (s0 + 2, t0), and the position The DCT coefficient Qj (S, T) of the spatial frequency component in the coordinates is Qj (s0 + 2, t0), and the position coordinates are (s0, t0 + 1), (s0 + 1, t0 + 1), (s0 + 2, t0 + 1) in order from the left side. Spatial frequency The DCT coefficients Qj (S, T) of the minute are respectively Qj (s0, t0 + 1), Qj (s0 + 1, t0 + 1), Qj (s0 + 2, t0 + 1), and further down one line, the position coordinates (s0, t0 + 2), ( When DCT coefficients Qj (S, T) of spatial frequency components corresponding to (s0 + 1, t0 + 2) and (s0 + 2, t0 + 2) are Qj (s0, t0 + 2), Qj (s0 + 1, t0 + 2), and Qj (s0 + 2, t0 + 2), respectively , Qj (s0 + μ, t0 + ν) (where μ, ν = 0, 1, 2, and μ = ν ≠ 0).

The frequency component determination unit 13 determines whether or not the DCT coefficient Qj (S, T) in the edge block determination region satisfies the following expression.
| Qj (s0 + μ, t0 + ν) × Qj (s0, t0) | ≧ α1 (7)
(However, μ, ν = 0, 1, 2, and μ = ν ≠ 0, α1 = 34000)

  The frequency component determination unit 13 multiplies the DCT coefficient Qj (S, T) of the edge block determination region by the DCT coefficient Qj (s0, t0) of the DC component among the spatial frequency components, and the absolute value Aqμν = | Qj (s0 + μ, t0 + ν) × Qj (s0, t0) | (where μ, ν = 0, 1, 2, μ = ν ≠ 0) are calculated, and μ satisfying the relationship of the above equation (7) When there is a set of ν, that is, any one of the products obtained by multiplying the DCT coefficient Qj (S, T) of the edge block determination region by the DCT coefficient Qj (s0, t0) of the DC component Is equal to or greater than a predetermined value α1, it is determined that the image processing target block is a block having an edge portion (hereinafter referred to as an edge block). When it is determined that the image processing target block is an edge block, 1 is set in the determination data F, and the determination data F is output to the changing unit 14.

  Further, when any magnitude of the product is less than the predetermined value α1, the frequency component determination unit 13 determines that the processing block is a block having no edge portion. If it is determined that the image processing target block is not an edge block, 0 is set in the determination data F, and the determination data F is output to the changing unit 14.

  The determination data F is data indicating whether or not the image processing target block is an edge block.

  The changing unit 14 changes the DCT coefficient Qj (S, T) by calculation in accordance with the value of the determination data F output from the frequency component determination unit 13. That is, when the determination data F = 1 is output, the changing unit 14 increases the DCT coefficient Qj (S, T) of the AC component to perform image enhancement processing, and the determination data F = 0 is output. In this case, the image smoothing process is performed by reducing the DCT coefficient Qj (S, T) of the AC component. The DCT coefficient Qj (S, T) is changed for each unit block.

  When determination data F having a value of 1 is output from the frequency component determination unit 13, the changing unit 14 is a real number greater than 1, for example, 1.3, with respect to the DCT coefficients Qj (S, T) of all AC components. And the DCT coefficient Qk (S, T) obtained by multiplying by 1.3 is output to the noise adding unit 15.

  When the determination data F having a value of 0 is output, the changing unit 14 divides the DCT coefficient Qj (S, T) of the predetermined AC component by one or more real numbers, for example, 350, thereby obtaining the DCT coefficient Qj (S , T) is reduced, and the reduced DCT coefficient Qk (S, T) is output to the noise adding unit 15.

  Next, the process procedure of the control part 10 is demonstrated using a flowchart. FIG. 21 is a flowchart illustrating a processing procedure related to image processing of the control unit 10. First, the control unit 10 sets the number of unit blocks when the image data Pi (X, Y) that is input and stored in the image data storage unit 11 is divided into 8 × 8 pixels as unit blocks, as a variable n. . For example, in the case of image data with 256 × 256 pixels, 32 × 32 is set as the variable n (step S301). Then, the control unit 10 reads the image data Pi (X, Y) in a unit block of 8 × 8 pixels, and outputs the read image data Pi (X, Y) to the frequency conversion unit 12 (step S302).

  Next, the control unit 10 performs discrete cosine transform on Pi (X, Y) in the frequency transform unit 12 and outputs the DCT coefficient Qj (S, T) obtained by discrete cosine transform to the frequency component determination unit 13 and the change unit 14. (Step S303).

  Then, the control unit 10 multiplies each DCT coefficient Qj (s0 + μ, t0 + ν) in the edge block determination region by the DC component value Qj (s0, t0) in the frequency component determination unit 13, and multiplies the product obtained by the multiplication. The absolute value Aqμν (where μ, ν = 0, 1, 2 and μ ≠ ν) is calculated, and it is determined whether any of the calculated absolute values Aqμν is greater than or equal to the predetermined value α1 (step S304).

If any of the absolute values Aqμν is greater than or equal to the predetermined value α1 (step S304: YES), 1 is set in the determination data F, and the determination data F in which 1 is set is output to the changing unit 14 (step S305). Then, the control unit 10 performs enhancement processing on the value of the DCT coefficient Qj (S, T) in the changing unit 14 from which the determination data F = 1 is output, and outputs the result to the noise adding unit 15 (step S306).
That is, the control unit 10 changes the DCT coefficient Qj (S, T) to Qk (S, T) by the equation (2) in the changing unit 14 and outputs the Qk (S, T) to the noise adding unit 15.

When all of the absolute values Aqμν are less than the predetermined value α1 (step S304: NO), the control unit 10 sets the determination data F to 0, and outputs the determination data F set to 0 to the changing unit 14 (step S304). S307). The control unit 10 performs a smoothing process on the value of the DCT coefficient Qj (S, T) in the change unit 14 to which the determination data F = 0 is output (step S308), and outputs the result to the noise addition unit 15.
For example, the DCT coefficient Qj (S, T) is changed to Qk (S, T) by Expression (4) and output to the noise adding unit 15.

  When the process of step S306 or step S308 is completed, the control unit 10 adds the DCT coefficient related to the blue noise to the DCT coefficient Qk (S, T) in the noise addition unit 15, and adds the noise-added DCT coefficient Ql (S, T) is output to the inverse frequency converter 16 (step S309).

  Next, the control unit 10 performs inverse discrete cosine transform on the DCT coefficient Ql (S, T) to Pm (X, Y) in the inverse frequency transform unit 16, and sets Pm (X, Y) obtained by inverse discrete cosine transform as a threshold value. The data is output to the processing unit 17 (step S310). Then, in the threshold processing unit 17, the control unit 10 converts Pm (X, Y) into image data Po (X, Y) that is quaternized by the equation (3), and outputs it to the outside (step S311). Next, the control unit 10 subtracts 1 from the variable n (step S312), and determines whether or not the subtracted variable n is 0 (step S313). That is, it is determined whether or not image processing has been completed for all unit blocks. When it is determined that the variable n is 0 (step S313: YES), the control unit 10 ends the image processing. If it is determined that the variable n is not 0 (step S313: NO), the control unit 10 returns the process to step S302, and performs the image processing of steps S302 to S311 for the remaining unit blocks.

The operation of the image processing apparatus according to the sixth embodiment will be described. First, the process of step S304 makes it possible to determine whether or not the unit block image has an edge portion.
Since the DCT coefficient Qj (S, T) of the low frequency component excluding the direct current component has more information on the edge portion than the high frequency component, the absolute value Aqμν, that is, the DCT coefficient Qj of the edge block determination region By comparing the product obtained by multiplying (S, T) by the DCT coefficient Qj (s0, t0) of the DC component with a predetermined value α1, it is determined whether or not the unit block has an edge portion. Can be determined.

  Next, in addition to the DCT coefficient Qj (S, T) of the edge block determination area having information related to the edge portion, the DCT coefficient Qj (s0, t0) of the DC component is also added to the determination element of the image having the edge portion. Will be described.

  When the block has an edge portion, the DCT coefficient Qj (S, T) in the edge block determination region has a larger value than that of a flat image. Therefore, it is basically determined whether or not the processing block is an edge block by determining whether or not the magnitude of the DCT coefficient Qj (S, T) in the edge block determination region is equal to or greater than a predetermined value. be able to.

However, the recognition characteristics for the human edge portion depend on the average gradation value of the image. That is, the DCT coefficient Qj (s0, t0) of the DC component corresponding to the average gradation value is large (small) even if the images have the same DCT coefficient Qj (S, T) in the edge block determination area. As a result, it becomes easier (or difficult) for a person to recognize the edge portion. Accordingly, the DCT coefficient Qj (S, T) in the edge block determination area is equal to or less than a predetermined value, but the DCT coefficient Qj (s0, t0) of the DC component is large, so that an image recognized as an edge portion is displayed. On the other hand, there arises a problem that the emphasis processing is not performed.
For example, for an image containing a thin character on the highlight, emphasis processing should be performed before quaternization, and the character should be emphasized, but it is a thin character, and the background highlight portion and the thin character portion Since the difference between the gradation values is small, the image processing target block is not determined as an edge block. Therefore, when the emphasis process is not performed and the gradation value is decreased, it may be difficult to see the character.
Further, when the predetermined value is simply reduced, it becomes easy to determine an edge block, but even an image that should not be emphasized is subjected to enhancement processing, and there is a problem that the image becomes rough.

  Therefore, a DC component DCT coefficient Qj (s0, t0) having information on the average gradation value of the image is added to the determination element of the edge block so that even a thin character on the highlight can be determined as the edge block.

  More specifically, is the absolute value of the product obtained by multiplying the DCT coefficient Qj (s, t0) of the DC component multiplied by the DCT coefficient Qj (s0, t0) of the DC component of the predetermined spatial frequency component equal to or greater than the predetermined value α1? By determining whether or not the edge block is determined, even if the DCT coefficient Qj (S, T) in the edge block determination region is the same block, the DCT coefficient Qj (s0, t0) is larger (smaller). This makes it easier (harder) to determine that an image processing target block is an edge block.

  Therefore, it is possible to determine the presence or absence of an edge portion depending on the average gradation value of the image, and it is possible to perform enhancement processing on the block of the image having the edge portion. For example, thin characters on the highlight can be emphasized.

In step S306 and step S308, the DCT coefficient is changed depending on whether the image of the unit block has an edge portion, thereby performing an enhancement process on the image having the edge portion, and having an edge portion. Smoothing processing is performed on images that do not exist Image enhancement processing is performed by increasing the value of the DCT coefficient Qj (S, T) of the AC component, and image smoothing processing is performed by decreasing the DCT coefficient Qj (S, T) of the AC component. .
As described above, since the DCT coefficient Qj (S, T) of the AC component has information relating to the edge portion of the image, the value of the DCT coefficient Qj (S, T) of the AC component is set to the value of the original image. By making the value larger than the value of the DCT coefficient Qj (S, T), the edge portion can be emphasized, and the value of the DCT coefficient Qj (S, T) is changed to the value of the DCT coefficient Qj (S, T) of the original image. By making it smaller, the image can be smoothed. Therefore, it is possible to enhance the edge portion in the entire image, smooth the flat image, and reduce the gradation value while maintaining the characteristic portion of the original image in good condition.

  In the present invention, it is determined whether or not each unit block includes an edge portion. When the unit block has an edge portion, the DCT coefficient Qj (S, T) is changed so that the edge portion is emphasized. When there is no portion, since the DCT coefficient Qj (S, T) is changed so as to smooth the image, an image with a clear edge portion can be obtained, and the characteristic portion of the original image can be improved. The gradation value of the image can be reduced while maintaining it. In particular, it is effective for an image containing a lot of text or graphic art.

  Further, in the determination of the frequency component, by using the DC component of the block to be image-processed, a block having a thin character on the highlight can be determined as an edge block, and text reproducibility is improved.

  Although the matrix size of the edge block determination area is 3 × 3 pixels in this embodiment, it may be expanded to a wider range in the low frequency area if the hardware condition permits.

  The predetermined value α1 is not limited to 34000 as long as it is an appropriate positive value that provides a preferable image.

(Embodiment 7)
The image processing apparatus according to the seventh embodiment is similar to the image processing apparatus according to the first embodiment illustrated in FIG. 1, the control unit 10, the image data storage unit 11, the frequency conversion unit 12, the frequency component determination unit 13, and the change. Unit 14, noise addition unit 15, inverse frequency conversion unit 16, and threshold processing unit 17.

Similarly to the sixth embodiment, the frequency component determination unit 13 according to the seventh embodiment multiplies the DCT coefficient Qj (s0 + μ, t0 + ν) of the edge block determination region by the DCT coefficient Qj (s0, t0) of the DC component. The absolute value Aqμν of the obtained product is calculated.
Then, it is determined whether or not the calculated absolute value Aqμν is equal to or greater than a predetermined value α1. When it is determined that the absolute value Aqμν is less than the predetermined value α1, it is further determined whether or not the DCT coefficient Qj (S, T) in the edge block determination region satisfies the following expression.
| Qj (s0 + μ, t0 + ν) | ≧ α2 (8)
(However, μ, ν = 0, 1, 2, and μ = ν ≠ 0, α2 = 16)

  When there is no set of μ and ν that satisfy the relationship of the above equation (8), that is, when any of the DCT coefficients Qj (S, T) in the edge block determination region is less than the predetermined value α2, the image processing target The block is determined to be a flat image block (hereinafter referred to as a non-edge block), and when any one of the absolute values is equal to or greater than a predetermined value α2, the image processing target block has an edge portion. And a block of an image that is not a flat image (hereinafter referred to as an intermediate block).

  When the frequency component determination unit 13 determines that any of the absolute values Aqμν is greater than or equal to α1, the frequency component determination unit 13 outputs determination data F = 01 to the changing unit 14, and both of the absolute values Aqμν are less than α1 and the absolute value qμν. Is greater than or equal to α2, the determination data F = 10 is output to the changing unit 14, and if any of the absolute values Aqμν is less than α1 and any of the absolute values qμν is less than α2, the determination data F = 00 is output to the changing unit 14.

FIG. 22 is a flowchart illustrating a processing procedure related to image processing of the control unit 10 according to the seventh embodiment. First, the control unit 10 executes the processing of steps S401 to S404 similar to the steps S301 to S304 shown in FIG.
That is, the control unit 10 sets the number of unit blocks to the variable n (step S401), reads out the image data Pi (X, Y) as a unit block of 8 × 8 pixels, and outputs it to the frequency conversion unit 12 ( Next, Pi (X, Y) is subjected to discrete cosine transform in the frequency converting unit 12 and output to the frequency component determining unit 13 and the changing unit 14 (step S403). Then, the control unit 10 uses the frequency component determination unit 13 to multiply the DCT coefficients Qj (s0 + μ, t0 + ν) in the edge block determination region by the DC component values Qj (s0, t0), and the absolute value Aqμν of the products obtained. (Note that μ, ν = 0, 1, 2 and μ ≠ ν) are calculated, and it is determined whether any of the calculated absolute values Aqμν is greater than or equal to a predetermined value α1 (step S404).

When it is determined that any one of the absolute values Aqμν is equal to or greater than the predetermined value α1 (step S404: YES), the control unit 10 sets 01 to the determination data F, and the determination data F set to 01 to the changing unit 14. Output (step S405).
Then, the control unit 10 performs enhancement processing on the value of the DCT coefficient Qj (S, T) in the changing unit 14 and outputs the result to the noise adding unit 15 (step S406).
That is, the control unit 10 changes the value of the DCT coefficient Qj (S, T) to Qk (S, T) by the equation (2) in the changing unit 14 and outputs the changed value to the noise adding unit 15.

  When any of the absolute values Aqμν is less than the predetermined value α1 (step S404: NO), the control unit 10 determines whether any of the absolute values qμν is greater than or equal to the predetermined value α2 (step S407). When it is determined that any one of the absolute values qμν is greater than or equal to the predetermined value α2 (step S407: YES), the control unit 10 sets 10 to the determination data F and transfers the determination data F set to 10 to the change unit 14. Output (step S408).

  When the determination data F = 10 is input to the changing unit 14, the changing unit 14 outputs the value to the noise adding unit 15 without changing the value of the DCT coefficient Qj (S, T).

  When it is determined that all of the absolute values qμν are less than the predetermined value α2 (step S407: NO), the control unit 10 sets the determination data F to 00, and sets the determination data F set to 00 to the change unit 14. Output (step S409).

  Then, the control unit 10 performs a smoothing process on the value of the DCT coefficient Qj (S, T) in the changing unit 14 (step S410) and outputs the result to the noise adding unit 15.

When the process of step S406, step S408, or step S410 is completed, the control unit 10 executes the processes of step S411 to step S415 similar to steps S309 to 313 illustrated in FIG.
That is, the control unit 10 adds the DCT coefficient related to blue noise to the DCT coefficient Qk (S, T) (step S411), and the inverse frequency conversion unit 16 converts the DCT coefficient Ql (S, T) to Pm (X , Y) (step S412), and the threshold processing unit 17 converts Pm (X, Y) into image data Po (X, Y) that has been quaternized by equation (3). (Step S413). Next, the control unit 10 subtracts 1 from the variable n (step S414), and determines whether or not the subtracted variable n is 0 (step S415). That is, it is determined whether or not image processing has been completed for all unit blocks. When it is determined that the variable n is 0 (step S415: YES), the control unit 10 ends the image processing. If it is determined that the variable n is not 0 (step S415: NO), the control unit 10 returns the process to step S402, and performs the image processing of steps S402 to S413 for the remaining unit blocks.

In the image processing apparatus according to the seventh embodiment, it is determined by the process of step S404 whether the image to be processed is an edge block, and an image determined not to be an edge block is a non-edge block. It is determined whether or not.
Then, the changing unit 14 performs enhancement processing on the image of the edge block and performs smoothing processing on the non-edge block.

  Since the image of the intermediate block does not have an edge portion but is not a flat image, it is difficult to uniformly determine whether the enhancement processing or the smoothing processing is appropriate. If the image is enhanced or smoothed, the image quality may be adversely affected. Therefore, the changing unit 14 does not perform any processing of enhancement processing and smoothing processing on the intermediate block.

  Therefore, it is possible to perform processing that captures the features of the image, and to reduce the gradation value of the image while maintaining a good feature portion of the image.

  The configuration, operation, and effect of the image processing apparatus according to the seventh embodiment other than the frequency component determination unit 13 and the change unit 14 are the same as the configuration, operation, and effect of the image processing apparatus according to the first embodiment. The same reference numerals are assigned to the parts to be described, and detailed description is omitted.

  The predetermined value α2 is not limited to 16 as long as it is an appropriate positive value so that a preferable image can be obtained.

(Embodiment 8)
An image forming apparatus including the image processing apparatus according to the present invention will be described in detail. FIG. 23 is a block diagram illustrating a configuration example of an image forming apparatus according to the present invention. The image forming apparatus is, for example, a digital color copying machine, and reduces the number of gradations of the color image input apparatus 2 to which image data having RGB color components are input and the image data input to the color image input apparatus 2. A color image processing apparatus 1 and a color image output apparatus 3 that outputs image data image-processed by the color image processing apparatus 1 are provided. The operation panel 4 includes a setting button or a numeric keypad for setting an operation mode of the image forming apparatus. The image forming apparatus also includes a CPU (Central Processing Unit) (not shown) that controls each of the devices included in the image forming apparatus.

  The color image processing apparatus 1 includes an A / D (analog / digital) conversion unit 101, a shading correction unit 102, an input tone correction unit 103, a region separation processing unit 104, a color correction unit 105, a black generation and under color removal unit 106, A spatial filter processing unit 107, an output tone correction unit 108, and a tone reproduction processing unit 109 are provided. The gradation reproduction processing unit 109 includes the image processing apparatuses according to the first to seventh embodiments described above.

  The color image processing apparatus 1 converts RGB analog signals output from the color image input apparatus 2 into image data of RGB digital signals, performs various image processing such as correction processing, and performs CMYK (C: cyan, M: magenta). (Y: Yellow, K: Black) Image data composed of digital signals is generated, and the number of gradations of the color component CMYK included in the generated image data is reduced to binary or quaternary. The binarized or quaternarized image data is temporarily stored in a storage unit (not shown) and output to the color image output device 3 at a predetermined timing.

  The color image input device 2 is a scanner having, for example, a CCD (Charge Coupled Device), and reads a reflected light image from an original as an RGB (R: red, G: green, B: blue) analog signal by the CCD. The read RGB analog signal is output to the color image processing apparatus 1.

  The A / D conversion unit 101 converts the RGB analog signal output from the color image input device 2 into RGB digital signal image data, and outputs the converted image data to the shading correction unit 102. The shading correction unit 102 performs processing for removing various distortions generated in the illumination system, the imaging system, and the imaging system of the color image input device 2 on the image data output from the A / D conversion unit 101, Output to the input tone correction unit 103. The input tone correction unit 103 adjusts the color balance of the image data output from the shading correction unit 102 and converts it into a density signal that can be easily processed by the image processing system employed in the color image processing apparatus 1. The converted image data is output to the region separation processing unit 104.

  The region separation processing unit 104 separates each pixel of the image related to the image data output from the input tone correction unit 103 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 106, the spatial filter processing unit 107, and the gradation reproduction processing unit 109. Also, the image data output from the input tone correction unit 103 is output to the color correction unit 105 as it is.

  The color correction unit 105 converts the RGB digital signal of the image data received from the input tone correction unit 103 into an image of a CMY (C: cyan, M: magenta, Y: yellow) digital signal in order to perform color reproduction faithfully. The data is converted into data and a process of removing color turbidity based on the spectral characteristics of the CMY color material including unnecessary absorption components is performed, and then output to the black generation and under color removal unit 106. The black generation and under color removal unit 106 generates a black signal, that is, a K signal, from three color signals of CMY digital signals included in the image data output from the color correction unit 105, that is, the C signal, the M signal, and the Y signal. Black generation is performed, and a new CMY digital signal is generated by subtracting the K signal obtained by black generation from the original CMY digital signal. Output to.

  As a general black generation process, there is a method of generating black using skeleton black. In the method of generating black using skeleton black, the input / output characteristics of the skeleton curve are y = f (x), the CMY digital signals of the input image data are C, M, Y, and the CMYK digital signals of the output image data Where C ′, M ′, Y ′, K ′ and the UCR (Under Color Removal) rate is α (0 <α <1), C ′, M ′, Y ′, K ′ It is represented by

K ′ = f {min (C, M, Y)}
C ′ = C−αK ′
M ′ = M−αK ′
Y ′ = Y−αK ′ (9)

  The spatial filter processing unit 107 performs a spatial filtering process using a digital filter on the image of the image data output from the black generation and under color removal unit 106 based on the region identification signal output from the region separation processing unit 104. The frequency characteristics are corrected to perform processing to improve image blurring or graininess degradation, and the processed image data is output to the output tone correction unit 108. The output tone correction unit 108 performs output tone correction processing on the image data output from the spatial filter processing unit 107 and outputs the image data to the tone reproduction processing unit 109. The gradation reproduction processing unit 109 binarizes the image data of the CMYK digital signal based on the image data output from the output gradation correction unit 108 and the region identification signal output from the region separation processing unit 104. Multilevel processing is performed.

  For example, for a region separated as a character by the region separation processing unit 104, the spatial filter processing unit 107 performs high-frequency enhancement by sharp enhancement processing included in the spatial filter processing in order to improve the reproducibility of black characters or color characters. Increase the amount of emphasis. Further, the gradation reproduction processing unit 109 performs high-resolution binarization or multilevel conversion suitable for high-frequency reproduction.

  Further, with respect to the region separated as halftone dots by the region separation processing unit 104, the spatial filter processing unit 107 performs low-pass filter processing for removing the input halftone component. The output tone correction unit 108 performs output tone correction processing for converting a signal such as a density signal into a halftone dot area ratio that is a characteristic value of the color image output device 3. Then, gradation reproduction processing for binarization or multi-value processing is performed so that the image is finally separated into pixels and each gradation can be reproduced.

  Further, with respect to the region separated as a photograph by the region separation processing unit 104, the gradation reproduction processing unit 109 performs binarization or multi-value processing that emphasizes gradation reproducibility.

  The image data of the CMYK digital signal that has been binarized or multi-valued by the gradation reproduction processing unit 109 is output to the color image output device 3. The color image output device 3 is a device that forms an image on a recording medium such as paper based on CMYK digital signals of image data received from the color image processing device 1, for example, an electrophotographic or ink jet printer. .

  The operation panel 4 is an input means for an operator to input an instruction by a key operation or the like. An operator instruction is output as a control signal from the operation panel 4 to the color image input device 2, the color image processing device 1, and the color image output device 3. An original image is read by the color image input device 2 according to an operator's instruction, and after the data processing by the color image processing device 1, an image is formed on a recording medium by the color image output device 3 and functions as a digital color copying machine. The above processing is executed by the CPU.

  According to the present invention, since the gradation reproduction processing unit 109 performs the processing detailed in the first to seventh embodiments, a binary image or a four-value image with high original image reproducibility is generated. Is possible. That is, it is possible to generate an image with a sharper edge than the original image, and to form an image with a reduced number of gradations while maintaining a good characteristic portion of the original image. Further, by adding the DCT coefficient of blue noise, it is possible to form an image in which pixels are not connected in the highlight portion. Further, the pixels are not connected at the highlight portion, and an image holding a uniform black or white portion can be formed.

(Embodiment 9)
A computer program according to the present invention, a computer-readable recording medium 7 storing the computer program, and a computer 5 functioning as an image processing apparatus by executing the computer program will be described. FIG. 24 is a functional block diagram of an image forming system including a computer 5 and an image output device 6 that function as an image processing apparatus according to the present invention. FIG. 25 is a block diagram showing a configuration of a computer 5 that functions as an image processing apparatus according to the present invention. The image forming system includes a computer 5 and an image output device 6. The image output device 6 is a printer, for example, and performs image formation of an electrophotographic method or an inkjet method.

  The computer 5 performs a color correction function unit 50a that performs color correction processing of the input image data, and threshold processing that reduces the number of gradations of the color-corrected image data, for example, 256 gradations to binary or quaternary values. A gradation reproduction processing function unit 50b and a printer language translation function unit 50c that converts image data whose number of gradations has been reduced by threshold processing into a printer language are included. The color correction function unit 50a has the same functions as the color correction unit 105, the black generation and under color removal processing unit 106, the spatial filter processing unit 107, the output gradation correction unit 108, and the like shown in FIG. The processing function unit 50b has the same function as that of the image processing apparatuses of the first to seventh embodiments described above. The image data converted into the printer language by the printer language translation function unit 50c is output by the communication port driver 50d to the external image output device 6 via the RS232C standard terminal, the communication port 55 such as a LAN card or a LAN board. .

  A specific configuration of the computer 5 will be described. The computer 5 includes a CPU 50 connected to the bus 51. Connected to the bus 51 are a ROM 53 storing a control program necessary for the CPU 50 to control various hardware described later, and a RAM 52 such as a DRAM for temporary storage. The storage means 54 such as a hard disk drive connected to the bus 51 stores a computer program according to the present invention, and the CPU 50 activates the computer program and expands a predetermined program portion in the RAM 52. By executing the processing related to the computer program, the computer 5 functions as an image processing apparatus having the same function as the image processing apparatus according to the first to seventh embodiments. The input unit 57 is configured by a keyboard, a mouse, and the like, and the display unit 58 is configured by a CRT display or a liquid crystal display that displays a processing result of the computer 5 and an image input / output to / from the computer 5.

  The external storage unit 56 is a flexible disk drive, a CD-ROM drive, or the like that reads the computer program according to the present invention from the recording medium 7 such as a flexible disk, CD-ROM, or DVD that records the computer program according to the present invention. Composed. The computer program read from the external storage unit 56 is stored in the storage unit 54. The communication port 55 is an interface for outputting an image processed by the computer 5 to the image output device 6.

  In the present invention, the computer program recorded on the recording medium 7 such as a CD-ROM is read by the external storage unit 56, stored in the storage means 54 or the RAM 52, and executed by the CPU 50, thereby implementing the computer 5. It becomes possible to function as the image processing apparatus according to the first to seventh embodiments and the image forming apparatus according to the eighth embodiment.

  That is, an image processing apparatus that can obtain an image with sharper edges than the original image, and can reduce the number of gradations of the image while maintaining a good characteristic portion of the original image. In addition, by adding the DCT coefficient of blue noise, it is possible to realize an image processing apparatus that can prevent pixels from being connected in a highlight portion having a large gradation value while minimizing image degradation. Furthermore, it is possible to achieve an image processing apparatus that can obtain uniform black or white blocks and prevent a pixel from being connected to each other in a highlight portion having a large gradation value, while obtaining a uniform black or white block.

  Furthermore, when the image processing apparatus corresponding to the fifth embodiment is functioned, it is possible to effectively prevent a block pattern from being generated in the curved portion of the image, compared to a case where all the AC component coefficients are changed. It is possible to reduce the number of gradations of the image while keeping the characteristic portion of the original image better. Furthermore, when the image processing apparatus corresponding to the fifth embodiment is functioned, the edge portion is more effective without losing the characteristic portion of the image than when all the AC components are multiplied by the same real number larger than 1. The number of gradations of the image can be reduced while keeping the characteristic part of the original image good.

  In this embodiment, the computer program is read from the storage medium storing the computer program according to the present invention and stored in the storage means. However, the computer or workstation connected to the network via the communication port, etc. The computer program according to the present invention may be received from the apparatus and stored in the hard disk drive or RAM.

  The recording medium may be a storage medium that can be read directly or indirectly by a computer, for example, a semiconductor element such as a ROM or a flash memory, or a magnetic storage medium such as a flexible disk, a hard disk, an MD, or a magnetic tape. But it ’s okay. Further, an optical storage medium such as a CD-ROM, MO, or DVD may be used. The recording method and reading method of the recording medium 7 are not limited.

  Further, the image output apparatus may be a digital multifunction machine having a copy function and a facsimile function in addition to the printer function.

1 is a block diagram showing an image processing apparatus according to Embodiment 1 of the present invention. It is the conceptual diagram which showed the DCT coefficient converted into the spatial frequency component in the frequency conversion part using the 8x8 division. It is a conceptual diagram which shows the frequency component which changes a DCT coefficient. It is a figure which shows an example of a blue noise mask. It is a figure which shows an example of the DCT coefficient which carried out the discrete cosine transform of the blue noise mask. It is a flowchart which shows the process sequence which concerns on the image process of a control part. 6 is a block diagram illustrating an image processing apparatus according to Embodiment 2. FIG. 6 is a block diagram showing an image processing apparatus according to Embodiment 3. FIG. It is a flowchart which shows the process sequence which concerns on noise addition. FIG. 10 is a block diagram illustrating an image processing apparatus according to a fourth embodiment. It is a conceptual diagram which shows the frequency area | region of the DCT coefficient which a 2nd change part changes, when the determination data whose value is 0 are output from the frequency component determination part. It is a conceptual diagram which shows the two-dimensional matrix data used when a 2nd change part performs the process of a change. FIG. 10 is a block diagram illustrating an image processing apparatus according to a fifth embodiment. It is the conceptual diagram which showed the spatial frequency component of DCT coefficient Qj (S, T) from which the magnitude | size of an absolute value is determined in a frequency component determination part among 8 spatial components. It is a conceptual diagram which shows the frequency component which the change part which concerns on Embodiment 5 changes a DCT coefficient. It is a conceptual diagram which shows the frequency information data for changing a DCT coefficient. 10 is a flowchart illustrating a processing procedure related to image processing of a control unit according to a fifth embodiment. 10 is a flowchart illustrating a processing procedure related to image processing of a control unit according to a fifth embodiment. It is explanatory drawing which shows the difference in the image quality by the frequency domain of the DCT coefficient changed in a change part. It is a conceptual diagram which shows the block of the image data after converted into the spatial frequency component in the frequency conversion part. 14 is a flowchart illustrating a processing procedure related to image processing of a control unit according to a sixth embodiment. 18 is a flowchart illustrating a processing procedure related to image processing of a control unit according to a seventh embodiment. 1 is a block diagram illustrating a configuration example of an image forming apparatus according to the present invention. 1 is a functional block diagram of an image forming system including a computer functioning as an image processing apparatus and an image output apparatus according to the present invention. It is a block diagram which shows the structure of the computer which functions as an image processing apparatus which concerns on this invention. It is a figure which shows an example of the 4x4 dither matrix used for the dither method which binarizes a 256 gradation image. It is a figure which shows an example of the weighting coefficient matrix used for an error diffusion method.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Color image processing apparatus 2 Color image input apparatus 3 Color image output apparatus 4 Operation panel 5 Computer 6 Image output apparatus 7 Recording medium 10 Control part 11 Image data memory | storage part 12 Frequency conversion part 13 Frequency component determination part 14a 2nd change part 14 Change unit 15 Noise addition unit 16 Inverse frequency conversion unit 17 Threshold processing unit 18 DC component determination unit 19 LUT unit 50 CPU
51 bus 52 RAM
53 ROM
54 storage means 55 communication port 56 external storage unit 57 input unit 58 display unit 101 A / D conversion unit 102 shading correction unit 103 input tone correction unit 104 region separation processing unit 105 color correction unit 106 black generation lower color removal unit 107 space Filter processing unit 108 Output tone correction unit 109 Tone reproduction processing unit

Claims (17)

In an image processing method for converting image data into image data having a spatial frequency component and changing the coefficient of the spatial frequency component to perform image processing,
Determining whether the absolute value of the coefficient of the predetermined frequency component of the converted image data is greater than or equal to the first predetermined value;
When it is determined that the absolute value is greater than or equal to the first predetermined value, the AC component coefficient of the image data is multiplied by a real number greater than 1, or the AC component coefficient of the image data is a positive value less than 1. Change the coefficient by dividing by the real number ,
Add the coefficient value of the spatial frequency component of the blue noise to the coefficient of the AC component of the image data changed by the calculation,
The image data added with the coefficient value is inversely converted into image data having a spatial coordinate component,
An image processing method comprising: comparing a gradation value of inversely converted image data with a second predetermined value and converting the gradation value to a gradation value according to the comparison result. In an image processing method for converting image data into image data having a spatial frequency component and changing the coefficient of the spatial frequency component to perform image processing,
Determining whether the absolute value of the product obtained by multiplying the coefficient of the DC component by the coefficient of the predetermined frequency component of the converted image data is equal to or greater than the first predetermined value;
When it is determined that the absolute value is greater than or equal to the first predetermined value, the AC component coefficient of the image data is multiplied by a real number greater than 1, or the AC component coefficient of the image data is a positive value less than 1. Change the coefficient by dividing by the real number ,
Add the coefficient value of the spatial frequency component of the blue noise to the coefficient of the AC component of the image data changed by the calculation,
The image data added with the coefficient value is inversely converted into image data having a spatial coordinate component,
An image processing method comprising: comparing a gradation value of inversely converted image data with a second predetermined value and converting the gradation value to a gradation value according to the comparison result. When it is determined that the absolute value is greater than or equal to the first predetermined value, a real number greater than 1 is set to the coefficient of each of the AC component having only the horizontal spatial frequency component and the AC component having only the vertical spatial frequency component. The image processing method according to claim 1, wherein the coefficient is changed by calculation by multiplication. When it is determined that the absolute value is equal to or greater than the first predetermined value, the coefficient is changed by calculation by multiplying the coefficient by a larger (smaller) real number as the frequency of the AC component is higher (lower). the image processing method according to any one of claims 1 to 3, characterized in that. In an image processing apparatus comprising conversion means for converting image data into image data having a spatial frequency component, and performing image processing by changing a coefficient of the spatial frequency component,
Determination means for determining whether an absolute value of a coefficient of a predetermined frequency component of the image data converted by the conversion means is equal to or greater than a first predetermined value;
When the determination means determines that the absolute value is greater than or equal to the first predetermined value, the AC component coefficient of the image data is multiplied by a real number greater than 1, or the AC component coefficient of the image data is set to 1. Arithmetic means for changing the coefficient by dividing by a small positive real number ;
Adding means for adding the coefficient value of the spatial frequency component of the blue noise to the coefficient of the alternating current component of the image data changed by the computing means;
Inverse conversion means for inversely converting the image data to which the coefficient values are added by the addition means into image data having a spatial coordinate component;
An image processing apparatus comprising: means for comparing the gradation value of the image data inversely converted by the inverse conversion means with a second predetermined value and converting the gradation value into a gradation value according to the comparison result. In an image processing apparatus comprising conversion means for converting image data into image data having a spatial frequency component, and performing image processing by changing a coefficient of the spatial frequency component,
Determining means for determining whether an absolute value of a product obtained by multiplying a coefficient of a DC component by a coefficient of a predetermined frequency component of the image data converted by the converting means is equal to or greater than a first predetermined value;
When the determination means determines that the absolute value is greater than or equal to the first predetermined value, the AC component coefficient of the image data is multiplied by a real number greater than 1, or the AC component coefficient of the image data is set to 1. Arithmetic means for changing the coefficient by dividing by a small positive real number ;
Adding means for adding the coefficient value of the spatial frequency component of the blue noise to the coefficient of the alternating current component of the image data changed by the computing means;
Inverse conversion means for inversely converting the image data to which the coefficient values are added by the addition means into image data having a spatial coordinate component;
An image processing apparatus comprising: means for comparing the gradation value of the image data inversely converted by the inverse conversion means with a second predetermined value and converting the gradation value into a gradation value according to the comparison result. The converting means includes
The image data is converted into image data having a spatial frequency component in a predetermined frequency range,
The determination means includes
The determination is made as to whether or not an absolute value of a product obtained by multiplying the coefficient of the AC component on the low frequency side by the coefficient of the DC component is equal to or greater than the first predetermined value. 6. The image processing apparatus according to 6. The computing means is
When the determination means determines that the absolute value is equal to or greater than the first predetermined value, the coefficient of each of the alternating current component having only the horizontal spatial frequency component and the alternating current component having only the vertical spatial frequency component is The image processing apparatus according to any one of claims 5 to 7, wherein a real number greater than 1 is multiplied. The computing means is
The image processing apparatus according to any one of claims 5 to 8, wherein the coefficient is multiplied by a larger (smaller) real number as the frequency of the AC component is higher (lower). The computing means is
When the determination means determines that the absolute value is not equal to or greater than the first predetermined value, the AC component coefficient of the image data is multiplied by a positive real number smaller than 1, or the AC component coefficient of the image data is 1 the image processing apparatus according to any one of claims 5 to 9, characterized in that you have to be divided by a real number larger than. The determination means includes
Means for determining whether or not an absolute value of a coefficient of a predetermined frequency component of the image data converted by the conversion means is equal to or greater than the first predetermined value;
The adding means includes
When the determination unit determines that the absolute value of the product is not equal to or greater than the first predetermined value and determines that the absolute value of the coefficient is equal to or greater than the first predetermined value, the image converted by the conversion unit The coefficient value of the spatial frequency component of blue noise is added to the coefficient of the AC component of the data,
The computing means is
When the determination means determines that the absolute value of the product is not equal to or greater than the first predetermined value and the absolute value of the coefficient is not equal to or greater than the first predetermined value, the image data converted by the conversion means The coefficient of the alternating current component of the image data is multiplied by a positive real number smaller than 1, or the coefficient of the alternating current component of the image data is divided by a real number larger than 1. 8. Image processing apparatus. An image processing apparatus according to any one of claims 5 to 1 1, an image in which the image processing apparatus based on the image data obtained by image processing, characterized in that is adapted to form an image Forming equipment. In a computer program for causing a computer to convert image data into image data having a spatial frequency component and changing the coefficient of the spatial frequency component to perform image processing,
Causing the computer to determine whether the absolute value of the coefficient of the predetermined frequency component of the converted image data is greater than or equal to a first predetermined value;
If the computer determines that the absolute value is greater than or equal to the first predetermined value, the coefficient of the AC component of the image data is multiplied by a real number greater than 1, or the coefficient of the AC component of the image data is greater than 1. Changing the coefficient by dividing by a small positive real number ;
Causing the computer to add the coefficient value of the spatial frequency component of the blue noise to the coefficient of the AC component of the image data changed by the operation;
Causing the computer to inversely convert the image data to which the coefficient value has been added into image data having a spatial coordinate component;
A computer program comprising: causing the computer to compare the gradation value of the inversely converted image data with a second predetermined value and converting the gradation value to a gradation value corresponding to the comparison result. In a computer program for causing a computer to convert image data into image data having a spatial frequency component and changing the coefficient of the spatial frequency component to perform image processing,
Causing the computer to determine whether or not an absolute value of a product obtained by multiplying a coefficient of a predetermined frequency component of the converted image data by a coefficient of a DC component is equal to or greater than a first predetermined value;
If the computer determines that the absolute value is greater than or equal to the first predetermined value, the coefficient of the AC component of the image data is multiplied by a real number greater than 1, or the coefficient of the AC component of the image data is greater than 1. Changing the coefficient by dividing by a small positive real number ;
Causing the computer to add the coefficient value of the spatial frequency component of the blue noise to the coefficient of the AC component of the image data changed by the operation;
Causing the computer to inversely convert the image data to which the coefficient value has been added into image data having a spatial coordinate component;
A computer program comprising: causing the computer to compare the gradation value of the inversely converted image data with a second predetermined value and converting the gradation value to a gradation value corresponding to the comparison result. The step of causing the computer to change the coefficient by multiplying the coefficient of the alternating current component of the image data by a real number larger than 1 or dividing the coefficient of the alternating current component of the image data by a positive real number smaller than 1 comprises:
If it is determined that the absolute value is greater than or equal to the first predetermined value, the computer has an AC component having only a horizontal spatial frequency component and an AC component having only a vertical spatial frequency component, respectively. It claims 1 to 3 or a computer program according to claim 1 4, characterized in that it comprises the step of multiplying the real number larger than 1. The step of causing the computer to change the coefficient by multiplying the coefficient of the alternating current component of the image data by a real number larger than 1 or dividing the coefficient of the alternating current component of the image data by a positive real number smaller than 1 comprises:
The computer program according to claim 15 , comprising: causing the computer to multiply the coefficient by a larger (smaller) real number as the frequency of the AC component is higher (lower). A computer-readable recording medium characterized in that are recorded thereon a computer program as claimed in any one of claims 1 3 to claims 1 to 6.

Images