JP2010166448A - Image processing method, image processing apparatus, image forming apparatus with the same, image processing program and recording medium - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image processing apparatus for providing an excellent output image by improving image quality deterioration that has occurred in compressed image data. <P>SOLUTION: The image processing apparatus 3 includes; a region separating section 13 for separating image data into regions such as a character region and a dot region; an image quality level determining section 11 for determining an image quality level of compressed image data; a JPEG expanding section 12 for applying decoding processing to the compressed image data; and a spatial filtering section 16 for applying filtering processing to the image data decoded by the JPEG expanding section 12 by setting a coefficient of filtering processing on the basis of a region separation result obtained in the region separating section 13 and the image quality level determined in the image quality level determining section 11. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to an image processing method for decoding compressed image data with high image quality, an image processing apparatus, an image forming apparatus including the image processing apparatus, an image processing program, and a recording medium.

In recent years, multifunction peripherals (abbreviated as MFP) are equipped with a function for filing image data to a hard disk, an image transmission function such as scan-to-email, and the like. In such a device, when image data is stored and transmitted, compression processing is performed on the image data. As a method, mainly JPEG (Joint
Photographic Experts Group) compression method is widely used.

  In the standard JPEG compression method, several parameters such as a quantization table and a sampling ratio can be set. These parameters are responsible for the trade-off between image quality and compression rate of compressed images. For example, depending on the purpose and preference, such as emphasizing image quality, emphasizing the compression rate, or setting emphasizing maintaining both on average. For copiers and scanners that can be adjusted and can select image modes such as character mode and photo mode, select compression parameters (quantization tables) that can handle a variety of documents in consideration of image quality and compression rate. It is configured to be able to.

  When compression processing is performed on image data in this way, for example, noise typified by block noise or ringing noise may occur in decoded image data. Block noise is noise that makes an image appear to be divided into blocks, and ringing noise is also referred to as mosquito noise, and is foggy noise that appears around the contour of an image. Therefore, in order to reduce noise from the decoded image data, a process for switching the smoothing process and a process for switching the halftone process are performed on the decoded image data.

  In the conventional technology, it is determined whether block noise or ringing noise is generated in the decoded image data, and switching between filter processing and halftone processing (error diffusion processing) is performed based on the determination result (for example, Patent Document 1). Specifically, the decoded image data is divided into a plurality of blocks, and an average value of variance values for each block is calculated for the target block and peripheral blocks around the target block (for example, eight blocks), and block noise is detected. If it is determined that block noise has occurred, the noise level is determined by comparing the variance of all blocks with a threshold value, and the degree of smoothing is changed according to the noise level. To perform filtering. If it is determined that no block noise has occurred, ringing is performed using (1) the variance value of the peripheral block and the variance value of the target block, and (2) the variance value of all blocks and the variance value of all blocks. It is determined whether or not noise is generated. If ringing noise is generated, smoothing processing having characteristics different from smoothing for block noise is performed.

JP 2004-343334 A

  However, in the conventional technique, if it is determined that there is ringing noise, the image is uniformly smoothed, so that the image quality of characters may be deteriorated.

  Further, in the conventional technique, halftone dots and block noise cannot be distinguished by dispersion, and the halftone dots are uniformly smoothed. Therefore, for example, there is a problem that halftone originals such as image data obtained by scanning a printed matter cannot be handled. is there.

  Further, in the conventional technique, (a) noise cannot be correctly determined by local region dispersion, and is unnecessarily smoothed. (B) noise and image characteristics cannot be distinguished only by dispersion. (C) For example, Due to the problem that the filter coefficient cannot be switched appropriately according to the image area such as characters, halftone dots, photographic paper photographs, and solid and background, unintentional image quality degradation due to insufficient accuracy of noise determination. May cause.

  Accordingly, an object of the present invention is to improve an image quality degradation that has occurred in compressed image data and obtain a good output image, an image processing apparatus, an image forming apparatus including the image processing apparatus, an image processing program, and a recording To provide a medium.

The present invention comprises an image quality level judging means for judging an image quality level based on a value of a quantization table of compressed image data,
Decoding means for decoding the compressed image data;
Coefficient setting means for setting a filter processing coefficient based on the image quality level determined by the image quality level determination means;
An image processing apparatus comprising: filter means for performing filter processing on the image data decoded by the decoding means using the coefficient set by the coefficient setting means.

In addition, the present invention further comprises area separation means for separating the image data decoded by the decoding means into a plurality of areas including at least one of a character area and a halftone dot area,
The coefficient setting means sets a filter processing coefficient based on the region separation result obtained by the region separation means and the image quality level determined by the image quality level determination means.

In the present invention, the image quality level determination means includes:
An information analysis unit that extracts the value of the quantization table from the compressed image data;
It has a level determination unit that determines the image quality level by comparing the quantization table value extracted by the information analysis unit with a plurality of predetermined threshold values.

In addition, the present invention is an image forming apparatus including the image processing apparatus.
The present invention also includes an image quality level determining step for determining an image quality level based on a quantization table value of compressed image data;
A decoding step of decoding the compressed image data;
A coefficient setting step for setting a filter processing coefficient based on the image quality level determined in the image quality level determination step;
And a filtering step of performing a filtering process on the image data decoded in the decoding step using the coefficient set in the coefficient setting step.

The present invention further includes a region separation step of separating the image data decoded in the decoding step into a plurality of regions including at least one of a character region and a halftone dot region,
In the coefficient setting step, a filter processing coefficient is set based on the region separation result obtained in the region separation step and the image quality level determined in the image quality level determination step.

  Further, the present invention is an image processing program for realizing the image processing apparatus, and is an image processing program for causing a computer to function as each of the means.

  The present invention is also a computer-readable recording medium on which the image processing program is recorded.

  According to the present invention, when the compressed image data is decoded and output in the image processing apparatus, the image quality level determining means automatically sets the image quality level based on the value of the quantization table of the compressed image data. to decide. Then, the coefficient setting means sets a filter coefficient for performing filter processing based on the image quality level determined by the image quality level determination means. Then, the filter means performs a filtering process on the image data decoded by the decoding means using the filter coefficient set by the coefficient setting means. As described above, the image processing apparatus automatically switches the filter coefficient for performing the filter process based on the image quality level automatically determined by the image quality level determination unit, and therefore performs an appropriate filter process according to the image quality level. be able to.

  In the conventional image processing apparatus, after the compressed image data is decoded, the filter coefficient is switched by judging the noise level from the local characteristics of the pixel value. For this reason, in the conventional image processing apparatus, the accuracy of determining the noise is insufficient, and accordingly, unintentional filter processing may be performed and image quality may be deteriorated. In contrast, the image processing apparatus of the present invention determines the image quality level (that is, the noise level) based on the value of the quantization table of the compressed image data, and switches the filter coefficient. That is, the image processing apparatus of the present invention is also used when decoding the quantization table value, which is a parameter responsible for the image quality set when compressing the image data, and therefore can accurately determine the image quality level. It becomes possible. Since the image processing apparatus of the present invention switches the filter coefficient based on the image quality level determined as described above, an optimal filter process for appropriately removing block noise and ringing noise in the decoded image data is performed. Is possible. Therefore, the image processing apparatus of the present invention can significantly improve image quality degradation as compared with the image processing apparatus of the prior art. Therefore, the image processing apparatus of the present invention can always obtain a good output image regardless of whether the image data is highly compressed or low-compressed.

  According to the present invention, when the compressed image data is decoded and output by the image processing apparatus, the image quality level determining means automatically sets the image quality level based on the value of the quantization table of the compressed image data. Judgment. The coefficient setting means sets a filter coefficient for performing filter processing based on the region separation result obtained by the area separation means and the image quality level determined by the image quality level determination means. Then, the filter means performs a filtering process on the image data decoded by the decoding means using the filter coefficient set by the coefficient setting means. As described above, the image processing apparatus automatically switches the filter coefficient for performing the filter processing based on the type of the region separated by the region separating unit and the image quality level automatically determined by the image quality level determining unit. Therefore, it is possible to perform appropriate filter processing according to the level of image quality degradation occurring in the compressed image data and a plurality of areas including the character area and the halftone dot area, and image data consisting of various components (areas). In contrast, a high-quality output image can be obtained. Therefore, mosquito noise around characters generated in compressed image data and block noise in photographs such as halftone dots can be properly removed, and even low-compressed images can be compressed. Even with data, it is possible to always obtain a good output image.

  According to the invention, the information analysis unit extracts the value of the quantization table from the compressed image data in the image quality level determination means. Then, the level determination unit determines the image quality level by comparing the quantization table value extracted by the information analysis unit with a plurality of predetermined threshold values. In this way, in the image quality level determination means, the level determination unit determines the image quality level based on the value of the quantization table, so that it is possible to easily and accurately determine the degree of image quality degradation that has occurred in the compressed image data. Can do.

  According to the invention, an image forming apparatus includes the image processing apparatus. The coefficient setting means included in the image processing apparatus can perform optimum filter processing according to the image quality level determined by the image quality level determination means, so that even if the image data is highly compressed, the image data is low-compressed. However, it is possible to provide an image forming apparatus that can always obtain a good output image.

  According to the present invention, when the compressed image data is decoded and output, the image quality level determining step automatically determines the image quality level based on the value of the quantization table of the compressed image data. In the coefficient setting step, a filter coefficient for performing filter processing is set based on the image quality level determined in the image quality level determination step. In the filtering step, the image data decoded in the decoding step is filtered using the filter coefficient set in the coefficient setting step. As described above, according to the image processing method of the present invention, the filter coefficient for performing the filter processing is automatically switched based on the image quality level automatically determined in the image quality level determination step. Processing can improve image quality degradation that occurs in compressed image data, and always obtain a good output image regardless of whether it is highly compressed image data or low-compressed image data Is possible.

  According to the present invention, when the compressed image data is decoded and output, the image quality level determining step automatically determines the image quality level based on the value of the quantization table of the compressed image data. In the coefficient setting step, a filter coefficient for performing filter processing is set based on the region separation result obtained in the region separation step and the image quality level determined in the image quality level determination step. In the filtering step, the image data decoded in the decoding step is filtered using the filter coefficient set in the coefficient setting step. As described above, in the image processing method of the present invention, the filter coefficient for performing the filter processing is automatically set based on the type of the region separated in the region separation step and the image quality level automatically determined in the image quality level determination step. Therefore, it is possible to perform appropriate filtering according to the level of image quality degradation that has occurred in the compressed image data and multiple areas including the character area and the dot area, and from various components (areas) A high-quality output image can be obtained for the image data. Therefore, mosquito noise around characters generated in compressed image data and block noise in photographs such as halftone dots can be properly removed, and even low-compressed images can be compressed. Even with data, it is possible to always obtain a good output image.

  According to the invention, the image processing program can realize the image processing apparatus by a computer.

  According to the invention, the recording medium is a computer-readable medium on which the image processing program is recorded. By causing the general-purpose computer such as a personal computer to read the image processing program recorded on the recording medium, it is possible to perform optimum filtering according to the image quality level of the input compressed image data, and to obtain a good image. Can be output.

1 is a block diagram illustrating a configuration of an image forming apparatus 1 according to an embodiment of the present invention. 3 is a block diagram illustrating a configuration of an image quality level determination unit 11. FIG. It is a figure which shows the coordinate of the quantization table Q contained in the header information of a JPEG code. It is a figure which shows the mode of separation of the high density side and low density side of a character area. It is a figure which shows an example of a filter coefficient. 3 is a block diagram showing a configuration of a spatial filter unit 16. FIG. It is a graph which shows the frequency characteristic of each filter coefficient shown by FIG. It is a graph which shows the frequency characteristic of each filter coefficient shown by FIG. 3 is a flowchart showing an image processing operation in the image forming apparatus 1. It is a block diagram which shows the structure of the image forming apparatus 30 of other embodiment of this invention. It is a block diagram which shows the structure of the image forming apparatus 40 of the further another form of implementation of this invention. 4 is a flowchart showing an image processing operation in the image forming apparatus 40. 1 is a configuration diagram illustrating an example of a computer system 50. FIG.

  FIG. 1 is a block diagram showing a configuration of an image forming apparatus 1 according to an embodiment of the present invention. The image forming apparatus 1 is realized, for example, as a digital copying machine or a multi-function machine having a copier function, a printer function, a facsimile transmission function, a Scan to E-mail function, and the like. The image forming apparatus 1 includes an interface 2, an image processing device 3, and an image output device 4.

  The image processing apparatus 3 includes an image quality level determination unit 11, a JPEG decompression unit 12, a region separation unit 13, a color correction unit 14, a black generation and under color removal unit (hereinafter “black generation / under color removal unit”). 15), a spatial filter unit 16, and a halftone generation unit 17.

  Compressed image data (hereinafter referred to as “compressed image data”) input to the image processing device 3 via the interface 2 is delivered to the image quality level determination unit 11 and the JPEG decompression unit 12. The image quality level determination unit 11 may be provided before the JPEG decompression unit 12.

Here, the compressed image data input to the image processing device 3 via the interface 2 is an RGB (R: red, G: green, B: blue) digital color image signal (hereinafter referred to as “RGB digital signal”). Image data consisting of the standard JPEG (Joint
Photographic Experts Group) A JPEG code to which header information is added, which is generated by encoding processing using a compression algorithm. The header information added to the JPEG code is in a standard data format, and the image width / height, quantization table, Huffman coding table, sampling ratio, etc. are divided by the specified marker symbols. Are described in order.

  FIG. 2 is a block diagram illustrating a configuration of the image quality level determination unit 11. FIG. 3 is a diagram showing the coordinates of the quantization table (Q-table) included in the header information of the JPEG code. The image quality level determination unit 11 is an image quality level determination unit, and includes a header analysis unit 111 and a quantization table determination unit 112.

  The header analysis unit 111 is an information analysis unit, which first extracts header information from the JPEG code passed from the interface 2, and then extracts a quantization table value from a predetermined section of the header information. To the conversion table determination unit 112. Note that the header information and the quantization table write section in the standard JPEG standard are defined, so if the JPEG code conforms to the JPEG standard, the header analysis unit 111 determines the value of the quantization table. Can be easily extracted.

  The quantization table determination unit 112 is a level determination unit, and determines the image quality level of the compressed image based on the received quantization table value. Here, the quantization table is divided into two tables for luminance signal and color difference signal, each of which consists of 8 × 8 values. In the present embodiment, the quantization table determination unit 112 determines the image quality level using a luminance signal table that easily affects image quality, such as characters and halftone dots.

  Next, an example of the image quality level determination method in the image quality level determination unit 11 will be described. In the present embodiment, the header analysis unit 111 extracts the value of the coordinates (4, 4) in the quantization table included in the header information. Then, the quantization table determination unit 112 performs threshold processing to change the image quality level to high image quality (image quality level “4”), relatively high image quality (image quality level “3”), and relatively low image quality (image quality level “4”). 2 ”) and low image quality (image quality level“ 1 ”). That is, based on the determination formula shown in Table 1, the quantization table determination unit 112 determines that the image quality level is “4” (high) when the value of the coordinate (4, 4) of the quantization table is 10 or less. If the image quality level is greater than 10 and 30 or less, it is determined that the image quality level is “3” (relative image quality). If it is greater than 30 and 50 or less, It is determined that the image quality level is “2” (relatively low image quality), and if it is greater than 50, it is determined that the image quality level is “1” (low image quality).

  As described above, the image quality level determination unit 11 is configured to determine the image quality level using the value of the coordinate (4, 4) among the plurality of coordinates in the quantization table. The level of mosquito noise that tends to occur can be determined accurately and easily.

  The value of the quantization table represents the roughness of the quantization applied at the time of compression. The larger the quantization table value, the coarser the quantization, and the smaller the value, the finer the quantization. is there. Therefore, it can be determined that the larger the quantization table value is, the more the image quality is degraded (that is, the mosquito noise and block noise are strongly generated), and the smaller the quantization table value is, the less the image quality degradation (that is, the mosquito noise). It can be determined that this is an image in which noise and block noise are not so much generated. Note that the value of coordinates (0, 0) represents the DC component of the image, and the other coordinates represent the roughness of quantization for the AC component of the image.

  The information of the image quality level automatically determined by the image quality level determination unit 11 as described above is transferred to the spatial filter unit 16 and is one of the consideration conditions when setting the filter coefficient used in the filter processing. Become one.

  In the present embodiment, the image quality level determination unit 11 is configured to determine the image quality level using only the luminance signal quantization table. However, the image quality level determination unit 11 determines using the color difference signal table. It may be configured. The image quality level determination unit 11 is configured to determine the image quality level using only the value of the coordinates (4, 4) in the quantization table. However, the image quality level is determined by combining a plurality of other coordinate values. It may be configured to determine. For example, the image quality level determination unit 11 determines the image quality level using the values of the coordinates (5, 5) corresponding to a relatively high frequency band when the image quality of a small character is to be emphasized. If the image quality is to be emphasized on average, the image quality level is determined by combining the values of coordinates (3, 3) and coordinates (5, 5) corresponding to the relatively low to high frequency bands. Configured. Furthermore, in the present embodiment, the number of stages of the image quality level is shown in four stages, but the number of stages is not limited to this.

  The JPEG decompression unit 12 is a decoding unit that performs processing such as decoding the code information by decoding the header information of the JPEG code, and decompresses the image data into an RGB digital signal. Note that all the processing performed here is based on a standardly defined JPEG decompression algorithm. The RGB digital signal from the JPEG decompression unit 12 is given to the region separation unit 13 and the color correction unit 14.

  The region separation unit 13 is a region separation unit, and separates the image data of the RGB digital signal generated by the JPEG decompression unit 12 into a plurality of regions including at least one of a character region and a halftone region. Here, the process of separating the image data into a plurality of areas is specifically a process of determining what kind of area each pixel of the image data belongs to. An example of the region separation method by the region separation unit 13 will be described later. The region separation unit 13 converts each pixel of the image data into a character region, a halftone dot region, and other regions (for example, a background region, a photographic paper photograph region, etc.). And a signal representing the separation result (hereinafter referred to as “region separation signal”) is delivered to the black generation / under color removal unit 15, the spatial filter unit 16, and the halftone generation unit 17, respectively. As a result, the black generation / undercolor removal unit 15, the spatial filter unit 16, and the halftone generation unit 17 perform appropriate process switching according to various regions.

  The color correction unit 14 is a digital color image signal of CMY (C: cyan, M: magenta, Y: yellow) that is a complementary color of the RGB digital signal with respect to the image data of the RGB digital signal generated by the JPEG decompression unit 12. (Hereinafter referred to as “CMY digital signal”) and processing for improving color reproducibility is performed. The color correction unit 14 performs a process of removing color turbidity based on the spectral characteristics of CMY color materials including unnecessary absorption components in order to faithfully reproduce colors. As a processing method, a method of having a correspondence relationship between an input RGB digital signal and an output CMY digital signal as an LUT (Look Up Table), a color masking method using a conversion matrix as shown in the following formula (1), etc. There is.

For example, when the color masking method is used, an L ^* a ^* b ^* value (CIE 1976 L ^* a ^* b ^* signal (CIE: Commission ^{) of} a color output when a certain CMY image data is given to the image output device 4. International de l'Eclairage; the International Lighting Commission, and RGB image data in which color patches having the same L ^* a ^* b ^* as L ^* : brightness, a ^* , b ^* : chromaticity)) are read, A large number of sets of CMY image data given to the image output apparatus 4 are prepared, and coefficients of a transformation matrix from a ₁₁ to a _{33 in} Expression (1) are calculated from the combinations, and color correction is performed using the calculated coefficients. Process. In order to increase the accuracy, it is only necessary to add a second or higher order term. The CMY digital signal from the color correction unit 14 is given to the black generation / undercolor removal unit 15.

  The black generation / under color removal unit 15 generates a black (K) signal from the CMY three-color CMY digital signals after color correction, and subtracts a portion where the black signals generated from the original CMY digital signals overlap, A process for generating a simple CMY digital signal is performed. Thus, the black generation / under color removal unit 15 converts the CMY three-color CMY digital signals into CMYK four-color CMYK digital signals.

The black generation / undercolor removal unit 15 performs black generation using skeleton black as an example of black generation processing. In the black generation by the skeleton black, the input / output characteristics of the skeleton curve are y = f (x), the densities corresponding to the input C, M, and Y are respectively C, M, and Y, and the output C, When the density corresponding to each of M, Y, and K is C ′, M ′, Y ′, and K ′, and the UCR (Under Color Removal) rate is α (0 <α <1), black generation / under color In the removal process, the density signals of three colors of CMY are converted into density signals of four colors of CMYK by the following formulas (2) to (5).
K ′ = f {min (C, M, Y)} (2)
C ′ = C−αK ′ (3)
M ′ = M−αK ′ (4)
Y ′ = Y−αK ′ (5)

  The black generation / undercolor removal unit 15 sets the UCR rate of the character area to 1. As a result, when color information representing black is added to the character area, a portion corresponding to the character area in the image formed on the recording medium is formed only by the K recording agent.

  The spatial filter unit 16 functions as a coefficient setting unit, and based on the image quality level information delivered from the image quality level judgment unit 11 and the region separation signal delivered from the region separation unit 13, an optimum filter coefficient for each pixel. Select and set. The spatial filter unit 16 functions as a filter unit, and performs filter processing such as enhancement processing and smoothing processing on the image data of the CMYK digital signal using the set filter coefficients. The spatial filter unit 16 can prevent blurring of the output image and deterioration of graininess by correcting the spatial frequency characteristics of the image data. The selection of the filter coefficient in the spatial filter unit 16 will be described later. The CMYK digital signal from the spatial filter unit 16 is given to the halftone generation unit 17.

  The halftone generation unit 17 performs output gradation correction processing for converting a signal such as a density signal into a halftone dot area ratio that is a characteristic value of the image output device 4. The halftone generation unit 17 performs pseudo gradation reproduction processing on the image data of the CMYK digital signal based on the region identification signal so that the image can finally reproduce the gradation in a pseudo manner. In this embodiment, a systematic dither method, an error diffusion method, or the like is used as a method used for the pseudo gradation reproduction process.

  The image data that has been subjected to each processing described above is output to the image output device 4. The image output device 4 can output a color image on a sheet material such as paper, and is a device that reproduces an image such as an electrophotographic printer or an inkjet printer. The image output device 4 can form an image on a recording medium using recording agents of two or more colors, and in the present embodiment, paper of each color of C, M, Y, K is used to make paper An image can be formed on a sheet body. The image output device 4 forms an image using the C recording agent for the image data corresponding to the C digital signal, and uses the M recording agent for the image data corresponding to the M digital signal. For image data corresponding to a Y digital signal, an image is formed using a Y recording agent, and for image data corresponding to a K digital signal, an image is formed using a K recording agent. To do.

  Next, an example of a region separation method by the region separation unit 13 will be described. As a method of performing the region separation process in the region separation unit 13, for example, a method described in Japanese Patent Laid-Open No. 2002-232708 can be used. Hereinafter, this method will be briefly described. This method is the sum of the maximum density difference which is the difference between the minimum density value and the maximum density value in an n × m block (for example, 7 × 7 pixels) including the target pixel and the absolute value of the density difference between adjacent pixels. A certain sum density busyness is calculated and compared with a plurality of predetermined threshold values to separate a character (character edge) region / halftone dot region and other regions (background region / photographic paper photograph region). In performing the region separation, the following feature amount is used.

  (1) Since the density distribution of the base region is usually small in density change, both the maximum density difference and the total density busyness become very small. The density distribution of the photographic paper photograph area (for example, a continuous tone area such as a photographic paper photograph is expressed as a photographic paper photograph area here) has a smooth density change, and the maximum density difference and the total sum. The density complexity is both small and slightly larger than the ground area. That is, in the background area and the photographic paper photograph area (other areas), the maximum density difference and the total density busyness are small values.

  (2) In the density distribution of the halftone dot area, the maximum density difference varies depending on the halftone dot. However, since the total density busyness varies by the number of halftone dots, the ratio of the total density busyness to the maximum density difference Becomes larger. Therefore, when the total density busyness is larger than the product of the maximum density difference and the character / halftone determination threshold, it can be determined that the pixel is a halftone pixel.

  (3) The density distribution of the character area has a large maximum density difference, and accordingly, the total density busyness increases. However, since the density change is smaller than that of the halftone dot area, the total density busyness becomes smaller than that of the halftone dot area. . Therefore, when the total density busyness is smaller than the product of the maximum density difference and the character / halftone dot determination threshold, it is possible to determine that the pixel is a character area pixel.

Next, a processing method will be described.
(A) The calculated maximum density difference and the maximum density difference threshold, and the calculated total density busyness and the total density busyness threshold are compared. When it is determined that the maximum density difference is smaller than the maximum density difference threshold and the total density busyness is smaller than the total density busyness threshold, the target pixel is the other region (background / photographic paper photo region). If not, it is determined that the area is a character / halftone area.

  (B) When it is determined that the region is the character / halftone dot region, the calculated total density busyness is compared with a value obtained by multiplying the maximum density difference by a character / halftone determination threshold, and the total density busyness is calculated. If it is smaller, it is determined to be a character region, and if the total density busyness is larger, it is determined to be a halftone dot region.

  (C) For a pixel determined to be a character area, an average value of pixel values in the block is obtained. If the pixel value of the target pixel is equal to or less than the average value, it is determined that the pixel is on the high density side. If it is larger, it is determined that it is on the low concentration side. (In this embodiment, since RGB digital signals are used, the smaller the pixel value, the darker the pixel value, and the larger the pixel value, the brighter the pixel.)

  The above processing is performed using any one of the RGB digital signals (for example, G signal), and the determination result obtained for each pixel is output as a region separation signal. Note that processing may be performed using two or three of the RGB digital signals. In that case, a region separation signal corresponding to each signal may be output, or may be output as one region separation signal by giving a priority order for each pixel. . In this case, the priority order may be given on the basis of which region the judgment result is given priority, or may be given on the basis of which signal the judgment result is given priority.

  Through the above processing, the image data can be separated into a character area, a halftone dot area, and other areas. Further, the character area is divided into a high density side character area which is a high density side and a low density side which is a low density side. It can be separated into character areas.

  FIG. 4 is a diagram illustrating the separation of the high density side and the low density side of the character area. FIG. 4A is a schematic diagram showing the image 21 at the end of the character, and FIG. 4B is a schematic diagram showing the result of the region separation processing performed by the region separation unit 13. In FIG. 4B, each pixel is shown in a grid pattern. Pixels that are near the edge (the boundary between the background 23 and the character 22) 24 of the character 22 and that are inside the character 22 are determined to be on the high density side, and pixels that are outside the character 22 (background 23) are determined to be on the low density side. Is done. Mosquito noise tends to stand out outside the character 22 and tends to occur at a position slightly away from the edge 24. According to the region separation method, the character region can be detected with a certain pixel width with respect to the edge 24, and further the low density side can be detected up to several pixels away from the edge 24. is there. When mosquito noise is generated around the character, the noise can be removed by performing smoothing processing on the low-density character region 25, and at that time, the high-density character region 26 is emphasized. By applying the processing, the characters can be reproduced clearly.

  Next, filter coefficients set in the spatial filter unit 16 based on the image quality level determined by the image quality level determination unit 11 and the region separation signal generated by the region separation unit 13 will be described. Table 2 shows an example of setting filter coefficients. FIG. 5 is a diagram illustrating an example of the filter coefficient. In FIG. 5, the filter name, the filter coefficient, and the division coefficient are shown in association with each other.

  Table 2 shows filter names of filter coefficients corresponding to each level of the image quality level and each region represented by the region separation signal. The image quality level can be selected from four levels 1 to 4. The image quality level “1” is low image quality, the image quality level “2” is relatively low image quality, the image quality level “3” is relatively high image quality, and the image quality level “ “4” corresponds to high image quality. As filter coefficients, there are provided enhancement filters 1, 2, 3 and smoothing filters 1, 2, 3, 4, 5 as shown in FIG. In Table 2, “other area” of the area separation signal is an area that is not included in either the character area or the halftone dot area. As for other regions, three examples of Examples 1, 2, and 3 are shown. In addition, when “through” is described, it indicates that the filtering process is not performed.

  FIG. 6 is a block diagram illustrating a configuration of the spatial filter unit 16. The spatial filter unit 16 includes a filter coefficient selection unit 161 and a calculation unit 162. The filter coefficient selection unit 161 includes a memory such as a ROM, and stores a plurality of sets of coefficients as a table shown in Table 2 in advance. The filter coefficients stored in the filter coefficient selection unit 161 are adjusted in advance so that a preferable output image can be obtained using a large number of input images with various conditions and sample images with image quality levels.

  The filter coefficient selection unit 161 selects a filter coefficient with reference to the table shown in Table 2 based on the image quality level and the region separation signal. The calculation unit 162 performs a convolution operation on the image data using the filter coefficient selected by the filter coefficient selection unit 161. Specifically, the product sum of the value of each pixel of the image data and the filter coefficient at the corresponding position is taken, and the value divided by the division coefficient is taken as the output value of the center pixel.

  7 and 8 are graphs showing the frequency characteristics of the filter coefficients shown in FIG. FIG. 7 shows the frequency characteristics of the emphasis filters 1, 2, and 3, the horizontal axis indicates the frequency, and the vertical axis indicates the gain. In FIG. 7, the “enhancement filter” is simply abbreviated as “emphasis”. FIG. 8 shows the frequency characteristics of the smoothing filters 1, 2, 3, 4, and 5, with the horizontal axis representing frequency and the vertical axis representing gain. In FIG. 8, “smooth filter” is simply abbreviated as “smooth”.

  The spatial filter unit 16 applies the same enhancement process to the high-density character region regardless of the image quality level. This is because mosquito noise is less noticeable inside the character, which is the character region on the high density side, and it is preferable to emphasize the sharpness of the character and to perform an emphasis process. By such processing, the characters are reproduced clearly and clearly. However, the high-density character region is not limited to the same enhancement filter regardless of the image quality level. If the image quality level is extremely low, mosquito noise may be conspicuous inside the character. Therefore, the degree of emphasis may be set to decrease as the image quality level decreases.

  The spatial filter unit 16 performs a smoothing process on the low-density character region when the image quality level is low (2 or less), and when the image quality level is lower, the degree of smoothness is increased and the image quality level is higher. In the case of (3 or more), the emphasis process is performed, and the degree of emphasis increases as the image quality level increases. This is because mosquito noise is conspicuous around the character, and it is necessary to sharpen the character and suppress mosquito noise. When the image quality level is lowered to some extent, the degree of mosquito noise increases. Therefore, it is possible to improve the quality of the output image by performing smoothing and blurring and removing the noise. On the other hand, since the degree of mosquito noise decreases when the image quality level increases to some extent, the output image can be improved in image quality by performing emphasis processing to make the characters clearer.

  Note that by applying an emphasis filter around the character, the edge becomes sharp and the character can be reproduced clearly, but if the mosquito noise is intense, the noise itself will be emphasized, causing unintended image quality degradation. . In order to eliminate this trade-off, the degree of enhancement and the degree of smoothing are switched according to the image quality level (that is, the degree of mosquito noise here). Here, the smoothing filter 3 smoothes the image data more than the smoothing filter 2. Further, the enhancement filter 1 is sharper than the enhancement filter 2. By the processing as described above, it is possible to remove mosquito noise and reproduce clear characters.

  The spatial filter unit 16 performs a smoothing process on the halftone dot region, and increases the degree of smoothness as the image quality level decreases. This is because block noise is conspicuous in a halftone image, and it is necessary to remove block noise. Therefore, it is possible to improve the quality of the output image by performing smoothing and blurring and removing the noise. Note that halftone dot components can be blurred by applying a smoothing filter to suppress moiré (interference fringes), and block noise can be removed. Will become extreme and cause unintentional image quality degradation. In order to eliminate this trade-off, the degree of smoothing is switched according to the image quality level (that is, the degree of block noise here). Here, the smoothing filter 3 smoothes the image data more than the smoothing filter 2. Furthermore, the smoothness level of the smoothing filter 4 and the smoothing filter 5 is increased. By such processing, it is possible to remove block noise and reproduce a clear image.

  The spatial filter unit 16 performs filter processing on the other regions not included in the character region and the halftone dot region, using filter coefficients as in Example 1, Example 2, and Example 3 of Table 2. For example, Example 1 is applied when the parameters of the compression process are generally close to high image quality. Since the degree of block noise and mosquito noise due to compression is generally small and it is not necessary to remove noise excessively, it is important to sharpen the image by enhancement processing. Further, in order to eliminate the trade-off described above, the degree of emphasis increases as the image quality level increases.

  On the other hand, when the parameters of the compression process are generally close to low image quality, Example 2 is applied. Since the degree of block noise and mosquito noise due to compression is generally large and it is necessary to positively remove noise, emphasis is placed on blurring by smoothing processing. In addition, in order to eliminate the trade-off described above, the degree of smoothness increases as the image quality level decreases.

  Furthermore, when the compression processing parameters are set so as to widen the image quality difference from high image quality to low image quality, Example 3 is applied. When the image quality level is high (in the case of 3 or more), importance is attached to the sharpening of the image. The higher the image quality level, the higher the degree of enhancement. Here, the enhancement filter 2 sharpens the image data more than the enhancement filter 3. When the image quality level is low (in the case of 2 or less), importance is placed on noise removal, and the lower the image quality level, the greater the degree of smoothness. Here, the smoothing filter 2 smoothes the image data more than the smoothing filter 1.

  In other areas, there are areas other than characters and halftone dots, such as areas with continuous gradation, such as photographic paper photographs, and solid areas with uniform pixel values over a wide range. An area such as a base area not to be included is included. In these areas, block noise tends to be noticeable. Therefore, it is necessary to achieve clear image reproduction and to suppress block noise. When the image quality level is lowered to some extent, the degree of block noise increases. Therefore, the image quality of the output image can be improved by performing smoothing processing and blurring and removing the noise. On the other hand, since the degree of block noise decreases as the image quality level increases to some extent, it is possible to improve the quality of the output image by performing enhancement processing to make the photographic image clearer. Note that photographic paper photographs can be sharpened by applying an emphasis filter and the image can be reproduced clearly. However, if the block noise is intense, the noise itself is emphasized, causing unintended image quality degradation. . In order to eliminate this trade-off, the degree of enhancement and the degree of smoothing are switched according to the image quality level (that is, the degree of block noise here). With the above processing, block noise and mosquito noise can be removed and clear pictures and characters can be reproduced.

  FIG. 9 is a flowchart showing an image processing operation in the image forming apparatus 1. The image processing operation in the image forming apparatus 1 proceeds to step a1 when the image processing apparatus 3 is activated to start processing. In step a1, the image processing apparatus 3 acquires a JPEG code to which header information, which is compressed image data, is added via the interface 2. The acquired JPEG code is delivered to the image quality level determination unit 11 and the JPEG decompression unit 12, and the process proceeds to step a2.

  In step a2, which is an image quality level determination step, the header analysis unit 111 of the image quality level determination unit 11 first extracts header information from the JPEG code, and then calculates a quantization table value from a predetermined section of the header information. Extracted and handed over to the quantization table determination unit 112. Then, the quantization table determination unit 112 determines the image quality level of the compressed image based on the passed quantization table value. Information on the image quality level determined by the image quality level determination unit 11 is delivered to the spatial filter unit 16.

  In step a3, which is a decoding process, the JPEG decompression unit 12 performs processing such as decoding the header information of the JPEG code and combining the code code, and decompresses the image data into an RGB digital signal. The image data expanded by the JPEG expansion unit 12 is delivered to the region separation unit 13 and the color correction unit 14.

  In step a4, which is a region separation process, the region separation unit 13 separates the image data expanded by the JPEG expansion unit 12 into a plurality of regions including at least one of a character region and a halftone region, and generates a region separation signal. Is generated. The region separation signal generated by the region separation unit 13 is delivered to the black generation / undercolor removal unit 15, the spatial filter unit 16, and the halftone generation unit 17, respectively.

  In step a5, the color correction unit 14 generates a CMY digital signal from the RGB digital signal that is the image data expanded by the JPEG expansion unit 12. The CMY digital signal generated by the color correction unit 14 is delivered to the black generation / under color removal unit 15.

  In step a6, the black generation / under color removal unit 15 generates a K signal from the CMY digital signal generated by the color correction unit 14, and subtracts a portion where the K signal generated from the original CMY digital signal is overlapped to obtain a new signal. A CMY digital signal is generated. As a result, the black generation / under color removal unit 15 converts the CMY digital signal into a CMYK digital signal.

  In step a7, which is a coefficient setting process, the spatial filter unit 16 is optimal for each pixel based on the image quality level information delivered from the image quality level judgment unit 11 and the region separation signal delivered from the region separation unit 13. Select and set the correct filter coefficient. Next, in step a8, which is a filtering process, the spatial filter unit 16 performs a filtering process on the image data of the CMYK digital signal using the set filter coefficient.

  In step a9, the halftone generation unit 17 performs pseudo tone reproduction processing on the image data of the CMYK digital signal so that the image can finally reproduce the tone in a pseudo manner based on the region identification signal. The image is output to the image output device 4 and the operation process is terminated.

  According to the image forming apparatus 1 as described above, the filter coefficient for performing the filter processing based on the region separation signal generated by the region separation unit 13 and the image quality level automatically determined by the image quality level determination unit 11. Therefore, it is possible to achieve good image reproduction with little image quality degradation due to compression processing for the image finally output from the image output device 4. In this embodiment, the number of stages of the image quality level is shown in four stages of 4: high image quality, 3: relatively high image quality, 2: relatively low image quality, and 1: low image quality, but the number of stages is limited to this. It is not a thing. In the present embodiment, eight sets of coefficients are shown as filter coefficients, but the number of sets is not limited.

  FIG. 10 is a block diagram showing a configuration of an image forming apparatus 30 according to another embodiment of the present invention. The image forming apparatus 30 is configured when the image output apparatus 4 is an image display apparatus 32 such as a liquid crystal display in the above-described embodiment shown in FIG. The image forming apparatus 30 and the image forming apparatus 1 shown in FIG. 1 are different from the image display apparatus 32 and the image processing apparatus 31. The image processing apparatus 31 has a black generation / undercolor removal unit due to the configuration of the image processing apparatus 3. 15 and the halftone generation unit 17 are omitted, and the same components are denoted by the same reference numerals and the description thereof is omitted.

  The image processing apparatus 31 includes an image quality level determination unit 11, a JPEG decompression unit 12, a region separation unit 13, a color correction unit 14, and a spatial filter unit 16. The processing content of each part is the same as that of the above-described embodiment. However, the color correction unit 14 does not perform conversion from the RGB digital signal to the CMY digital signal, and performs only the process of correcting the color for the RGB digital signal in order to faithfully reproduce the color by the image display device 32. The spatial filter unit 16 is supplied with image data of RGB digital signals, and the spatial filter unit 16 performs a filtering process on the image data and outputs the image data to the image display device 32.

  According to the above configuration, even when the input image data is displayed, it is possible to perform the optimum filter processing according to the image quality level and the region separation signal, and the noise generated by the compression processing is not noticeable. Display images can be obtained. This configuration can also be applied to a preview of image data stored on the hard disk.

  FIG. 11 is a block diagram showing a configuration of an image forming apparatus 40 according to still another embodiment of the present invention. The image forming apparatus 40 is different from the image forming apparatus 1 shown in FIG. 1 in that the image processing apparatus 41 is different, and the image processing apparatus 41 has a configuration in which the region separation unit 13 is removed from the configuration of the image processing apparatus 3. The same reference numerals are assigned to the configurations of and the description thereof is omitted.

  The image processing apparatus 41 includes an image quality level determination unit 11, a JPEG decompression unit 12, a color correction unit 14, a black generation / undercolor removal unit 15, a spatial filter unit 16, and a halftone generation unit 17. Yes. The processing content of each part is the same as that of the above-described embodiment. The characteristics of the image forming apparatus 40 are different from those of the image forming apparatus 1. In the spatial filter unit 16, the filter coefficient selection unit 161 uses the table shown in Table 2 based only on the image quality level determined by the image quality level determination unit 11. It is a point which becomes a structure which selects a filter coefficient with reference. In this case, the image processing apparatus 41 stores, for example, only the “other area” table in Table 2 in advance and refers to it. The calculation unit 162 performs a convolution operation on the image data using the filter coefficient selected by the filter coefficient selection unit 161. Specifically, the product sum of the value of each pixel of the image data and the filter coefficient at the corresponding position is taken, and the value divided by the division coefficient is taken as the output value of the center pixel.

  FIG. 12 is a flowchart showing an image processing operation in the image forming apparatus 40. The image processing operation in the image forming apparatus 40 proceeds to step b1 when the image processing apparatus 41 is activated to start processing. In step b1, the image processing apparatus 41 acquires a JPEG code to which header information, which is compressed image data, is added via the interface 2. The acquired JPEG code is delivered to the image quality level determination unit 11 and the JPEG decompression unit 12, and the process proceeds to step b2.

  In step b2, which is an image quality level determination step, the header analysis unit 111 of the image quality level determination unit 11 first extracts header information from the JPEG code, and then calculates a quantization table value from a predetermined section of the header information. Extracted and handed over to the quantization table determination unit 112. Then, the quantization table determination unit 112 determines the image quality level of the compressed image based on the passed quantization table value. Information on the image quality level determined by the image quality level determination unit 11 is delivered to the spatial filter unit 16.

  In step b3, which is a decoding process, the JPEG decompression unit 12 performs processing such as decoding the header information of the JPEG code and combining the code code, and decompresses the image data into an RGB digital signal. The image data expanded by the JPEG expansion unit 12 is delivered to the color correction unit 14.

  In step b4, the color correction unit 14 generates a CMY digital signal from the RGB digital signal that is the image data expanded by the JPEG expansion unit 12. The CMY digital signal generated by the color correction unit 14 is delivered to the black generation / under color removal unit 15.

  In step b5, the black generation / under color removal unit 15 generates a K signal from the CMY digital signal generated by the color correction unit 14, and subtracts a portion where the K signal generated from the original CMY digital signal is overlapped to obtain a new signal. A CMY digital signal is generated. As a result, the black generation / under color removal unit 15 converts the CMY digital signal into a CMYK digital signal.

  In step b6, which is a coefficient setting process, the spatial filter unit 16 selects and sets an optimum filter coefficient for each pixel based on the image quality level information delivered from the image quality level determination unit 11. Next, in step b7, which is a filtering process, the spatial filter unit 16 performs a filtering process on the image data of the CMYK digital signal using the set filter coefficient.

  In step b8, the halftone generation unit 17 performs pseudo gradation reproduction processing on the image data of the CMYK digital signal so that the image can finally reproduce the gradation, and the image output device 4 To complete the operation process.

  According to the image forming apparatus 40 as described above, since the filter coefficient for performing the filter processing is appropriately switched based on the image quality level automatically determined by the image quality level determination unit 11, the image output device 4 finally performs the switching. Therefore, it is possible to achieve good image reproduction with little image quality degradation due to compression processing.

  As still another embodiment of the present invention, in order for a computer to function as any of the image processing apparatuses of the above-described embodiments, program code (execution format program, intermediate code program, and It is also possible to provide a computer-readable recording medium on which the program code is recorded, and at least one of the source programs. According to the present embodiment, it is possible to provide a portable recording medium on which a program code for performing the above-described image processing method is recorded. In addition, a general-purpose computer such as a personal computer is read through a recording medium such as a CD-ROM (Compact Disc-Read Only Memory) or downloaded from a network, and an appropriate filtering process is performed on the input image. Thus, a good image can be output. Similarly, for digital copiers and multi-function machines that perform software processing with a DSP (Digital Signal Processor) or the like, the program is read into a flash memory or a rewritable recording medium, and an appropriate filtering process is applied to the input image. As a result, a good image can be output.

  As a recording medium for recording the program code, a memory (not shown) such as a ROM itself may be a program medium because processing is performed by a microcomputer. However, it may be a program medium provided with a program reading device as an external storage device and readable by inserting a recording medium therein.

  In any case, the stored program code may be configured to be accessed and executed by the microprocessor, or in any case, the program code is read and the read program code is the microcomputer. The program code may be downloaded to a program storage area (not shown) and executed. It is assumed that this download program is stored in the main device in advance.

The program medium is a recording medium configured to be separable from the main body, and is a tape system such as a magnetic tape or a cassette tape, a magnetic disk such as a floppy (registered trademark) disk or a hard disk, or a CD-ROM / MO (Magneto Optical disc). ) / MD (Mini Disc) / D
Optical discs such as VD (Digital Versatile Disc), IC (Integrated)
Circuit) cards (including memory cards) / optical cards, mask ROM, EPROM (Erasable Programmable Read Only Memory), EEPROM (
Electrically Erasable Programmable Read Only Memory), or a medium carrying a fixed program including a semiconductor memory such as a flash ROM.

  In addition, the medium may have a system configuration capable of connecting a communication network including the Internet so that the program code is fluidly carried so as to download the program code from the communication network. When the program code is downloaded from the communication network in this way, the download program may be stored in the main device in advance, or may be installed from another recording medium. The present invention can also be realized in the form of a computer data signal embedded in a carrier wave in which the program code is embodied by electronic transmission.

The recording medium is read by a digital color image forming apparatus or a program reading apparatus provided in the computer system 50, whereby the above-described image processing method is executed. FIG. 13 is a configuration diagram illustrating an example of the computer system 50. The computer system 50 includes an image input device 51 such as a flatbed scanner, a film scanner, or a digital camera, a computer 52 that performs various processes such as the above-described image processing method by loading a predetermined program, and processing results of the computer 52. CRT to display (
(Cathode Ray Tube) An image display device 53 such as a display or a liquid crystal display, a printer 54 that outputs a processing result of a computer to paper, and an input device 55 such as a keyboard and a mouse are configured. Furthermore, the computer 52 is provided with a modem or the like as a communication means for connecting to a server or the like via a network.

  In particular, when the above-described image processing method is executed by the computer system 50 shown in FIG. 13, this image processing method can be used according to the user's preference. For example, the user can adjust the quantization table used for compression processing, filter coefficients used for filter processing, multiple threshold values used for region separation, etc. (by rewriting the program) to obtain an output image that meets user preferences Can do.

DESCRIPTION OF SYMBOLS 1,30,40 Image forming apparatus 2 Interface 3,31,41 Image processing apparatus 4 Image output apparatus 11 Image quality level judgment part 12 JPEG decompression part 13 Area separation part 16 Spatial filter part 32 Image display apparatus 50 Computer system 111 Header analysis part 112 Quantization table determination unit 161 Filter coefficient selection unit 162 Calculation unit

Claims (8)

An image quality level judging means for judging an image quality level based on a value of a quantization table of compressed image data;
Decoding means for decoding the compressed image data;
Coefficient setting means for setting a filter processing coefficient based on the image quality level determined by the image quality level determination means;
An image processing apparatus comprising: filter means for performing filter processing on the image data decoded by the decoding means using the coefficient set by the coefficient setting means. The image data decoded by the decoding means further comprises an area separating means for separating the image data into a plurality of areas including at least one of a character area and a halftone dot area,
2. The image processing according to claim 1, wherein the coefficient setting unit sets a filter processing coefficient based on a region separation result obtained by the region separation unit and an image quality level determined by the image quality level determination unit. apparatus. The image quality level determining means includes
An information analysis unit that extracts the value of the quantization table from the compressed image data;
The image processing according to claim 1, further comprising: a level determination unit that determines an image quality level by comparing a quantization table value extracted by the information analysis unit with a plurality of predetermined threshold values. apparatus.   An image forming apparatus comprising the image processing apparatus according to claim 1. An image quality level determining step of determining an image quality level based on a value of a quantization table of compressed image data;
A decoding step of decoding the compressed image data;
A coefficient setting step for setting a filter processing coefficient based on the image quality level determined in the image quality level determination step;
And a filtering step of performing a filtering process on the image data decoded in the decoding step using the coefficient set in the coefficient setting step.
The image data decoded in the decoding step further comprises a region separation step for separating the image data into a plurality of regions including at least one of a character region and a halftone dot region,
6. The image processing according to claim 5, wherein, in the coefficient setting step, a filter processing coefficient is set based on the region separation result obtained in the region separation step and the image quality level determined in the image quality level determination step. Method.   The image processing program for implement | achieving the image processing apparatus as described in any one of Claims 1-3, Comprising: The image processing program for functioning a computer as said each means.   A computer-readable recording medium on which the image processing program according to claim 7 is recorded.

Images