JP5446525B2 - Image processing apparatus and image processing method - Google Patents

Description

  The present invention relates to an image processing apparatus and an image processing method in an image forming apparatus such as a printer, a copying machine, and a facsimile.

  Image forming apparatuses such as printers, copiers, and facsimiles are supplied with image data having gradation data corresponding to pixels from a host computer or a built-in scanner, and a controller unit that generates image reproduction data from the input gradation values ( An image processing apparatus) and a print engine for forming an image based on the image reproduction data. The image reproduction data is, for example, binary data indicating an ink adhesion region or multi-value data indicating the area of the ink adhesion region.

  The controller unit has, for example, a halftone processing unit that outputs image reproduction data from an 8-bit input gradation value, and the halftone processing unit expresses the gradation of the input gradation value as an area ratio of the ink adhesion region. Based on a predetermined gamma characteristic such as an AM screen or an FM screen expressed by the density of the ink adhesion area, image reproduction data suitable for the print engine is output. On the other hand, in the case of electrophotography, for example, in the case of electrophotography, a latent image is formed by irradiating a charged photosensitive drum with energy light based on image reproduction data, and toner, which is dry ink, is attached to the latent image. A toner image is formed and transferred to a printing medium such as paper. The toner image is transferred to a sheet to form an ink adhesion area. In the case of inkjet, an ink adhesion region is formed by directly spraying ink that is wet ink onto a sheet.

  In such an image forming apparatus, it is desired that the input gradation value of the image data and the output density on the print medium have an ideal linear relationship. At the design stage, there is a conversion table for converting the input tone value of the halftone processing unit into image reproduction data so that the input tone value and output density have a linear relationship based on the characteristics of the print engine. Designed. However, the linear relationship may not be maintained due to variations in the characteristics of the print engine at the time of shipment from the factory, or due to variations in the characteristics of the print engine due to aging or environmental changes.

  Therefore, based on the image data for forming the patch image, a patch image having a plurality of gradations is formed on the photosensitive drum or intermediate transfer medium of the print engine, and the output density of the patch image is read by the density sensor. The relationship between the input tone value and the read output density (exactly the output density on paper) is obtained, and a correction gamma table is created so that the relationship becomes an ideal linear characteristic, and halftone processing is performed based on this It has been proposed to correct some conversion tables. Alternatively, it has been proposed to correct an input tone value using a correction gamma table and supply the corrected input tone value to a halftone processing unit.

  Alternatively, it is also proposed that a patch image is printed on a sheet and the user manually corrects the conversion table of the halftone processing unit based on the characteristics.

  The above density correction control is described in the following Patent Documents 1 and 2.

  FIG. 1 is a diagram showing linear output density characteristics. FIG. 1A shows output density characteristics with respect to input gradation values of 0 to 255. In this example, the paper white density with no ink attached is 0.1, and the output density in the state of solid printing is 1.5, and the paper white density is 0.1 at the input gradation “0”. , The input gradation value “255” is formed to a solid density of 1.5, and the input gradations “1” to “254” are converted so that the output density has a linear relationship with the input gradation value. Is designed or corrected.

  FIG. 1B shows output density characteristics in the low density region (input gradation values “0” to “17”) in FIG. The conversion table is designed or corrected so that a black circle output density is obtained for each input tone value so that the linear relationship shown by the broken line is obtained even in the low density region.

  FIG. 2 is a chart showing the output density with respect to the input gradation value in the low density region shown in FIG. As can be understood from the specific values, the output density has a linear relationship with the input gradation value.

  Thus, recent image forming apparatuses such as printers and copiers can form the output density of the output image linearly from the minimum gradation value to the maximum gradation value by improving the printing engine and ink (toner). It is possible. Accordingly, as shown in FIG. 1, the halftone conversion table is designed so that the output density has a linear characteristic corresponding to all the gradation values of the input gradation values 0 to 255.

JP 2008-131114 A JP 2006-33402 A

  FIG. 3 is a diagram showing image quality degradation due to engine characteristic changes. FIG. 3A shows a printing result obtained by printing a sphere with the linear output density characteristic shown in FIG. This spherical image is an image representing a shadow when light is irradiated on the sphere from the upper right, and the upper right area of the sphere is a highlight area with a very low output gradation.

  FIG. 3B shows a sphere print result when the characteristics of the print engine of the image forming apparatus fluctuate. In this example, as in the output density characteristic shown on the right side, the solid density fluctuates from 1.5 to 1.6, and the output density characteristic of the broken line fluctuates to the output density characteristic of the solid line.

  According to the conventional density correction method, the conversion table is corrected so that an output density on a straight line connecting the paper white density 0.1 and the solid density 1.6 is formed. As a result of this correction, the ink adhesion area is not formed at the input gradation value 0, and the paper white density does not change to 0.1. However, for example, the ink adhesion area is first formed for the input gradation value 1, and its output The density is higher than the dashed line before correction. In other words, the output density in the extremely low input gradation area where the ink adhesion area is formed for the first time becomes higher than before correction, and a sudden density change is formed in the spherical highlight area, resulting in pseudo contours and tone jumps. To do. In the figure, although it is not easy to discriminate, a circle is formed. In particular, in a highlight area with an extremely low input gradation value, a large tone jump due to the minimum ink output density when the ink adhesion area is formed for the first time as the input gradation value increases is particularly noticeable to the human eye. It becomes a conspicuous presence and causes a decrease in image quality.

  Accordingly, an object of the present invention is to provide an image processing apparatus and an image processing method that improve the image quality in a low density region.

A first aspect of the image processing apparatus is an image processing apparatus that generates image reproduction data indicating an ink adhesion region from image data having an input gradation value.
A controller having a halftone processing unit for generating the image reproduction data from the input gradation value based on a predetermined gamma characteristic;
The controller generates the image reproduction data from the patch image data by the halftone processing unit, causes the print engine to form a patch image based on the image reproduction data, and inputs the input gradation value of the patch image data and the patch A correction gamma table is generated based on the input gradation / output density characteristic indicating the relationship with the output density of the image so that the output density characteristic with respect to the input gradation value becomes the first characteristic,
The halftone processing unit generates image reproduction data from the input gradation value based on the correction gamma table,
The first characteristic is a medium input gradation value area, and has a linear characteristic that linearly connects a minimum output density and a maximum output density, and a low input gradation value area lower than the medium input gradation value area Thus, the ink has a low input gradation characteristic in which the minimum ink output density corresponding to the minimum ink adhesion region is lower than the output density of the linear characteristic.

  According to the first aspect, since the minimum ink output density is suppressed, the image quality in the highlight area is improved.

  In the first aspect, according to a preferred embodiment, the low input gradation characteristic is obtained by the linear characteristic from the minimum ink output density as the input gradation value increases in the low input gradation value region. It has a characteristic that asymptotically approaches the output density.

A second aspect of the image processing apparatus is an image processing apparatus that generates image reproduction data indicating an ink adhesion region from image data having an input gradation value.
A controller having a halftone processing unit for generating the image reproduction data from the input gradation value based on a gamma table;
The gamma table has a linear characteristic in which an output density characteristic with respect to the input gradation value linearly connects between a minimum output density and a maximum output density in a middle input gradation value region, In the low input gradation value region lower than the value region, the minimum ink output density corresponding to the minimum ink adhesion region is formed to have a low input gradation characteristic that is lower than the output density of the linear characteristic.

  According to the second aspect, since the minimum ink output density is suppressed, the image quality in the highlight area is improved.

It is a figure which shows a linear output density characteristic. 3 is a chart showing output density with respect to input gradation values in a low density region shown in FIG. It is a figure which shows the image quality fall by the characteristic change of an engine. 1 is a configuration diagram of an image forming apparatus having an image processing apparatus in the present embodiment. It is a flowchart figure which shows the control procedure by an automatic image density control program. It is a figure which shows the various graphs produced | generated in automatic image density control. It is a figure which shows the various graphs produced | generated in automatic image density control. It is a figure which shows the halftone process part using a correction | amendment gamma table. FIG. 6 is a graph summarizing automatic image density control. It is a graph which shows the automatic image density control in this Embodiment. FIG. 6 is a characteristic diagram of an input tone value and a target output density of the image processing apparatus according to the present embodiment. This is a specific example of the target output density with respect to the input gradation value of the low input gradation characteristic 40 of FIG. It is a figure which shows the low input gradation characteristic 40 in this Embodiment. It is a figure which shows the change of the inclination of the low input gradation characteristic 40 of FIG. It is a figure which shows the example of the corrected screen gamma table which the halftone process part in this Embodiment refers. It is a figure which shows the example of the corrected screen gamma table which the halftone process part in this Embodiment refers. It is a figure which shows an example of the threshold value matrix in this Embodiment.

  Hereinafter, the present embodiment will be described with reference to the drawings.

  FIG. 4 is a configuration diagram of an image forming apparatus having the image processing apparatus according to the present embodiment. FIG. 4 shows the configuration of a printer as an example of the image forming apparatus. In addition to printers, copiers and facsimiles with built-in scanner functions are similar image forming apparatuses.

  The printer 20 is connected to the host computer 10 via a network or the like, and receives print data from the printer driver 12 installed in the host computer 10. The printer 20 includes a controller unit 200 that generates image reproduction data based on print data, and a print engine 250 that forms an image based on image reproduction data. This controller unit 200 is an image processing apparatus.

  The controller unit 200 includes a CPU 202 that controls the entire controller unit, a memory 206 that stores image data, an interface IF that receives print data, and a color conversion unit that converts image data based on the print data from RGB to CMYK. 208, a halftone processing unit 210 that converts input gradation values of color-converted CMYK image data into image reproduction data based on a predetermined gamma characteristic, and a pulse width modulator 212 that performs pulse width modulation on the image reproduction data. And have. Further, the non-volatile memory 214 stores a screen gamma table 216, a correction gamma table 218, and a sensor value / on-paper density value conversion table 220.

  The halftone processing unit 210 refers to a screen gamma table 216 having characteristics corresponding to an AM screen or an FM screen, and converts an input gradation value of image data into image reproduction data for image formation. The image reproduction data is data defining an ink adhesion area for each pixel, and is, for example, multi-gradation value or binary data. The multi-tone value indicates the area of the ink adhesion region in the pixel, and the binary value indicates whether ink is adhered to the pixel.

  The print engine 250 includes an engine controller 252, light emitting means 254 including a laser diode or a line-shaped light emitting element, a photosensitive drum 256, a developing unit, a transfer unit, and a fixing unit (not shown). The engine controller 252 forms a latent image on the photosensitive drum 256 by irradiating a beam from the light emitting means 254 based on the pulse of the image reproduction data generated by the controller unit 200, and the developing unit converts the latent image to the toner. The image is developed, the toner image is transferred to a printing medium, for example, paper by a transfer unit, and the toner image is fixed on the paper by a fixing unit. The toner is dry ink, and the area where the toner adheres is the ink adhesion area.

  The print engine 250 further detects the density of the toner image formed on the photosensitive drum or an intermediate transfer medium (not shown) by a built-in density sensor 260, and the sensor interface S-IF in the controller unit 200 detects the density of the toner image. The value is output.

[Automatic image density control]
The controller unit 200 of the printer has an automatic image density control program 204, and by executing this program, control is automatically performed so that the output image density with respect to the input gradation value maintains an optimum characteristic. The outline of this automatic image density control is as follows.

  First, a patch image having a predetermined gradation step from a low density to a high density is formed by the print engine 250 at a predetermined timing such as when the printer is turned on or when the photosensitive drum or the developing unit is replaced. Specifically, the controller unit 200 generates image reproduction data from the patch image data by the halftone processing unit 210, and the print engine 250 forms a patch image 258 corresponding to the image reproduction data on the photosensitive drum 256. The patch image 258 is the above-described toner image, and may be formed on an intermediate transfer medium (not shown). Then, the density sensor 260 measures the density of the patch image 258 and the sensor value is supplied to the controller unit 200.

  The automatic image density control program 200 obtains an output density value based on the sensor value, and obtains an input tone / output density characteristic indicating the relationship between the input tone value of the patch image data and the output density value. Then, a correction gamma table 218 for correcting the input gradation value or the output gradation value is generated so that the input gradation / output density characteristic has an ideal linear relationship, for example, and is stored in the nonvolatile memory 214. Record. Then, the halftone processing unit 210 refers to the corrected screen gamma table 216 obtained by correcting the screen gamma table 216 based on the corrected gamma table 220, and converts the input gradation value data into image reproduction data. Alternatively, the halftone processing unit 210 corrects the input tone value data with reference to the correction gamma table 218, and converts the corrected input tone value data into image reproduction data with reference to the screen gamma table 216.

  FIG. 5 is a flowchart showing a control procedure by the automatic image density control program. 6 and 7 are graphs showing various graphs generated in the automatic image density control. The automatic image density control will be described in detail with reference to these, and then the problem and automatic image density control according to the present embodiment will be described.

  In FIG. 5, the automatic image density control program 204 of the controller unit generates a patch image (S10). The patch image has an input gradation value obtained by dividing 256 printable gradations at equal intervals of 255 / (Pmax-1) according to the number of patch gradations Pmax, using a screen gamma table used during normal printing. It consists of a patch pattern.

  Next, a sensor value is acquired for the patch image in the photosensitive drum 256 in the print engine (S12). FIG. 6A is an example of a graph showing the sensor value with respect to the input tone value of the patch pattern. In FIG. 6A, the horizontal axis represents input gradation values from 0 to 255, and the vertical axis represents sensor values from 0 to 1023.

  Here, in this specification, the sensor value means a standardized sensor value. The standardized sensor value means that the detected sensor value (voltage value) is 10 bits (0 to 1023) between the standardized voltage of the sensor (for example, 5V) and the voltage when there is no received light (sensor offset voltage). This is a standardized value. Therefore, the sensor normalization voltage is normalized to 1023, the sensor offset voltage is normalized to 0, and the toner is not adhered to the photoconductor (paper white, corresponding to the minimum density) is set to a high value on the 1023 side. The solid state with 100% area ratio (corresponding to the maximum density) is normalized to a low value on the 0 side. Therefore, the normalized sensor value can be used as an absolute value substantially the same as the output voltage of the sensor, that is, the sensor value. For example, if the normalized sensor value is 200, the on-paper density is 1.42, if the normalized sensor value is 120, the on-paper density is 1.63, and 240 corresponds to 1.38. Hereinafter, the normalized sensor value is simply referred to as a sensor value.

For the input tone values on the horizontal axis in FIG. 6A, the tone value Pi (0 ≦ i <Pmax) of the patch pattern is obtained by the following equation. In this expression, INT is a truncation process, and according to this expression, the input gradation values of the patch pattern having the number of gradations Pmax are arranged at equal intervals in the 0 to 255 gradation range.
P _i = INT ((i * 255.) / (Pmax-1) + 0.5)
As shown in FIG. 6A, the sensor value is about 720 at the input gradation value 0 corresponding to the minimum density, and the sensor value is about 160 at the input gradation value 255 corresponding to the maximum density.

  Next, the sensor value is converted into an output density value (S14). That is, the output density value (on-paper density value) of each patch pattern is obtained using a sensor value / output density value conversion table obtained in advance. FIG. 6B is an example of a sensor value / output density value conversion table. The horizontal axis is a value obtained by converting the sensor value of 0 to 1023 into the sensor value of 0 to 255, and the vertical axis is the output density value. This conversion table is generated based on the acquired output density and the sensor value of the patch pattern after the patch image is formed on the paper of the printing medium in advance and the output density on the paper is measured by the colorimeter.

  FIG. 7A is a graph showing the correspondence between the input tone value and the output density value of the patch pattern by converting the sensor value into the output density value from FIGS. 6A and 6B. A solid line 30 in FIG. 7A is an output density value with respect to an input gradation value obtained from a patch pattern, and indicates the characteristics of the print engine. This output density value is a value obtained by converting the sensor value using the sensor value / output density conversion table of FIG. The alternate long and short dash line changes from the minimum output density 0.1 for the minimum input gradation value 0 to the maximum output density 1.5 for the maximum input gradation value 255 in a linear relationship with respect to the input gradation. This is an ideal linear characteristic IDEA. The engine characteristic indicated by the solid line 30 in FIG. 7A is an example in which the output density is higher than the ideal linear characteristic.

  Next, the automatic image density control program generates a correction gamma table so that the input tone / output density characteristic 30 of FIG. 7A becomes a linear characteristic IDEA (S16). This correction gamma table is a table for correcting the output density obtained in FIG. 7A to a desired output density. Here, the correction is made so that the output density becomes linear according to the input gradation value. Create a table to do FIG. 7B shows an example of the correction gamma table. In this example, the corrected gamma table table CGT has a downwardly convex characteristic. In subsequent normal printing, the input gradation is corrected based on the correction gamma table CGT (S18).

  This corrected gamma table CGT means that the output density 74 can be obtained if the input gradation value 64 before correction is used as the input gradation value 74 after correction. In other words, before correction, the output density value obtained for a certain input gradation value is higher than the ideal value according to the input gradation / output density characteristic of FIG. As a result, the input gradation value 74 is corrected and converted to 64 using the correction gamma table, and the input gradation value 64 subjected to the correction conversion is input to the halftone processing unit, so that a lighter image can be output. It becomes like this.

  FIG. 8 is a diagram illustrating a halftone processing unit using a correction gamma table. The halftone processing unit 210 converts the corrected input gradation value INa = 74 into the uncorrected input gradation value INb = 64 by the correction gamma table 218 and converts it by the screen gamma table 216 to convert the image reproduction data. Generate. The output density D printed by this image reproduction data is 74, which is the same gradation value 74 as the corrected input gradation value 74, which is an ideal linear characteristic.

  The halftone processing unit 210 may correct the screen gamma table 216 based on the corrected gamma table 218 and convert the corrected input tone value INa into image reproduction data using the corrected screen gamma table.

  FIG. 9 is a graph summarizing the above automatic image density control. FIG. 9A is a graph of the output density value with respect to the input tone value based on the patch image, which is the same as FIG. 7A. The input gradation value on the horizontal axis corresponds to the input gradation value INb before correction in FIG. FIG. 9B is a graph for simply converting the output density on the vertical axis to the output density on the horizontal axis. FIG. 9C is a graph of ideal input tone values and output density values. The maximum value (D = 1.5) of the output density value converted from the sensor value and the minimum value of the output density value. (D = 0.1) and the linear relationship IDEA.

  Finally, FIG. 9D shows the relationship between the input gradation value INb before correction and the input gradation value INa after correction so that the input gradation value and the output density value have an ideal linear relationship. This is a correction gamma table CGT, which is the same as FIG. As described with reference to FIG. 8, the post-correction input tone value INa is converted to the pre-correction input tone value INb by the correction gamma table, and is converted to image reproduction data by the screen gamma table. FIG. 9D also shows the relationship between the input gradation value INa = 64 before correction and the input gradation value INb = 74 after correction.

[This embodiment]
FIG. 10 is a graph showing automatic image density control in the present embodiment. Each graph of FIGS. 10A to 10D is the same as FIG. 9 is different from FIG. 9 in that the output density with respect to the maximum input gradation value in FIG. 10A is 1.6, and the input gradation and output density after correction of the one-dot chain line in FIG. The characteristic is the characteristic IDEA of the output gradation value linear with respect to the input gradation in the intermediate gradation value area and the high gradation value area other than the low gradation value area, and in the low gradation value area. The low input gradation characteristic 40 has an output density lower than the linear characteristic IDEA.

  Accordingly, the characteristics of the target output density with respect to the output density detected by the sensor in FIG. 10C are the low input gradation characteristics 40 in the low density area, and the medium density area and the high density area other than the low density area are linear. Characteristic IDEA. Further, the correction gamma table CGT of FIG. 10D has a low input gradation characteristic 42 in the low input gradation value region, and the corrected input gradation value INa is set to a lower input gradation value INb before correction. It is corrected.

  As shown in FIG. 3B, when the output density with respect to the maximum input gradation value increases from 1.5 to 1.6 due to the change in the print engine characteristics, the output density and the maximum input level with respect to the minimum input gradation value are increased. If a correction gamma table is created so that the linear characteristic connecting the output density to the tone value, the minimum ink adhesion area when the ink first adheres as the input gradation value increases, especially in the low input gradation area. Corresponding minimum ink output density becomes higher than the intended design density, resulting in image quality degradation.

  In the present embodiment, in order to avoid such a situation, the first characteristic corrected by the correction gamma table (the one-dot chain line and the broken line in FIG. 10A) has a minimum value in the middle input gradation value region. It has a linear characteristic IDEA that linearly connects the minimum output density for the input gradation value “0” and the maximum output density for the maximum input gradation value “255”, and has a low input floor lower than the middle input gradation value region. In the tone value area, a low input gradation characteristic 40 is obtained in which the minimum ink output density corresponding to the minimum ink adhesion area is lower than the output density of the linear characteristic IDEA. Further, the low input gradation characteristic 40 has a characteristic that gradually approaches the output density of the linear characteristic IDEA from the minimum ink output density as the input gradation value increases in the low input gradation value region. Further, the linear characteristic IDEA may also be provided in a high input gradation value region higher than the middle input gradation value region and up to the maximum input gradation value.

  FIG. 11 is a characteristic diagram of the input tone value and the target output density of the image processing apparatus according to the present embodiment. FIG. 11A shows the target output density characteristics for the input gradation values from the minimum value to the maximum value, and FIG. 11B shows the target output density characteristics in the low input gradation value region. In the present embodiment, the characteristics of the input tone value and the target output density of the image processing apparatus are the minimum input tone value in the medium input tone value region and the high input tone value region other than the low input tone value region. The linear characteristic IDEA connects the output density with respect to the value and the output density with respect to the maximum input gradation value. On the other hand, the low input gradation value region has a low input gradation characteristic 40 in which the minimum ink output density corresponding to the minimum ink adhesion area is lower than the output density of the linear characteristic IDEA.

  FIG. 11B specifically shows the low input gradation characteristic 40. First, the target output density 40-0 with respect to the input gradation value “0” is the paper white density of 0.1. That is, the image reproduction data for the input gradation value “0” is “0”, and ink such as toner is not attached on the paper. Then, the image reproduction data becomes a value other than “0” with respect to the input gradation value “1”, and the ink adhesion area formed thereby is the minimum ink adhesion area formed first. The minimum ink output density 40-1 due to this minimum ink adhesion region is lower than the output density of the linear characteristic IDEA. Furthermore, the output densities 40-2 to 40-7 for the input gradation values “2” to “7” are also lower than the output density of the linear characteristic IDEA. However, these output densities 40-2 to 40-7 have characteristics that gradually approach the output density of the linear characteristic IDEA from the minimum ink output density 40-1 due to the minimum ink adhesion area as the input gradation value increases. .

  FIG. 12 is a specific example of the target output density with respect to the input gradation value of the low input gradation characteristic 40 of FIG. As is clear from comparison of FIG. 12 with FIG. 2, the target output density for the input gradation values “1” to “8” shown in FIG. 12 is lower than the output density of the linear characteristic of FIG. The target output density for “9” or more is the same as the output density of the linear characteristic of FIG. In particular, since the target output density (minimum ink output density) for the input gradation value “1” is set to “0.101”, which is lower than the output density “0.105” of the linear characteristic of FIG. It is possible to suppress deterioration in image quality in the highlight area (B).

  FIG. 13 is a diagram showing the low input gradation characteristic 40 in the present embodiment. FIG. 13 shows target output densities 40-2, 40-4, 40-6 for input gradation values “2”, “4”, “6”, and “8” in the low input gradation characteristic 40 of FIG. For 40-8, the straight slope connected with the paper white density 40-0 with respect to the minimum input gradation value is indicated by a two-dot chain line. In the low input gradation characteristic 40, the output density 40-2 to 40-8 has a characteristic that gradually approaches the output density of the linear characteristic IDEA. Therefore, the slope of the straight line indicated by the two-dot chain line is the slope “of the linear characteristic IDEA. Gradually approach 1 ”.

  FIG. 14 is a diagram showing a change in the slope of the low input tone characteristic 40 of FIG. When the input gradation values are “1” to “8”, the gradient gradually approaches the gradient “1” of the linear characteristic, that is, gradually approaches.

  FIGS. 15 and 16 are diagrams showing examples of the corrected screen gamma table referred to by the halftone processing unit in the present embodiment. This example is an example of a table reference type dither matrix in which a screen gamma table having a correspondence between an input gradation value and image reproduction data is provided corresponding to each pixel.

  FIG. 15A shows five types of screen gamma tables. The horizontal axis corresponds to the input gradation value, and the vertical axis corresponds to the image reproduction data. This image reproduction data is converted into a pulse width by the PWM 212 as described in FIG. The screen gamma table GT0 has a characteristic that the image reproduction data rises from 0 to 255 at a low input gradation value, and the other screen gamma tables GT1 to GT4 have image reproduction data from 0 at a higher input gradation value. It has a characteristic of rising up to 255. Therefore, in this screen gamma table, the characteristic of the low input gradation value of the table GT0 has the low input gradation characteristic 40.

  FIG. 15B shows an index table indicating which screen gamma table should be referred to in accordance with the pixel position. This index table 50 is a 5 × 5 table, and each element indicates a table number indicating which screen gamma table should be referred to. In the index table 50, cells 52 composed of five cross-shaped elements are tiled. Then, as the input gradation value increases from “0” to “255”, an ink adhesion region is first formed in the pixel of the table GT0 at the center of the cell 52, and sequentially in the tables GT1, GT2, GT3, GT4. An ink adhesion region is formed on the pixel. That is, the cell 52 corresponds to a halftone dot formation region, and the halftone dot area due to the ink adhesion region increases sequentially in response to an increase in input gradation value.

  FIG. 16 is an enlarged view of the low input gradation value region 40 of the screen gamma table GT0 that forms the ink adhesion region with the low input gradation value. The enlarged view of FIG. 16 is similar to FIG. That is, the pulse width (image reproduction data) corresponding to the input gradation values “1” to “9” is lower than the pulse width IDEA corresponding to the linear characteristic. Thereby, in particular, since the pulse width (image reproduction data) for the input gradation value “1” is lower than the pulse width IDEA corresponding to the linear characteristic, the minimum ink output density becomes lower than the linear output density characteristic, and the highlight. It is possible to suppress image quality degradation in the region.

  The table reference type screen corresponding to the gamma table referred to by the halftone processing unit of the present embodiment can also be realized by a threshold matrix having threshold values of 1 to 256 in each element. In the case of the threshold matrix, the screen processing unit compares the input gradation value with the threshold value of the threshold matrix, and generates image reproduction data indicating that an ink adhesion region is to be formed in a pixel whose input gradation value is equal to or greater than the threshold value. , Image reproduction data indicating that an ink adhesion region is not formed in a pixel whose input gradation value is less than the threshold value is generated. That is, the image reproduction data is binary data indicating whether or not to form an ink adhesion region in the pixel. However, when the input gradation value is “0”, the ink adhesion region is not formed.

  Such a threshold matrix can be corrected by an automatic density correction algorithm. For example, by correcting the threshold value of the threshold matrix lower, the number of pixels on which the ink adhesion area is formed can be increased and the target output density can be corrected higher. Conversely, the threshold value of the threshold matrix is corrected higher. As a result, the number of pixels on which the ink adhesion area is formed can be reduced and the target output density can be corrected to be low.

  In order to have the low gradation region characteristic 40 of the present embodiment in such a threshold matrix, the number of pixels in which the ink adhesion region is formed with the input gradation value “1” is set to the pixel number in the case of the linear characteristic. The number may be less than the number. Therefore, the minimum ink output density can be lowered by correcting the threshold value matrix so as to reduce the number of threshold values “1”.

  FIG. 17 is a diagram illustrating an example of a threshold matrix in the present embodiment. FIG. 17A shows a threshold matrix before correction, and FIG. 17B shows a threshold matrix after correction. This threshold matrix is 16 × 16 and includes two cells (center and four corners) having thresholds 1 to 128. Therefore, unlike the above embodiment, the input gradation value is assumed to be 7 bits from 0 to 127.

  In the threshold matrix before correction in FIG. 17A, the threshold value Vth = 1 is arranged for the center pixels of the two cells, and the number of pixels in which the ink adhesion region is formed with the input gradation value “1” is two. become. That is, the minimum ink output density is formed by the ink adhesion region of the two pixels with the threshold value Vth = 1.

  On the other hand, in the threshold matrix after correction in FIG. 17B, the threshold Vth = 1 at the center of the center cell of the two cells, but the threshold Vth = 2 at the center of the upper left cell. Yes. That is, no ink adhesion region is formed in the upper left cell center pixel even with the input gradation value “1”. As a result, the number of pixels in which the ink adhesion area is formed with the input gradation value “1” is one, and the minimum ink output density is kept low.

  In the above embodiment, it has been described that when the correction gamma table is formed in the automatic density correction, the low input gradation characteristic is obtained. However, this embodiment can be applied not only to automatic density correction but also to the design of a halftone gamma table referred to by the halftone processing unit. In other words, the halftone gamma table is a table that converts input tone values into image reproduction data that defines the ink adhesion formation region, but the target output density for the input tone values is low in the low tone value region. It is designed to have a tone characteristic 40 and a linear characteristic IDEA that connects the paper white density and the solid density in the gradation value area higher than the low gradation value area. By doing so, the minimum ink output density when ink is first deposited can be lowered compared with the case where the linear characteristic IDEA is provided in the low gradation value region, and the image quality in the highlight region is reduced. The decrease can be suppressed. In a region other than the low gradation value region, a linear output density characteristic that faithfully reproduces the input gradation value can be provided.

  Furthermore, as described above, when the print engine characteristics fluctuate toward a higher solid density, the halftone gamma table indicates that the target output density for the input tone value is low in the low tone value region. It is particularly effective to perform correction so as to have a linear characteristic IDEA that connects the white density and the solid density in the gradation value region having the gradation region characteristic 40 and higher than the low gradation value region. Such correction can suppress an increase in the minimum ink output density when the ink first adheres.

  In addition, density correction is performed by measuring the output density of the formed patch image with a colorimeter, and manually correcting the printer driver characteristics, for example, based on the output density characteristics with respect to the input gradation values obtained thereby. The present embodiment can also be applied when doing so.

IDEA: ideal linear characteristic 40: low gradation characteristic CGT: correction gamma table

Claims (8)

In an image processing apparatus for generating image reproduction data indicating an ink adhesion region from image data having an input gradation value,
A controller having a halftone processing unit for generating the image reproduction data from the input gradation value based on a predetermined gamma characteristic;
The controller generates the image reproduction data from the patch image data by the halftone processing unit, causes the print engine to form a patch image based on the image reproduction data, and inputs the input gradation value of the patch image data and the patch A correction gamma table is generated based on the input gradation / output density characteristic indicating the relationship with the output density of the image so that the output density characteristic with respect to the input gradation value becomes the first characteristic,
The halftone processing unit generates image reproduction data from the input gradation value based on the correction gamma table,
In the first characteristic, the input tone value area among higher than a predetermined input tone values, has a linear characteristic connecting between the minimum output density and the maximum output density linearly, the minimum input tone value or more The output density corresponding to the minimum input gradation value is the output density of the linear characteristic in the low input gradation value area below the predetermined input gradation value , and the minimum ink output density corresponding to the minimum ink adhesion area is An image processing apparatus having a low input gradation characteristic that is lower than an output density of a linear characteristic. In claim 1,
The image processing apparatus having a characteristic that the low input tone characteristic gradually approaches the output density of the linear characteristic from the minimum ink output density as the input tone value increases in the low input tone value region. In claim 2,
The low input gradation characteristic in accordance with the inclination connecting the output density corresponding to the paper white output density and the input tone value corresponding to the minimum input tone value, the input gradation value increases, the An image processing apparatus having characteristics that asymptotically approach the slope of linear characteristics. In claim 1,
The image processing apparatus according to claim 1, wherein the first characteristic has the linear characteristic even in a high input gradation value region that is higher than the middle input gradation value region and reaches a maximum input gradation value. In claim 1,
The image processing apparatus, wherein the image reproduction data forming the minimum ink adhesion region is image reproduction data corresponding to a first input gradation value that is one gradation value higher than the minimum input gradation value. In an image processing apparatus for generating image reproduction data indicating an ink adhesion region from image data having an input gradation value,
A controller having a halftone processing unit for generating the image reproduction data from the input gradation value based on a gamma table;
The gamma table has a linear characteristic that linearly connects the minimum output density and the maximum output density in the middle input tone value region where the output density property with respect to the input tone value is higher than a predetermined input tone value. The output density corresponding to the minimum input tone value becomes the output density of the linear characteristic in the low input tone value region that is greater than the minimum input tone value and less than the predetermined input tone value , and the minimum ink adhesion An image processing apparatus formed to have a low input tone characteristic in which a minimum ink output density corresponding to a region is lower than an output density of the linear characteristic. In an image processing method for generating image reproduction data indicating an ink adhesion region from image data having an input gradation value,
A halftone processing step of generating the image reproduction data from the input gradation value based on a predetermined gamma characteristic;
The halftone processing step generates the image reproduction data from the patch image data, causes the print engine to form a patch image based on the image reproduction data, and sets the input gradation value of the patch image data and the output density of the patch image. A correction step of generating a correction gamma table based on an input gradation / output density characteristic indicating a relationship between the input gradation value and the output density characteristic with respect to the input gradation value so as to be a first characteristic;
In the halftone processing step, image reproduction data is generated from the input gradation value based on the correction gamma table,
In the first characteristic, the input tone value area among higher than a predetermined input tone values, has a linear characteristic connecting between the minimum output density and the maximum output density linearly, the minimum input tone value or more The output density corresponding to the minimum input gradation value is the output density of the linear characteristic in the low input gradation value area below the predetermined input gradation value , and the minimum ink output density corresponding to the minimum ink adhesion area is An image processing method having a low input gradation characteristic that is lower than an output density of a linear characteristic. In an image processing method for generating image reproduction data indicating an ink adhesion region from image data having an input gradation value,
A halftone processing step of generating the image reproduction data from the input gradation value based on a gamma table;
The gamma table has a linear characteristic that linearly connects the minimum output density and the maximum output density in the middle input tone value region where the output density property with respect to the input tone value is higher than a predetermined input tone value. The output density corresponding to the minimum input tone value becomes the output density of the linear characteristic in the low input tone value region that is greater than the minimum input tone value and less than the predetermined input tone value , and the minimum ink adhesion An image processing method formed so as to have a low input gradation characteristic in which a minimum ink output density corresponding to a region is lower than an output density of the linear characteristic.