JP6464826B2 - Image processing device - Google Patents

Description

  The present invention relates to an image processing apparatus, and more particularly to halftoning.

  Conventionally, various methods such as a dither method and an error diffusion method are known as a halftone processing method for converting a gradation value into a value indicating ejection (on) and non-ejection (off) of ink dots of a printer. . Generally, in order to speed up halftone processing, a pipeline processor for halftone processing is employed.

  Patent Document 1 discloses a technique for improving the image quality performance of halftone processing by combining a dither method and an error diffusion method. In the error diffusion processing using the result of dither processing described in Patent Document 1, the diffusion error generated for other pixels that have been subjected to error diffusion processing in advance is added to the tone value of the pixel of interest, A diffusion error corresponding to the difference between the gradation value added with the diffusion error of another pixel and the maximum gradation value or the minimum gradation value is derived.

JP 2014-78980 A

However, as described in Patent Document 1, when adding a diffusion error that has occurred for other pixels that have been previously subjected to error diffusion processing to the gradation value of the pixel of interest, the error diffusion processing is performed in advance. The error diffusion process for the pixel of interest cannot be started until the diffusion error occurring for the applied pixel is determined. Therefore, a pipeline processor that executes halftone processing that combines the dither method and error diffusion method described in Patent Document 1 is focused on until a diffusion error that occurs for a pixel that has been subjected to error diffusion processing in advance is determined. A pipeline buffer that delays pixel error diffusion processing is required, which increases the number of clocks required for halftone processing.
An object of the present invention is to increase the speed of halftone processing that provides high image quality.

  An image processing apparatus for achieving the above object determines a threshold value according to a gradation value of a target pixel, and an added value obtained by adding a diffusion error generated in another pixel to the gradation value of the target pixel is the threshold value. Is greater than the threshold, the comparison unit determines that no dot is formed when the added value is smaller than the threshold value, and the comparison unit determines the threshold value and the threshold value for the target pixel. In a period for determining dot formation, an ON error corresponding to the difference between the added value and the maximum gradation value of the pixel of interest and an OFF corresponding to the difference between the addition value and the minimum gradation value of the pixel of interest An error calculation unit that performs calculation for each of the errors, and when the comparison unit determines to form a dot, the ON error is selected as the diffusion error that has occurred in the pixel of interest, and the comparison unit does not form a dot And decision And a selection unit for the off error is selected as the diffusion errors occurring in the target pixel when the.

  According to the present invention, error diffusion with respect to another pixel (subsequent pixel) is performed during a period in which processing for determining dot-on (forming a dot) or dot-off (not forming a dot) of a pixel of interest is performed. The diffusion error of the target pixel required in the initial processing step in the processing is derived in advance for both the dot-on and dot-off cases, and when the dot-on or dot-off is determined for the target pixel, Select one of the derived diffusion errors. Thereby, since the start of the error diffusion process for other pixels (subsequent pixels) can be accelerated, the halftone process can be speeded up.

  Note that the function of each means described in the claims is realized by hardware resources whose function is specified by the configuration itself, hardware resources whose function is specified by a program, or a combination thereof. The functions of these means are not limited to those realized by hardware resources that are physically independent of each other.

The block diagram concerning embodiment of this invention. The block diagram concerning embodiment of this invention. The flowchart concerning embodiment of this invention. The flowchart concerning embodiment of this invention. The graph which shows the data table concerning embodiment of this invention. The circuit diagram concerning the embodiment of the present invention.

Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. In addition, the same code | symbol is attached | subjected to the corresponding component in each figure, and the overlapping description is abbreviate | omitted.
1. Apparatus Configuration A printer 1 as an embodiment of the present invention shown in FIG. 1 is a serial inkjet printer that performs bidirectional printing. The printer 1 has a mechanism for transporting the print medium P by the paper feed motor 32, a mechanism for reciprocating the carriage 40 in the longitudinal direction of the platen by the carriage motor 31, and a print head 90 mounted on the carriage 40 to drive the ink. A mechanism for discharging dots and a control unit 10 are provided.

  A mechanism for reciprocating the carriage 40 in the longitudinal direction of the platen includes a guide that is laid in parallel with the platen axis and slidably holds the carriage 40, and a pulley that is driven by a carriage motor 31 over an endless belt. Etc.

  The carriage 40 is mounted with ink cartridges 41 to 46 that contain cyan ink C, magenta ink M, yellow ink Y, black ink K, light cyan ink Lc, and light magenta ink Lm, respectively, as color inks. Hereinafter, a data format represented by C, M, Y, K, Lc, and Lm gradations is referred to as a CMY format. In the print head 47 below the carriage 40, nozzle rows corresponding to the above-described color inks are formed. When ink cartridges 41 to 46 filled with these inks are mounted on the carriage 40 from above, ink can be supplied from each cartridge to the print head 47.

  Each color ink constitutes a color component for expressing a full color by area gradation. That is, by changing the ratio of the area occupied by the ink dots of C, M, Y, K, Lc, and Lm, a full color is expressed by the six color components of C, M, Y, K, Lc, and Lm. Since the print head 47 can only control for each lattice point whether ink dots are ejected (ON) or not (OFF), the halftone processing unit 16 (to be described later) uses a CMY format gradation. The value is converted into halftone data indicating on and off of the ink dots, and the print head 47 is driven based on the halftone data.

  The control unit 10 includes a CPU 13, an external storage 12 including a nonvolatile storage medium such as a flash memory and an EEPROM, a main storage 52 including a volatile storage medium, an input / output unit 11, and an image processing IC (Integrated Circuit) 19. Are connected to each other by a bus. The control unit 10 loads the program stored in the external storage 12 into the main storage 14 and executes it by the CPU 13 to execute image processing on the input image data to generate halftone data. Control overall operation.

  The input / output unit 11 includes a data input interface compliant with standards such as USB, Ethernet (registered trademark), IrDA, Bluetooth, and the like, and can input image data from a removable memory or an external device. In this embodiment, the image data input from the input / output unit 11 has gradation values of three color components of red (R), green (G), and blue (B) for each pixel.

  The external storage 12 stores various data tables such as a color conversion table 120, a dither mask 121, and an error diffusion threshold table 122, in addition to various programs such as an OS and a print control program. In the color conversion table 120, RGB gradation values and CMYKLcLm gradation values are stored in association with each other in advance. The dither mask 121 is used for halftone processing by a systematic dither method, and includes a plurality of dither threshold values. In the present embodiment, the dither mask 121 has a so-called blue noise characteristic. The error diffusion threshold value table 122 is a table in which a diffusion threshold value used to determine whether dots are turned on or off in error diffusion processing is stored, and details thereof will be described later.

  The image processing IC 19 is configured such that a program for realizing the functions of the color conversion unit 15, the halftone processing unit 16, the dot processing unit 17, and the discharge control unit 18 described below is described in a predetermined hardware description language. It is a designed integrated circuit.

  The color conversion unit 15 performs color conversion from RGB to CMYKLcLm by referring to the color conversion table 120. The image data after color conversion has 256 steps (0 to 255) of gradation values Dn for each color component in the CMY format.

The halftone processing unit 16 receives CMY format image data from the color conversion unit 15 and converts the image data into binary halftone data indicating on / off of the ink dots in units of pixels. Details of the halftone processing unit 16 will be described later.
The dot processing unit 17 inputs binary halftone data, converts the dot arrangement data into dot arrangement data indicating the arrangement of dots formed on the print medium, and converts the dot arrangement data into dot arrangement data in the dot ejection order. Output.
The ejection controller 18 inputs the dot ejection data, and controls the print head 47, the carriage motor 31 and the paper feed motor 32 based on the dot ejection data so that dots corresponding to the dot arrangement data are formed on the print medium. A control signal for driving is output.

Here, the configuration of the halftone processing unit 16 will be described in detail with reference to FIG.
The dither unit 61 functioning as the first processing unit, the second processing unit, and the third processing unit compares the dither threshold value with the gradation value Dn of the target pixel for each color component while referring to the dither mask 121, Output the comparison result. The output signal of the dither unit 61 indicating the comparison result is a binary value indicating a dither threshold value (dot on) or less than the dither threshold value (dot off), and is hereinafter referred to as a dither comparison value. The dither unit 61 performs this dither processing for each color component in parallel. The dither comparison values of all the color components are input to the first selector 62 and the second selector 67. In addition, the gradation values of all the color components input to the dither unit 61 are also input to the first selector 62 for use in error diffusion units 64, 65, and 66 described later.

  The first selector 62 functioning as a synthesis unit acquires dither comparison values and gradation values for all color components of the pixel of interest, and selects these output destinations for each color component. That is, the first selector 62 inputs a dither comparison value to the delay buffer 63 for the first color component that is not the target of error diffusion processing for the target pixel, and the second color component that is the target of error diffusion processing for the target pixel. Inputs the dither comparison value and the gradation value to the error diffusion units 64, 65, 66.

  The delay buffer 63 functioning as a combining unit is a delay circuit that holds dither comparison values for the other color components of the pixel of interest by the number of clocks necessary to complete the processing of the pixel of interest in the error diffusion units 64, 65, 66. Pipeline register). When error diffusion processing is not performed for all color components, it is not necessary to hold the dither comparison value in the delay buffer 63, and when error diffusion processing is performed for only one color component, five color components are used. The dither comparison value needs to be held in the delay buffer 63. Therefore, the number of signal lines connecting the first selector 62 and the delay buffer 63 can transmit dither comparison values for five color components that are one less than six, which is the number of color components that the printer 1 can handle. You only need to prepare as many. In this embodiment, the first selector 62 and the delay buffer 63 are connected by three signal lines that transmit binary dither comparison values for the three color components.

  The first error diffusion unit 64, the second error diffusion unit 65, and the third error diffusion unit 66 functioning as the second processing unit have the same configuration, and each of the one color component of the target pixel has a dither comparison value. The tone value of the target pixel is acquired from the first selector 62, and an error diffusion threshold value that is a threshold value used for determining dot on / off in the error diffusion method is determined based on the dither comparison value and the tone value of the target pixel. This is a pipeline processor that controls the ease with which dots are formed by the error diffusion method by changing. Each error diffusion unit includes an error addition unit 661, a threshold comparison unit 662, an on error calculation unit 663, an on error storage unit 664, a diffusion range coefficient table 665, an off error calculation unit 666, an off error storage unit 667, and a selection unit. The third selector 668 and the diffusion error storage unit 669 are provided. Error addition unit 661, threshold comparison unit 662, on error calculation unit 663, on error storage unit 664, diffusion range coefficient table 665, off error calculation unit 666, off error storage unit 667, third selector 668, diffusion error storage unit 669 Will be described in detail later.

  The printer 1 having the above hardware configuration drives the carriage motor 31 to reciprocate the print head 47 in the main scanning direction with respect to the print medium P, and drives the paper feed motor 32. Thus, the print medium P is moved in the sub-scanning direction. The control unit 10 drives the nozzle at a timing based on the dot ejection data in accordance with the movement (main scanning) in which the carriage 40 reciprocates or the paper feeding movement (sub-scanning) of the printing medium. Ink dots of each color are ejected. In this way, the printer 1 can print a color image on the print medium P.

2. Printing Process Next, a printing process in the printer 1 will be described with reference to FIG. The printing process here starts when the user performs a print instruction operation on desired image data using the operation panel 20 or the like. When the printing process is started, the CPU 13 first reads RGB image data to be printed into the main memory 14 via the input / output unit 11 (step S1).

  When the image data is input, the CPU 13 causes the color conversion unit 15 to perform color conversion (step S2). The color conversion unit 15 refers to the color conversion table 120 stored in the external storage 12 and converts image data in RGB format into CMY format for the image data. The CPU 13 may cause the color conversion unit 15 to perform color conversion after performing image processing such as resolution conversion processing or smoothing processing on image data in RGB format or CMY format, or may perform resolution conversion processing after color conversion. Or image processing such as smoothing processing may be performed.

  When the color conversion process is performed, the CPU 13 causes the halftone processing unit 42 to execute the halftone process (S3).

  When the halftone process is performed, the CPU 13 causes the dot processing unit 43 to execute the dot process (step S4). The dot processing unit 43 rearranges the dot arrangement data into the dot ejection data in a single main scanning unit according to the nozzle arrangement of the printer 1 and the paper feed amount.

  When the dot processing is performed, the CPU 13 causes the ejection control unit 18 to execute printing (step S5). The discharge controller 18 outputs a control signal based on the dot discharge data to the print head 47, the carriage motor 31, the paper feed motor 32, and the like to execute printing.

3. Details of Halftone Processing Here, the halftone processing (step S3) will be described in detail with reference to FIGS. As shown in FIG. 4, when starting halftone processing, the CPU 13 first sets a color component to be subjected to error diffusion processing for image data to be subjected to halftone processing (S31). Specifically, the CPU 13 sets the destination to which the first selector 62 outputs the dither comparison value and the gradation value for each color component. That is, the CPU 13 outputs the dither comparison value and tone value of the color component to be subjected to the error diffusion process to the error diffusion units 64, 65, and 66, and the dither comparison value of the color component not to be subjected to the error diffusion process is delayed. The first selector 62 is set so as to be output to the buffer 63.

  A halftone process in which a dither process and an error diffusion process are combined is performed on a color component to be subjected to the error diffusion process. In addition, only the dither processing is performed as the halftone processing for the color components that are not subjected to the error diffusion processing. The output of the dither unit 61 is input not only to the first selector 62 but also to the second selector 67 for outputting the final halftone processing result. Accordingly, it is possible to perform only dither processing as high-speed halftone processing for all color components without using the error diffusion units 64, 65, 66 and the delay buffer 63. When only dither processing is performed as halftone processing for all color components, the CPU 13 may set the second selector 67 so that the dither comparison value output from the first selector 62 is selected and output.

  Which color component is to be subjected to error diffusion processing is an arbitrary design matter in the present invention. For example, when the printing speed is more important than the printing quality, only the dithering process may be performed as a halftone process for all color components. In addition, when most of the image data colors are concentrated in a specific color gamut, such as a monotone image, the color components close to the color gamut where the colors are concentrated are printed as error diffusion processing targets. Quality may be increased.

  Note that the units for setting the color component that is the target of error diffusion processing may be set, for example, for each image data, or may be set for each page. It can be set for each file, for each band, for each start-up, or set once by the user during setup after shipment. Alternatively, it may be set only once during factory assembly. That is, the process in step S31 may be performed before the printing process shown in FIG. Even in the case of setting only once at the time of factory assembly, by providing the first selector 62, the number of error diffusion units can be easily changed according to the design specifications, so that the design cost can be reduced.

  After setting the color components to be subjected to error diffusion processing, when the CPU 13 instructs the halftone processing unit 16 to execute halftone processing on the CMY format image data output from the color conversion unit 15, the dither unit 61 obtains the coordinates n (x, y) indicating the position of the target pixel and the gradation value Dn of the target pixel for the CMY format image data (step S31).

  When the coordinate n (x, y) of the target pixel and the gradation value Dn are acquired, the dither unit 61 performs dither processing (step S33). Here, the dither processing includes the threshold value THn_d corresponding to the coordinate n (x, y) of the target pixel and the gradation value Dn among the plurality of threshold values constituting the dither mask 121 stored in the external storage 12. This is a process for comparing the magnitude relationship with. This process is the same as the process for determining dot on / off using the dither method that is normally performed. However, the dither processing of the present embodiment is provisional dither processing in that there is a possibility that dot on / off may be finally determined by an error diffusion method to be described later. It is also a process for determining (error diffusion threshold).

  When ON or OFF as a dither comparison value is output from the dither unit 61 for each pixel and each color component, the first selector 62 selects an output destination for each color component based on the setting in step S31 and performs dither. The comparison value and the gradation value are output (step S34). That is, when the first selector 62 performs halftone processing different for each color component, the output destination of the dither comparison value and the gradation value Dn is set to the error diffusion unit for the second color component that is the target of the error diffusion processing. 64, 65, or 66, and for the first color component that is not subject to error diffusion processing, the output destination of the dither comparison value is the delay buffer 63.

  When the delay buffer 63 obtains the dither comparison value of the color component that is not subjected to the error diffusion process, the delay buffer 63 holds the dither comparison value by the number of clocks corresponding to the number of pipeline steps necessary for the error diffusion process described below (step S35). ).

  When the error diffusion units 64, 65, and 66 obtain the dither comparison value and the gradation value Dn of the target color component for the error diffusion processing, the error diffusion processing is executed by pipeline processing. Since a plurality of processes are executed simultaneously by pipeline processing, they are not actually executed in the order shown in FIG. 4, but in order to make it easy to understand the procedure of error diffusion processing applied to one target pixel. After describing the error diffusion processing along the flow described in FIG. 4, the actual processing sequence will be described with reference to FIG.

  First, the threshold comparison unit 662 determines whether the dither comparison value determined by the dither processing in step S33 is dot-on or dot-off (step S36), and if it is dot-on, the threshold THe used for error diffusion processing. Is set to the lower threshold THe_L corresponding to the gradation value Dn (step S37). On the other hand, if the dither comparison value is dot off, the threshold value THe used for the error diffusion process is set to the high threshold value THe_H corresponding to the gradation value Dn (step S38). As described above, in this embodiment, the threshold value THe used in the error diffusion method is changed according to the dither processing result and the gradation value Dn. The threshold THe is set with reference to the error diffusion threshold table 122 stored in the external storage 12.

  A specific example of the error diffusion threshold table 122 is conceptually shown in FIG. As shown in the figure, in the error diffusion threshold table 122, the gradation value Dn (here, 0 to 255) is associated with the lower threshold THe_L and the higher threshold THe_H. The threshold value THe_N shown in the figure is displayed as an example of a threshold value used in a normal error diffusion method. In this example, the threshold value THe_N becomes a constant value 127.5 regardless of the gradation value Dn. ing.

  In the example shown in FIG. 5, the high threshold THe_H takes a value of 127.5 when the gradation value Dn is 0, and increases as the gradation value Dn increases from 0. Finally, the gradation value Dn When the value is 255, the value is 207.5. The low threshold value THe_L takes the same value 127.5 as the high threshold value THe_H when the gradation value Dn is 0, and decreases as the gradation value Dn increases from 0, and when the gradation value Dn is 255. The value is 47.5.

  In other words, the difference between the high threshold THe_H and the low threshold THe_L (hereinafter also referred to as threshold difference ΔTHe) is 0 when the gradation value Dn is 0, and increases as the gradation value Dn increases. When the gradation value Dn is 255, the value is 80. That is, in the example of FIG. 5, the error diffusion threshold value table 122 indicates that the threshold value difference ΔTHe increases as the gradation value increases over the entire range of the gradation value Dn of the target pixel (here, the relationship is in a directly proportional relationship). ) Is set. Thus, the magnitude relationship between the high threshold THe_H and the low threshold THe_L is set such that the mutual values are equal or the high threshold THe_H is relatively larger than the low threshold THe_L. However, the case where the high threshold THe_H and the low threshold THe_L are the same in all gradation values is excluded. Note that the values of the high threshold THe_H and the low threshold THe_L for each gradation value can be set in any way as long as such a magnitude relationship is maintained. In this embodiment, the high threshold value THe_H and the low threshold value THe_L corresponding to the gradation value Dn are obtained by referring to the table, but these may be obtained by a function.

  Next, the error addition unit 661 adds the diffusion error Edn of other pixels stored in the diffusion error storage unit 669 to the gradation value Dn, and outputs a gradation value Dn + (step S39). Here, the diffusion error Edn is determined in step S44, which will be described later, for the preceding pixel in the vicinity of the target pixel, and the content thereof will be described later. The gradation value Dn + is a gradation value as a result of adding the diffusion error Edn generated in the preceding pixel to the gradation value Dn of the target pixel.

On the other hand, the on error calculation unit 663 and the off error calculation unit 666 estimate the binarization error and the diffusion error generated in the target pixel, respectively (step S40). The binarization error En is a difference between the gradation value Dn + obtained by adding the diffusion error Edn generated in the preceding pixel and the dot on / off result (in this case, the gradation value 255 or 0). The diffusion error Edn is an error added to the gradation value Dn of another pixel (subsequent pixel) in step S36. In this embodiment, the binarization error En is set to 7/16 for the pixel adjacent to the right of the pixel of interest, which is a peripheral pixel for which dot on / off has not been determined by the threshold comparison unit 662, and for the pixel on the lower left. The diffusion error Edn is distributed at a ratio of 3/16, 5/16 for the lower pixel, and 1/16 for the lower right pixel. These coefficients are stored in the diffusion range coefficient table 665. That is, the on error calculation unit 663 and the off error calculation unit 666 add the binarization error En, which is the difference between the gradation value Dn + and the gradation value 255 or 0, as in the following equations (1) to (4). The diffusion error Edn is derived for each neighboring pixel of the pixel of interest by multiplying the coefficients stored in the diffusion range coefficient table 665. The diffusion range coefficient table 665 is not limited to this, and different numbers may be used, and the diffusion range may be widened or narrowed.
E1n = En × 7/16 (1)
E2n = En × 3/16 (2)
E3n = En × 5/16 (3)
E4n = En × 1/16 (4)

  The threshold value comparison unit 662 that has set the error diffusion threshold value compares the gradation value Dn + to which the diffusion error Edn generated in the preceding pixel is added with the threshold value THe set in step S38 or step S39 (step S41). As a result, if the gradation value Dn + is equal to or greater than the threshold value THe, the dot of the target pixel is determined to be on (step S42), and if the gradation value Dn + is less than the threshold value THe, the dot of the target pixel is determined to be off. (Step S43).

  When the threshold comparison unit 662 determines dot on or dot off for the pixel of interest, the third selector 668 is set according to the determination, and the diffusion error Edn according to the determination is output from the third selector 668 (step S44). . That is, when the threshold comparison unit 662 determines dot on, the diffusion error Edn held by the on error storage unit 664 is output from the third selector 668. When the threshold comparison unit 662 determines dot off, the diffusion error Edn held by the off error storage unit 666 is output from the third selector 668. The diffusion error Edn determined in this way is stored in the diffusion error storage unit 669.

  The processing in steps S36 to S44 described above is halftone processing by the error diffusion method, and is executed by the error diffusion units 64, 65, and 66 by pipeline processing for each color component. Since the error diffusion method is a well-known technique, a detailed description thereof is omitted. As described above, while adding the quantization error of the gradation value of the target pixel to the surrounding pixels at a predetermined distribution ratio, This is a method of comparing a gradation value of a pixel with a predetermined threshold value and quantizing the gradation value.

  When the error diffusion processing by the error diffusion units 64, 65, and 66 is completed for the target pixel, the second selector 67 performs the dot on or dot off stored in the delay buffer 63 in step S35, and the third selector 668 in step S44. The output dot on or dot off is output as a result of halftone processing of one target pixel (S45). That is, the second selector 67 synthesizes and outputs, for each pixel, the result of the halftone process of the color component that is the object of the error diffusion process and the result of the halftone process of the color component that is not the object of the error diffusion process. To do. Here, as already described, the delay buffer 63 is configured to hold dot on or dot off by the number of clocks corresponding to the number of clocks (the number of pipeline steps) of the error diffusion units 64, 65, 66. The second selector 67 receives ON / OFF of dots of each color component of one target pixel simultaneously, and the second selector 67 outputs ON / OFF of dots of each color component of one target pixel simultaneously. As a result, halftone data indicating dot on / off for each pixel is output from the halftone processing unit 16.

  Here, the halftone processing unit 16 may output in parallel the signals indicating ON or OFF of the dots of the respective color components output from the second selector 67, or may collectively output the signals for each pixel. In the case of parallel wiring, ON or OFF of all color component dots is simultaneously output for each pixel in parallel. In the case of serial wiring, on / off of dots of all color components are collected in series and output for each pixel.

  Then, the halftone processing unit 16 repeats the processes of steps S32 to S45 with all the pixels as the target pixel sequentially (step S46). When the processes in steps S32 to S44 are finished for all the pixels, the halftone process in step S3 is finished.

  Since the output of the dither unit 61 is also directly input to the second selector 67, when the error diffusion processing is not performed on all the color components, the above-described processing of step S34 to step S44 is executed. Instead, the output of the dither unit 61 is directly output from the second selector 67 as a result of the halftone process. For this reason, when error diffusion processing is not performed on all color components, high-speed halftone processing is possible.

  Next, the principle of halftone processing combining the above-described dither method and error diffusion method will be described below. As described above, in the processing of steps S37 to S39, if the tone value Dn of the target pixel is equal to or larger than the threshold value THn_d of the dither processing, the threshold value THe used for the error diffusion method is set to the lower threshold value THe_L. If the tone value Dn is less than the threshold value THn_d for dither processing, the threshold value THe is set to the high threshold value THe_H. The threshold difference ΔTHe (= THe_H−THe_L) is a value of 0 or more.

  Here, a case where the threshold difference ΔTHe is 0 (THe_H = THe_L) is considered. In this case (in this embodiment, when the gradation value Dn = 0), the result of the dithering process does not affect the threshold value THe, so the process of step S33 is performed by the error diffusion method (steps S36 to S44). It will have no meaning for the final dot on / off decision. This means that in the halftone process of step S3, the final dot on or off is determined only by the error diffusion method.

  Next, consider a case where the threshold difference ΔTHe is greater than 0 (THe_H> THe_L). In this case (in this embodiment, when the gradation value Dn ≠ 0), when the CPU 13 determines that the dot is on by the dither process (when the gradation value Dn is equal to or greater than the threshold value THn_d), the CPU 13 relatively sets the threshold value THe. A small low threshold THe_L is set. On the other hand, when it is determined that the dot is off by the dither processing (when the gradation value of the gradation value Dn is less than the threshold value THn_d), the threshold value THe is set to a relatively large high threshold value THe_H. That is, the CPU 13 performs control so that the dot is likely to be turned on by the error diffusion method when it is determined that the dot is on by the dither process, and if the dot is easily turned off by the error diffusion method, the CPU 13 is likely to be turned off by the error diffusion method. Control. This is because the final dot ON / OFF determination result by the error diffusion method is determined as the dot ON / OFF determination by the dither processing as the gradation value Dn is larger than when the threshold difference ΔTHe is 0. It means approaching the result. That is, the final dot ON / OFF is determined by combining the elements of the error diffusion method and the dither method.

  As the threshold difference ΔTHe increases, the dither method becomes dominant. When the threshold difference ΔTHe becomes infinite, only the on / off state of the dot is determined by the dither method alone. . When the threshold difference ΔTHe is infinite, if it is determined that the dot is turned on by the dithering process, it is always determined that the dot is turned on by the subsequent error diffusion method, and if the dot is turned off by the dithering process, the subsequent error diffusion is performed. This is because the dot is always determined to be off by law.

  In short, by changing the threshold value THe of the error diffusion process according to the result of the dither process, specifically, by changing the magnitude of the threshold difference ΔTHe, the elements of the dither method and the error diffusion method in the halftone process It is possible to control the degree of contribution to each element. In this embodiment, by utilizing such a principle, the dither method element and the error diffusion method element in the halftone process are dynamically controlled according to the gradation value of the gradation value Dn.

  In the error diffusion method, the dot is more likely to be turned off as the threshold value THe is larger. Therefore, by controlling the threshold value THe according to the dither processing result, the degree of control of the ease of dot formation by the error diffusion method It can also be understood as controlling.

  Here, the order of the error diffusion processing in the error diffusion units 64, 65, 66 will be described in detail with reference to FIG. The diffusion error occurring in the target pixel is finally detected as a result of the threshold value comparison unit 662 comparing the tone value Dn + of the target pixel added with the diffusion error Edn generated in the preceding pixel with the threshold value THe. If off is not determined, it cannot be calculated. That is, the diffusion error occurring in the target pixel cannot be calculated until the diffusion error Edn occurring in the preceding pixel is determined.

  The diffusion error generated in the target pixel is added to the gradation value of the subsequent pixel in the error adding unit 661. If the ON / OFF of the dot is determined for the target pixel and the diffusion error generated in the target pixel is not calculated, Processing of subsequent pixels will be delayed. That is, if the threshold error comparing unit 662 waits until the dot on / off is determined and then derives the diffusion error occurring in the target pixel, the number of pipeline buffer clocks for waiting for the dot on / off determination increases. As a result, the number of clocks required for the entire error diffusion process increases. If the number of clocks required for error diffusion processing increases, the number of clocks for which the delay buffer 63 holds dither comparison values for color components that are not subject to error diffusion processing must also be increased, which is necessary for the entire halftone processing. The number of clocks increases. Therefore, by calculating the diffusion error occurring in the pixel of interest at an early stage, it is possible to advance the error diffusion processing for the subsequent pixels quickly.

  Therefore, the on error calculation unit 663 derives in advance a diffusion error that occurs in the target pixel when the threshold comparison unit 662 determines dot on before the threshold comparison unit 662 determines dot on or dot off of the target pixel. , And stored in the ON error storage unit 664. Further, the off error calculation unit 666 derives in advance a diffusion error that occurs in the target pixel when the threshold comparison unit 662 determines dot off before the threshold comparison unit 662 determines dot on or dot off of the target pixel. And stored in the OFF error storage section 667. In this way, during the period in which the threshold comparison unit 662 performs the process for determining whether the pixel of interest is dot on or dot off, the diffusion error is estimated in advance and held in advance, so that the subsequent pixels can be stored. The start of error diffusion processing can be accelerated. That is, the number of pipeline steps of the error diffusion units 64, 65, 66 can be reduced.

  In this way, the printer 1 performs a diffusion error that occurs in the target pixel when the threshold comparison unit 662 determines dot on during the period in which the process for determining dot on or dot off of the target pixel is being performed, By deriving the diffusion error that occurs in the pixel of interest when the threshold comparison unit 662 determines dot-off, the number of pipeline steps for error diffusion processing is reduced. As a result, the number of clocks of the pipeline buffer that holds the dither processing result for the color component not subjected to the error diffusion processing can be reduced. As a result, the number of pipeline steps required for halftone processing performed only for dither processing for each color component or in combination with dither processing and error diffusion processing is reduced, and the halftone processing can be speeded up.

  Since the printer 1 includes a delay buffer that can hold dither comparison values of color components that cannot be processed by the error diffusion unit, instead of including error diffusion units corresponding to the number of colors of ink, halftone processing Therefore, the circuit scale can be suppressed. In addition, since the printer 1 can select the output destination of the first selector 62 for each color component, the printer 1 requests whether to perform halftone processing only of dither processing or halftone processing combining dither processing and error diffusion processing. Each color component can be selected according to printing performance and specifications. Furthermore, since the dither comparison value, which is the result of the dither processing, can be held in the delay buffer 63 by the number of clocks necessary for performing the error diffusion processing by the error diffusion units 64, 65, 66, halftone of one pixel The processing results can be simultaneously input to the second selector 67, and the halftone processing results can be synthesized and output for each pixel.

  Further, the printer 1 can set the contribution degree between the dither method and the error diffusion method in the halftone process to a desired level by appropriately setting the threshold difference ΔTHe in the halftone process. Further, since the threshold THe is changed based on the result of the temporary dither process, the ease of dot formation by the error diffusion method is controlled, so the configuration is simple and contributes to the speeding up of the process.

  Further, the printer 1 changes the threshold value THe based on the gradation value of the gradation value Dn. Specifically, the printer 1 changes the threshold difference ΔTHe to change the dot density by the error diffusion method. Since the ease of formation is controlled, the degree of contribution between the dither element and the error diffusion element can be changed according to the gradation value of the image data. In addition, the contribution can be changed for each arbitrary region of the image data. As a result, it is possible to generate dot data with a good contribution according to the image data and the characteristics of the printing apparatus, and to improve the print image quality.

  In this embodiment, ON / OFF determination of dots by error diffusion processing is controlled by the magnitude of the threshold difference ΔTHe, and the threshold difference ΔTHe is set so as to increase toward the higher gradation side. In the error diffusion process, when it is determined that the dot is turned on by the dithering process, a lower threshold ΔTHe_L that sets the dot to be easily turned on by the error diffusion process is set as the gradation value Dn is larger. The higher threshold value ΔTHe_H is set that the dot is likely to be turned off by the error diffusion method as the tone value Dn is larger. In other words, the printer 1 determines the error diffusion threshold value THe that causes the error diffusion processing result to approach the dither processing result for high gradation image data than for low gradation image data. On the key side, the merit of the error diffusion method element can be obtained, while on the high gradation side, the merit of the dither method element can be obtained.

  As an advantage of the error diffusion method on the low gradation side, for example, it is possible to obtain a granularity with good print image quality. If a diffusion range switching error diffusion method or the like is used as the error diffusion method, further improvement in image quality can be expected. Since the diffusion range switching error diffusion method is a known technique, a detailed description is omitted. However, the error diffusion range is switched according to the combination of the input gradation value and the binarization result. By diffusing the error over a wide range only when the dots are turned on, it is possible to improve the granularity of the low gradation region and suppress the occurrence of undesired continuity of dots, so-called worms.

  As an advantage of the dither method on the high gradation side, for example, it is possible to suppress deterioration in image quality due to deviation of dot landing positions. This merit is due to the above-described predetermined dot formation characteristic of the dither mask 121. In the high gradation region, even if dot data is generated by the dither method, the graininess of the print image quality is not conspicuous due to ink bleeding, and this does not cause a big problem.

  In addition, the error diffusion threshold value table 122 is set so that the threshold difference ΔTHe changes stepwise based on the magnitude of the gradation value Dn. That is, the printer 1 gradually changes the degree of control of the ease of dot formation by the error diffusion method based on the magnitude of the gradation value. Therefore, it is possible to smoothly change the contributions of the dithering element and the error diffusion element according to the data gradation value. As a result, the change in the contribution between the dithering and error diffusion elements is difficult to see in the printed result, so the change in the contribution of the dithering and error diffusion elements in the same print image The accompanying deterioration in print image quality can be suppressed.

  In particular, in the present embodiment, since the dither mask 121 has a blue noise characteristic that is excellent in the granularity of the print image quality, the error diffusion method element that is also excellent in the granularity of the print image quality is used. The change in contribution can be seen more smoothly. Even if the dither mask 121 does not have a blue noise characteristic, the degree of control of the ease of dot formation by the error diffusion method is changed stepwise based on the magnitude of the gradation value. By doing so, it is possible to smoothly change the contribution of the dithering and error diffusion elements according to the data gradation value.

4). Other Embodiments The technical scope of the present invention is not limited to the above-described embodiments, and it goes without saying that various modifications can be made without departing from the scope of the present invention. For example, if the threshold comparison unit 662 determines dot on during the period from when the gradation value Dn + is input to the threshold comparison unit 662 until the threshold comparison unit 662 determines whether to turn on / off the dot, the pixel of interest is Since it is only necessary to derive the diffusion error that occurs and the diffusion error that occurs in the pixel of interest when the threshold comparison unit 662 determines dot off, the on error storage unit 664 and the off error storage unit 667 are not necessarily required, and the on error storage is not necessary. Instead of the unit 664 and the off-error storage unit 667, or in addition to the on-error storage unit 664 and the off-error storage unit 667, a buffer that holds the gradation value Dn or the gradation value Dn + output from the error addition unit 661 ( (Pipeline register) may be provided.

  In the above-described embodiments, halftone processing for determining only dot on / off has been described. However, multi-value processing for determining dot on / off for each of a plurality of types of dots having different sizes is performed as a dither unit. 61, may be performed by the error diffusion units 64, 65, 66.

  In the above-described embodiment, halftone processing for determining whether one dot is turned on or off for each pixel has been described. However, a plurality of dots may be turned on or off for each pixel. That is, the present invention can be applied to a configuration in which one pixel of image data corresponds to a plurality of dots of the printer.

  In the above-described embodiment, the dither method is exemplified as the halftone processing method applied to the first processing unit, but another halftone processing method that is neither the dither method nor the error diffusion method may be applied to the first processing unit. good. In the above-described embodiment, an example in which the first processing unit and the second processing unit share the dither unit 61 has been described. However, the dither unit 61 is the first processing unit, and the error diffusion units 64, 65, and 66 are the first processing unit. The two processing units may be configured to alternatively apply the dither method or the error diffusion method for each color component.

DESCRIPTION OF SYMBOLS 1 ... Printer, 10 ... Control unit, 11 ... Input / output part, 12 ... External storage, 14 ... Main memory, 15 ... Color conversion part, 16 ... Halftone processing part, 17 ... Dot processing part, 18 ... Discharge control part, DESCRIPTION OF SYMBOLS 20 ... Operation panel, 31 ... Carriage motor, 32 ... Paper feed motor, 40 ... Carriage, 41-46 ... Ink cartridge, 42 ... Halftone processing part, 43 ... Dot processing part, 47 ... Print head, 52 ... Main memory, 61: Dither unit, 62: First selector, 63: Delay buffer, 64, 65, 66 ... Error diffusion unit, 64 ... Error diffusion unit, 64 ... First error diffusion unit, 64, 65 ... Error diffusion unit, 64, 65, 66 ... error diffusion unit, 65 ... second error diffusion unit, 66 ... third error diffusion unit, 67 ... second selector, 90 ... print head, 1 DESCRIPTION OF SYMBOLS 0 ... Color conversion table, 121 ... Dither mask, 122 ... Error diffusion threshold value table, 661 ... Error addition unit, 662 ... Threshold comparison unit, 663 ... On error calculation unit, 664 ... On error storage unit, 665 ... Diffusion range coefficient table 665 ... Diffusion range coefficient table, 666 ... Off error calculation unit, 667 ... Off error storage unit, 668 ... Third selector, 669 ... Diffusion error storage unit, 19 ... Image processing IC, P ... Print medium

Claims (3)

A threshold value is determined in accordance with the tone value of the target pixel, and if an added value obtained by adding a diffusion error generated in another pixel to the tone value of the target pixel is larger than the threshold value, it is determined to form a dot. And a comparison unit that determines not to form dots when the added value is smaller than the threshold value;
An ON error that is a diffusion error according to a difference between the added value and the maximum gradation value of the target pixel in a period in which the comparison unit determines the threshold value and the dot formation for the target pixel; and the target pixel An error calculation unit that performs calculation for each of the off errors that are diffusion errors according to the difference between the added value and the minimum gradation value;
When the comparison unit determines to form a dot, the on error is selected as the diffusion error generated in the target pixel, and when the comparison unit determines not to form a dot, the off error is set to the target pixel. A selection unit for selecting the generated diffusion error;
An image processing apparatus comprising: The comparison unit controls the easiness of dot formation by changing the threshold according to the result of dithering the tone value of the pixel of interest and the tone value of the pixel of interest.
The image processing apparatus according to claim 1. When the result of the dither processing is a value for forming a dot, the comparison unit changes the threshold so that the comparison unit can easily determine to form the dot, and the result of the dither processing is a dot. If the value does not form the dot, the threshold value is changed so that the decision not to form the dot is easily made.
The image processing apparatus according to claim 2.

Images