JP2006173929A - Image processing method, program, image processing apparatus, and image forming apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To suppress rising of a cost for a recording head and to increase recording speed by solving the problem of irregular image such as white streak of a recorded image which is caused by failure in jetting, so that the white streaks and irregular image are hard to be recognized by human eyes even if the failure in jetting occurs. <P>SOLUTION: In an error dispersion process, an input value of an interested pixel is added with an error value of a peripheral pixel that has been processed according to the ratio of weight matrix, to calculate a correction pixel value Dxy. A non-jetting nozzle managing table is referenced, and it is discriminated whether the interested pixel position is at the position recorded by the non-jetting nozzle or not. If it is not non-jetting nozzle from the result of discrimination, a normal error dispersion process is continued. If it is the non-jetting nozzle a corrective process which forces dot-OFF is performed to process the entire correction pixel value Dxy as error values even if "threshold value is equal to or less than correction pixel value Dxy". Dot conversion is performed with peripheral pixels so that missing of dot due to non-jetting is reflected on peripheral pixels in the form of error value in such manner as correcting a missing part. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to an image processing method, a program, an image processing apparatus, and an image forming apparatus.

  2. Related Art As an image forming apparatus such as a printer, a facsimile machine, a copying machine, or a multifunction machine of these, for example, an ink jet recording apparatus using a droplet discharge head as a recording head is known. An ink jet recording apparatus is a paper (not limited to paper, but includes OHP, etc., and means that ink droplets, other liquids, etc. can adhere to the recording medium, recording medium, recording paper, etc. In other words, the recording liquid is ejected with ink to form an image (recording, printing, printing, and printing are also used synonymously).

  Such an ink jet recording apparatus, in particular, increases the image quality of color images using inks of a plurality of colors, increases the droplet discharge driving frequency of the head, and increases the speed by increasing the number of nozzles arranged in the recording head. Is progressing.

  However, in an inkjet head that is a liquid ejection head, due to dust that has entered the nozzles of the recording head at the time of manufacture, deterioration of nozzles due to long-term use, deterioration of elements for ejecting ink, etc., There is a case where a so-called “non-ejection” ink droplet cannot be ejected (such a nozzle is called “non-ejection nozzle”). In the case of deterioration of an element for ejecting ink, there is a possibility that non-ejection may occur accidentally especially during the use period of the recording apparatus.

  In addition, the ink droplet ejection direction is greatly deviated from the desired direction (hereinafter also referred to as “ejection variation”) or the ink droplet ejection amount is a desired amount without being in a completely non-ejection state. In some cases, the state became significantly different (hereinafter also referred to as “variation in droplet amount”). As described above, the nozzles deteriorated so as to greatly reduce the quality of the recorded image when used for recording are not equivalent to nozzles for recording, and will be described as the same treatment as “non-ejection nozzles”.

  Such non-ejection and the like can be suppressed by improving the manufacturing environment and the like, and has not been a major problem in the past. However, as described above, the non-ejection is arranged on the recording head for speeding up. When the number of nozzles is increased, it becomes a problem that cannot be ignored. In particular, in order to manufacture a recording head that does not include a nozzle in a non-ejection state or a good recording head in which non-ejection is unlikely to occur, the manufacturing cost increases, and as a result, the recording head becomes expensive.

  However, when these non-ejection nozzles occur, defects such as white stripes occur on the image.

Therefore, conventionally, as described in Patent Documents 1 and 2, in order to complement such white streaks due to such non-ejection nozzles, a multi-scan recording method that performs recording by scanning the recording head a plurality of times is used. There is one in which white streaks are recorded by complementing with other normal nozzles.
JP-A-5-309874 JP 2001-630008 A

  However, in order to achieve the high-speed recording as described above, it is preferable to perform so-called one-pass printing in which printing is completed by one scan. However, in this one-pass printing, recording is not performed due to non-ejection. It is very difficult to complement or make it inconspicuous.

  Also in the multi-scan recording method in which a recording head scans a predetermined area on a recording medium a plurality of times and performs recording, the position is complementarily recorded depending on the position and number of nozzles in which ejection failure has occurred. It can be difficult. In particular, in the recording mode where the number of passes of the multi-scan is relatively small, the nozzles that can be used for replacement are limited. May not be available.

Therefore, as described in Patent Documents 3 to 7, for the highlight portion image data having a low density, the density of the pixels placed near the position where the non-ejection nozzle is supposed to be printed is increased. There is a method of replacing the shadow portion image data in which the omission is not conspicuous and the density cannot be increased any more, with dots of different colors having close brightness.
Japanese Patent Laid-Open No. 2002-19101 JP 2003-136702 A JP 2003-136663 A JP 2003-136664 A JP 2003-205604 A

  However, even with the correction methods described in Patent Documents 3 to 7, it is not always possible to deal with all cases. Even if the pixel density in the vicinity of the pixel corresponding to the non-ejection nozzle (non-ejection pixel) is increased, the output data that can increase the density does not always exist at a desired position, and the correction point is the pole of the non-ejection pixel. Since it is applied only to the vicinity, there is a problem that the density increase may lead to deterioration of the graininess.

  Recently, there is also a device that employs a multi-dot method for controlling the dot diameter. As a feature of such a device, only a part of the dots are not ejected, and the remaining dots can be ejected normally. Cases also exist. In the correction methods disclosed in Patent Documents 3 to 7, binary representation is presupposed to the last, and thus the case of the multi-dot method for controlling the dot diameter is not assumed. Of course, it is possible to eliminate nozzles with some non-ejecting dots as non-ejecting nozzles, but correction will be applied even to normal dot parts, so it will be an overcorrection that degrades the quality. There is a problem that there may be cases.

  The present invention has been made in view of the above-described problem, and eliminates image unevenness such as white streaks in a recorded image that occurs when dots are not recorded due to non-ejection, and even when non-ejection occurs, Provided are an image processing method, a program, an image processing apparatus, and an image forming apparatus that make it difficult for human eyes to recognize white stripes and image unevenness, suppress the increase in cost of the recording head, and increase the recording speed. The purpose is to do.

  In order to solve the above-described problem, the image processing method according to the present invention corrects a defect in a recorded image due to a nozzle that cannot perform normal recording in halftone processing in which multi-value data is converted into a dot pattern. It was set as the structure which performs.

  Here, error diffusion processing is used as halftone processing to compensate for defects in the recorded image, and the correction processing uses a recording sequence table that shows the correspondence between nozzles and pixel positions reflecting non-ejection nozzle data as an error in dot generation. It is preferable to reflect in the diffusion process. In this case, when the recording sequence table reflecting the non-ejection nozzle data is reflected in the error diffusion process as the dot generation condition, the recording position by the non-ejection nozzle is forcibly turned off while the quantization error itself is directly It is preferable to develop the pixel. The error diffusion process is a reduction error diffusion process corresponding to n values (n ≧ 2), and it is preferable to apply the correction process for each dot size that can be expressed by each nozzle.

  The program according to the present invention is configured to cause a computer to execute the image processing method according to the present invention.

  The image processing apparatus according to the present invention is configured to include the program according to the present invention.

  The image forming apparatus according to the present invention is configured to perform a correction process to compensate for defects in a recorded image caused by nozzles that cannot perform normal recording in halftone processing for converting multi-value data into a dot pattern. .

  Here, error diffusion processing is used as halftone processing to compensate for defects in the recorded image, and the correction processing uses a recording sequence table that shows the correspondence between nozzles and pixel positions reflecting non-ejection nozzle data as an error in dot generation. It is preferable to reflect in the diffusion process. In this case, when the recording sequence table reflecting the non-ejection nozzle data is reflected in the error diffusion process as the dot generation condition, the recording position by the non-ejection nozzle is forcibly turned off while the quantization error itself is directly It is preferable to develop the pixel. The error diffusion process is a reduction error diffusion process corresponding to n values (n ≧ 2), and it is preferable to apply the correction process for each dot size that can be expressed by each nozzle.

  According to the image processing method, the program, the image processing apparatus, and the image forming apparatus according to the present invention, the correction processing for complementing the defect of the recorded image caused by the nozzle that cannot perform normal recording is converted into the dot pattern. Therefore, correction processing can be applied directly to processing that converts multi-valued data to dots without causing a significant decrease in recording speed, such as correction means that uses multi-pass recording. Thus, it is possible to prevent application of overcorrection that causes the balance of dot arrangement to be lost. This eliminates image irregularities such as white streaks in the recorded image caused by non-ejection of dots that are not recorded, and even when non-ejection occurs, the human eyes recognize white streaks and image irregularities. This makes it difficult to reduce the cost of the recording head and to increase the recording speed.

  Embodiments of the present invention will be described below with reference to the accompanying drawings. First, an example of an ink jet recording apparatus as an image forming apparatus will be described with reference to FIGS. 1 is a schematic configuration diagram of the entire mechanism of the recording apparatus, FIG. 2 is an explanatory plan view of the main part of the recording apparatus, FIG. 3 is a perspective explanatory view illustrating the head configuration of the recording apparatus, and FIG. FIG. 3 is a schematic cross-sectional explanatory diagram of a conveyance belt of a recording apparatus.

  This ink jet recording apparatus has an image forming unit 2 and the like inside the apparatus main body 1, and a paper feed tray 4 on which a large number of recording media (hereinafter referred to as “paper”) 3 can be stacked below the apparatus main body 1. The sheet 3 fed from the sheet feeding tray 4 is taken in, and a required image is recorded by the image forming unit 2 while the sheet 3 is conveyed by the conveying mechanism 5, and then mounted on the side of the apparatus main body 1. The paper 3 is discharged to the paper discharge tray 6.

  In addition, the ink jet recording apparatus includes a duplex unit 7 that can be attached to and detached from the apparatus main body 1. When performing duplex printing, the sheet 3 is transported in the reverse direction by the transport mechanism 5 after completion of one-surface (front surface) printing. The sheet is taken into the duplex unit 7, reversed, and sent to the transport mechanism 5 again as the other side (back side) as a printable side, and the sheet 3 is discharged to the discharge tray 6 after the other side (back side) printing is completed.

  Here, the image forming unit 2 slidably holds the carriage 13 on the guide shafts 11 and 12, and moves the carriage 13 in a direction orthogonal to the conveyance direction of the paper 3 (main scanning) by a main scanning motor (not shown). . The carriage 13 is equipped with a recording head 14 composed of a droplet discharge head in which nozzle holes 14n (see FIG. 3) as a plurality of discharge ports for discharging droplets are arranged. The ink cartridge 15 for supplying the ink is detachably mounted. Note that a sub tank may be mounted in place of the ink cartridge 15 so that ink is replenished and supplied from the main tank to the sub tank.

  Here, as the recording head 14, for example, as shown in FIGS. 2 and 3, droplets that eject ink droplets of each color of yellow (Y), magenta (M), cyan (C), and black (Bk). Although the four independent inkjet heads 14y, 14m, 14c, and 14k, which are ejection heads, are used, one or a plurality of heads having a plurality of nozzle arrays that eject ink droplets of each color may be used. The number of colors and the order of arrangement are not limited to this.

  As the ink-jet head constituting the recording head 14, a piezoelectric actuator such as a piezoelectric element, a thermal actuator that utilizes a phase change caused by film boiling of a liquid using an electrothermal transducer such as a heating resistor, and a metal phase change caused by a temperature change. A shape memory alloy actuator to be used, an electrostatic actuator using an electrostatic force, or the like as an energy generating means for discharging ink can be used.

  The paper 3 in the paper feed tray 4 is separated one by one by a paper feed roller (half-moon roll) 21 and a separation pad (not shown), fed into the apparatus main body 1 and sent to the transport mechanism 5.

  The transport mechanism 5 guides the fed paper 3 along the guide surface 23 a and guides the paper 3 fed from the duplex unit 7 along the guide surface 23 b and the paper 3. A conveying roller 24 that conveys, a pressure roller 25 that presses the sheet 3 against the conveying roller 24, a guide member 26 that guides the sheet 3 toward the conveying roller 24, and a sheet 3 that is returned when duplex printing is performed on the duplex unit 7. And a pressing roller 28 that presses the paper 3 fed from the conveying roller 24.

  Further, the transport mechanism 5 charges the transport belt 33 between the drive roller 31 and the driven roller 32 and the transport belt 33 so that the recording head 14 transports the paper 3 while maintaining the flatness of the paper 3. A charging roller 34 that is opposed to the charging roller 34, a guide member 35 (not shown) that guides the conveying belt 33 at a portion facing the image forming unit 2, and a conveying belt 33. A cleaning roller made of a porous material or the like, which is a cleaning means for removing the recording liquid (ink) adhering to the recording medium.

  Here, the conveyance belt 33 is an endless belt, and is stretched between the driving roller 31 and the driven roller (tension roller) 32 so as to circulate in the direction indicated by the arrow in FIG. 1 (paper conveyance direction). It is composed.

  The transport belt 33 can be configured as a single layer, or as shown in FIG. 4, a two-layer configuration of a first layer (outermost layer) 33a and a second layer (back layer) 33b, or a configuration of three or more layers. For example, the transport belt 33 is a surface layer that is a sheet adsorbing surface formed of a pure resin material having a thickness of about 40 μm that is not subjected to resistance control, for example, ETFE pure material, and resistance control by carbon using the same material as the surface layer. It consists of the back layer (medium resistance layer, earth layer) performed.

  The charging roller 34 is disposed so as to come into contact with the surface layer of the transport belt 33 and to rotate following the rotation of the transport belt 33. A high voltage is applied to the charging roller 34 in a predetermined pattern from a high voltage circuit (high voltage power source) (not shown).

  Further, on the downstream side from the transport mechanism 5, a paper discharge roller 38 for sending the paper 3 on which an image is recorded to the paper discharge tray 6 is provided.

  In the image forming apparatus configured as described above, the conveyance belt 33 rotates in the direction of the arrow and is positively charged by coming into contact with the charging roller 34 to which a high potential applied voltage (AC bias voltage) is applied. In this case, by switching the polarity from the charging roller 34 at predetermined time intervals, the conveying belt 33 is charged while switching the polarity at a predetermined charging pitch.

  Here, when the paper 3 is fed onto the conveying belt 33 charged at this high potential, the inside of the paper 3 is in a polarized state, and the electric charge having the opposite polarity to the electric charge on the conveying belt 33 is connected to the belt 33 of the paper 3. The charge on the belt 33 and the charge on the transported paper 3 are electrostatically attracted to each other, and the paper 3 is electrostatically attracted to the transport belt 33. . In this way, the sheet 3 strongly adsorbed to the transport belt 33 is calibrated for warpage and unevenness, and a highly flat surface is formed.

  Therefore, the recording belt 14 is moved around the conveyor belt 33 and the recording head 14 is driven according to the image signal while moving and scanning the carriage 13 in one direction or in both directions, as shown in FIGS. As shown, a droplet 14i is ejected (jetted) from the recording head 14, and ink droplets, which are droplets, are landed on the stopped paper 3 to form dots Di, thereby recording one line. After the predetermined amount of paper 3 is conveyed, the next line is recorded. When the recording end signal or the signal that the rear end of the paper 3 reaches the recording area is received, the recording operation is ended. FIG. 5B is an enlarged view of the dot Di formation portion of FIG.

  In this way, the sheet 3 on which the image is recorded is discharged to the discharge tray 6 by the discharge roller 38.

  In the ink jet recording apparatus of this embodiment, the four-color head configuration is described, but the present invention is not limited to this. That is, for example, a 6-color head configuration as shown in FIGS. 6 and 7 or a 7-color head configuration as shown in FIGS. Of course, it is not limited to the color in each of these head configurations and the arrangement order thereof.

  Here, the head configuration in FIG. 6 is a head (independent head or one or a plurality of heads) that ejects droplets of each color of yellow (Y), magenta (M), cyan (C), and black (Bk). Meaning of independent nozzle rows) In addition to 14y, 14m, 14c and 14k, red (R) and blue (B) heads 14r and 14b are added.

  In the head configuration of FIG. 7, in addition to the heads 14y, 14m, 14c, and 14k that discharge ink droplets of yellow (Y), magenta (M), cyan (C), and black (Bk), the density is reduced. Heads 14 lc and 14 lm that discharge droplets of light cyan (LC) and light magenta (LM) are added.

  The head configuration of FIG. 8 is obtained by adding a head 14r that discharges red (R) color droplets to the six-color head configuration of FIG. Further, the head configuration in FIG. 9 is obtained by adding a head 14 dy for ejecting dark yellow (DY) droplets with reduced saturation to the six-color head configuration in FIG. 7.

Next, an outline of the control unit of the image forming apparatus will be described with reference to FIG. This figure is an overall block diagram of the control unit.
The control unit 100 includes a CPU 101 that controls the entire apparatus, a ROM 102 that stores programs executed by the CPU 101 and other fixed data, a RAM 103 that temporarily stores image data, and the like, while the apparatus is powered off. Also, a non-volatile memory (NVRAM) 104 for holding data and an ASIC 105 for processing image processing for performing various signal processing and rearrangement and other input / output signals for controlling the entire apparatus are provided.

  The control unit 100 also includes an I / F 106 for transmitting and receiving data and signals to and from the host 90 such as a personal computer including the image processing apparatus according to the present invention, and a drive for controlling the drive of the recording head 14. Drive waveform generation unit 107 and head driver 108 for generating waveforms, main scanning motor driving unit 111 for driving main scanning motor 110, sub-scanning motor driving unit 113 for driving sub-scanning motor 112, and charging An AC bias supply unit 114 that supplies an AC bias to the roller 34, an environmental sensor 118 that detects environmental temperature and / or environmental humidity, an I / O 116 that inputs detection signals from various sensors (not shown), and the like are provided. The control unit 100 is connected to an operation panel 117 for inputting and displaying information necessary for the apparatus.

  Here, the control unit 100 prints data including image data from the host 90 side such as a data processing device such as a personal computer, an image reading device such as an image scanner, an imaging device such as a digital camera, and the like. The data is received by the I / F 106 via a cable or a net. It should be noted that the print data generation output for the control unit 100 is performed by the printer driver 91 according to the present invention on the host 90 side.

  Then, the CPU 101 reads and analyzes the print data in the reception buffer included in the I / F 106, performs data rearrangement processing by the ASIC 105, and transfers the image data to the head drive control unit 107. Note that the conversion of print data for image output into bitmap data is performed by developing the image data into bitmap data by the printer driver 91 on the host 90 side and transferring it to this apparatus as described above. For example, font data may be stored in the ROM 102.

  The drive waveform generation unit 107 includes a D / A converter that performs D / A conversion on the drive pulse pattern data, and includes a single drive pulse (drive signal) or a plurality of drive pulses (drive signals). The waveform is output to the head driver 108.

  The head driver 108 selectively records drive pulses constituting a drive waveform supplied from the drive waveform generator 107 based on image data (dot pattern data) corresponding to one row of the recording head 14 input serially. The recording head 14 is driven by being applied to the pressure generating means of the head 14. For example, the head driver 108 includes a shift register that inputs a clock signal and serial data, a latch circuit that latches a register value of the shift register with a latch signal, and a level conversion circuit (level shifter) that changes the output value of the latch circuit. In addition, an analog switch array (switch means) that is controlled to be turned on / off by the level shifter and the like, and a required drive waveform included in the drive waveform from the drive generator 107 by controlling the on / off of the analog switch array. Is selectively applied to the pressure generating means of the recording head 14 to drive the recording head 14. Here, the drive waveform is composed of a plurality of drive pulses, and by giving one or a plurality of drive pulses, four types of gradations of large droplets, medium droplets, small droplets, and no droplets can be reproduced. .

  Next, a program according to the present invention (hereinafter referred to as “printer driver”) on the host side for transferring image data (print data) to the image forming apparatus in order to form an image by the image forming apparatus. An example of an image processing apparatus (data processing apparatus) according to the present invention will be described with reference to FIG.

  The printer driver 91 of the host 90 converts the image data 130 supplied from application software or the like from a monitor display color space to a recording device color space (RGB color system → CMY color system). Color Management Module) processing unit 131, BG / UCR (black generation / under color removal) processing unit 132 that performs black generation / under color removal from CMY values, and input / output correction that reflects the characteristics of the recording device and user preferences A gamma correction unit 133 that performs the total amount regulation, a total amount regulation unit 134 that regulates the total amount, a halftone processing unit 135 that includes a dither matrix that replaces the image data with a pattern arrangement of dots ejected from the recording apparatus, and divides the dot pattern data into data for each scan And a rasterizing unit 136 for developing data in accordance with the positions of the nozzles for recording. Sends the output of the grayed portion 136 to the image forming apparatus described above as the image data 137.

  Note that all or part of the functions of the printer driver 91 can be provided in the image forming apparatus, and the image forming apparatus according to the present invention is configured by providing the function of the halftone processing unit.

Next, the flow of image processing up to halftone processing by the host-side printer driver 91 will be described with reference to the block diagram shown in FIG.
When a “print” instruction is issued from application software running on a data processing apparatus such as a personal computer, the printer driver 91 determines the type of object in the object determination process 201 for the input 200, and for each object, that is, Data is transferred for each of the character image data 202, the line drawing image data 203, the graphics image data 204, and the image image data 205, and processing is performed through the respective routes.

  That is, color adjustment processing 206 is performed for the character 202, the line drawing 203, and the graphics 204. For characters, color matching processing 207, BG / UCR processing 209, total amount regulation processing 211, and γ correction processing 213 are performed, and character dither processing (halftone processing) 215 is performed. Further, color matching processing 208, BG / UCR processing 210, total amount regulation processing 212, and γ correction processing 214 are performed for line drawings and graphics, and graphics dither processing (halftone processing) 216 is performed.

  On the other hand, for the image 205, color determination and compression method determination processing 221 is performed. In a normal case, color adjustment processing 222 and color matching processing 223 are performed, then BG / UCR processing 224, total amount regulation processing 225, γ correction processing 226 is performed, and error diffusion processing (halftone processing) 227 is further performed. In the case of two colors or less, after performing image thinning processing 231, color adjustment processing 232, color matching processing 233a or indexless processing (processing without color matching) 233b, BG / UCR processing 224, total amount regulation Processing 225 and γ correction processing 226 are performed, and error diffusion processing (halftone processing) 227 is further performed.

  Note that the line drawing and graphics may branch before reaching the color adjustment 206 process, and may proceed to the color matching process 232 in the case of an image via the ROP process 241.

  The image data processed for each object in this way is combined with the original one image data.

  The image processing method according to the present invention relates to processing in the halftone processing unit, and particularly relates to correction of missing nozzles using error diffusion processing as halftone expression means.

  As a halftone processing method, a so-called “dither processing” in which a threshold matrix and pixel data are compared on a one-to-one basis and quantization is performed, and a quantization error of pixels already processed at the time of quantization is reflected. A so-called “error diffusion process” in which threshold processing is performed is common. Both methods have different calculation speeds (dither processing), image quality (error diffusion processing), and characteristics, so that they are properly used depending on the recording mode.

  However, with the recent increase in speed of computers, “error diffusion processing”, which was disadvantageous in terms of calculation speed, can now be processed at a speed that is not so different from dither processing. Mostly, halftone processing is performed by “error diffusion processing”.

  An example of this error diffusion processing is shown in FIG. In FIG. 13, the error value of the calculated pixel is quoted in the calculation of the next pixel. However, as shown in FIG. It is also possible to adopt a method of distributing to unprocessed pixels.

  Next, FIG. 15 shows an example of the correspondence with the nozzles used when the number of printing passes is 1 pass, and FIG. 16 shows an example of the correspondence with the nozzles used when the number of printing passes is 2 passes. ing. 15 and 16 are described as heads having eight nozzles for the sake of simplicity, it is apparent that the present invention can be applied to a head having several hundred nozzles.

  If high-speed recording is attempted without reducing the amount of information, as shown in FIG. 15A, one-pass recording is performed in which recording is performed in one recording pass at the resolution of the nozzle pitch arranged in the head. , Show the best performance. However, looking at “used nozzles for the head width” shown in FIG. 5B, the same nozzles are always used in the horizontal direction. For this reason, the non-ejection of the nozzles immediately leads to the generation of horizontal white streaks as shown in FIG.

  On the other hand, correction means using multi-pass printing is proposed in which the printing passes are divided into a plurality of dots so that the vertical and horizontal dots are not always performed by the same nozzle, and the nozzles used for the dot positions in each row are switched. (For example, JP-A-5-309874, JP-A-2001-063008, etc.).

  An example of this will be described with reference to the two-pass recording shown in FIGS. In two-pass printing, as shown in FIG. 16A, one head is printed in two passes while using half of the head nozzles. Looking at the “used nozzles”, corresponding nozzles are alternately used in each pass. For example, No. 1 nozzle and No. 5 nozzle are used alternately.

  Therefore, for example, when the No. 3 nozzle becomes a non-ejection nozzle and recording is performed with the non-ejection nozzle as it is, as shown in FIG. 17B, the recording position (dot) by the non-ejection nozzle has a gap and a space. Will occur. Here, by using another nozzle forming the horizontal line (No. 7 in the figure) as an alternative to the No. 3 nozzle, as shown in FIG. It becomes possible to cover. In the figure, two-pass printing is described. However, it is obvious that the same correction can be performed in multi-pass printing with more passes.

  However, as a matter of course, in multi-pass recording, the recording speed decreases with the number of passes. Not only the high-speed recording mode in which the recording speed is first required but also the high-quality recording mode has recently become important, and recording with as few passes as possible is required.

  However, when the number of passes decreases, the nozzles that can be used for replacement are limited, and the load on the replacement nozzle increases. An increase in the load may lead to faster nozzle consumption and may lead to a second non-ejection nozzle. Also, there are cases where the design must be divided into as many paths as possible. For example, in a thermal type ink jet recording apparatus using film boiling due to heating, in order to avoid heat storage, multi-pass recording in which driving can be divided is preferred over one-pass recording that requires continuous driving. In addition, there is a problem that may occur in any type of ink jet recording apparatus, but there are cases in which the drive waveform for stably ejecting ink droplets cannot be accommodated within the ejection period of the head. In this case, continuous dot formation is practically impossible, and the two-pass printing correction shown in FIG. 17C cannot be realized.

  Even in the correction by multi-pass printing, it is possible to adjust the printing speed and the load of the substitute nozzle to some extent by intelligently managing the combination of the substitute nozzles and the frequency of use according to the image data. However, the smaller the number of passes (= the fewer alternative nozzles), the narrower the adjustable range, and a separate mechanism for storing / calculating the usage frequency according to the image data is required.

  Therefore, in consideration of this point, the present invention corrects a non-ejection nozzle by a simple method. Specifically, in the error diffusion process, the dot is forcibly turned off at the coordinates where the non-ejection nozzle is applied, and the error is diffused to surrounding pixels, thereby making the dot defect due to the non-ejection nozzle inconspicuous. is there.

  The defect nozzle correction processing according to the present invention will be described with reference to FIG. Here, the description is based on the error diffusion processing shown in FIG. 13, but it can also be used for the error diffusion processing of the method of distributing errors to unprocessed pixels shown in FIG.

  In this error diffusion process, input image data is processed in units of pixels. The corrected pixel value Dxy is calculated by adding the error value of the processed peripheral pixel to the input value of the target pixel in accordance with the weight matrix ratio. Then, referring to the non-ejection nozzle management table, which is a recording sequence table showing the correspondence between nozzles and pixel positions reflecting non-ejection nozzle data, whether or not the pixel position of interest is a pixel position recorded by the non-ejection nozzles Is determined. As a result of this determination, if the target pixel position is not a non-ejection nozzle, normal error diffusion processing is continued. If the target pixel position is a non-ejection nozzle, even if “threshold ≦ correction pixel value Dxy” is satisfied. Correction processing for forcibly turning off dots is performed, and all correction pixel values Dxy are processed as error values.

  As a result, the missing dot due to non-ejection is reflected in the surrounding pixels in the form of an error value, and the dot conversion of the surrounding pixels is performed so as to correct the missing part. As a result, the missing part becomes inconspicuous. FIG. 19 shows an example of an output result when the correction is not performed and when the correction is performed. FIG. 19A shows the case where the above correction is not performed in the two-pass printing, and FIG. 19B shows the case where the above correction is not performed in the two-pass printing. From this, the above correction is performed. It turns out that the missing part is not noticeable.

  Here, the non-ejection nozzle management table will be described. The non-ejection nozzle management table may be as simple as shown in FIG. 20, for example. Since the recording sequence table showing the correspondence between the nozzles and the pixel positions is originally incorporated in the driver process, it is based on the recording sequence table showing the correspondence between the nozzles and the pixel positions and the non-ejection nozzle information (non-ejection nozzle data). Thus, a non-ejection nozzle management table that is a recording sequence table reflecting non-ejection nozzle information can be created.

  This non-ejection nozzle information is obtained by reflecting the inspection data based on the non-ejection nozzle check pattern as shown in FIGS. 21A and 22A, thereby easily creating the non-ejection nozzle management table. be able to.

  The non-ejection nozzle detection pattern shown in FIG. 21A is an example of a check pattern used when checking the reproduction of the horizontal line corresponding to the nozzle and visually judging the non-ejection nozzle. When this non-ejection nozzle detection pattern is used, for example, as shown in FIG. 5B, if the third nozzle is non-ejection, the dot corresponding to the third nozzle is missing.

  Further, the non-ejection nozzle detection pattern shown in FIG. 22A is an example of a check pattern for automatically detecting a non-ejection nozzle using a scanner unit, a photo sensor, or the like. When this non-ejection nozzle detection pattern is used, for example, as shown in FIG. 4B, if the No. 7 nozzle is non-ejection, the density measurement value by the photosensor or the like of the pattern corresponding to the No. 7 nozzle is normal. It is possible to determine that no discharge occurs without reaching the level.

  In this way, correction using multi-pass printing is performed by performing correction processing to compensate for defects in the recorded image due to nozzles that are unable to perform normal printing in halftone processing that converts multi-value data into dot patterns. Overcorrection is applied without causing a significant decrease in recording speed as in the case of the method, and can directly apply correction processing to conversion processing from multi-value data to dots, thereby losing the balance of dot arrangement. Can be prevented. This eliminates image irregularities such as white streaks in the recorded image caused by non-ejection of dots that are not recorded, and even when non-ejection occurs, the human eyes recognize white streaks and image irregularities. This makes it difficult to reduce the cost of the recording head and to increase the recording speed.

  Such correction processing in halftone processing uses error diffusion processing as halftone processing, and in the correction processing, a recording sequence table reflecting non-ejection nozzle data is reflected in error diffusion processing as dot generation conditions. Correction processing can be performed pinpointed on pixels where image defects due to ejection occur, and distortion generated by correction processing is distributed to surrounding pixels and made conspicuous by performing correction using error diffusion processing. Can be eliminated.

  When a recording sequence table reflecting non-ejection nozzle data is reflected in error diffusion processing as a dot generation condition, the recording position by the non-ejection nozzle is forcibly turned off while the quantization error itself remains as it is in the surrounding pixels. Thus, the dot conversion of the surrounding pixels can be performed so as to reflect the missing dot due to the non-ejection to the surrounding pixels in the form of an error value and correct the missing portion.

  Next, the relationship with gradation expression in the ink jet recording apparatus will be described. Here, the binary value is a case of “1” and “0”, that is, “ON” and “OFF”, and “small value” is generally referred to as “multi-value”, “ It is used as a comparison for “binary”, and the amount of information has a relationship of “multivalue” ≧ “small value”> “binary”. In general, when image processing is performed, multi-value data having an information amount of about 8 bits (256 values) per pixel is used as input image data. However, in an apparatus that actually outputs the data, Since each pixel has only a descriptive power of about 1 to 3 bits, for convenience, in the present invention, a small amount of information is called “low value” in order to call it multi-value but with a small amount of information.

  FIG. 23 shows an example of a dot pattern when a binarization process and a binarization process that are generally used are applied. FIG. 6A shows an example in the case of performing dot reproduction by performing binarization processing, FIG. 9B shows an example in the case of performing dot reproduction by performing density modulation, and FIG. This is an example of performing dot reproduction.

  Among these, in the value reduction process employing the density modulation method, since the recording units are divided for different density inks, in the present invention, they can be handled in the same row as the recording units of different colors. Since the gradation represented by the dark and light ink is inherited without any problem in the error diffusion processing, the above-described correction method can be applied as it is.

  On the other hand, in the case of the dot size modulation method, there are cases where the nozzles do not necessarily completely fail to discharge. Since dots of a plurality of sizes are formed with one nozzle, depending on the case, there is a case where only a specific size is ejected in an unstable or non-ejection state, and other sizes can be ejected without any problem. In this case, if the nozzles are handled completely as non-ejections, the correction process is forcibly applied to a gradation pattern that can be normally recorded, resulting in excessive correction.

  Therefore, in an ink jet recording apparatus that can perform a value reduction process by dot size modulation, a non-ejection nozzle management table is created for each dot size, and the non-ejection nozzle management table is switched in accordance with the threshold level for the reduction in the value. Therefore, it is preferable that the correction processing according to the present invention is applied only to the problematic dot size.

  In this way, the error diffusion process is a reduction error diffusion process corresponding to n values (n ≧ 2), and dot size modulation is possible by applying correction processing for each dot size that can be expressed by each nozzle. In the case of a simple ink jet printing apparatus, even if non-ejection (or unstable ejection that is treated as non-ejection) occurs only in some size dots, correction processing is applied only to unstable dot sizes. Therefore, it is possible to prevent over-correction and the overall image quality from being deteriorated.

  In the above embodiment, the printer driver as a program performs the correction process by the halftone process. However, as described above, it can also be performed by, for example, an ASIC having the same function. In particular, in printers that support direct printing without using a host PC, fax machines and copiers equipped with an ink jet recording head as a recording unit, such an ASIC can be used to correct non-ejection nozzles according to the present invention. It becomes possible. In addition, when viewed as a program (including a printer driver) that implements the present invention, it is easy to distribute, install, and save the program via various storage media (CD-ROM, DVD-ROM, memory card, etc.) and a network. Can be performed.

1 is a schematic configuration diagram of a mechanism unit of an ink jet recording apparatus showing an example of an image forming apparatus according to the present invention. It is principal part plane explanatory drawing of the mechanism part. It is a perspective explanatory view explaining the head unit composition of the device. It is explanatory drawing explaining an example of the conveyance belt of the apparatus. FIG. 6 is an explanatory diagram for explaining an image forming operation by the apparatus. It is explanatory drawing which shows the other 1st example of the head structure which uses special color ink other than CMYK. It is explanatory drawing which similarly shows a 2nd example. It is explanatory drawing which similarly shows the 3rd example. It is explanatory drawing which similarly shows the 4th example. It is a block diagram which shows the outline | summary of the control part of the apparatus. FIG. 2 is a block diagram functionally illustrating an example of a configuration of a printer driver according to the present invention in an image processing apparatus according to the present invention. FIG. 3 is a block explanatory diagram illustrating details of a flow of image processing in a printer driver. It is explanatory drawing explaining the basic arithmetic processing of an error diffusion process. It is explanatory drawing explaining the other example of the propagation method of an error. It is explanatory drawing with which it uses for description of the relationship between 1 pass printing and a use nozzle. It is explanatory drawing with which it uses for description of the relationship between 2 pass printing and a use nozzle. It is explanatory drawing with which it uses for description of the dot defect | deletion by a non-ejection nozzle, and the complement by an alternative nozzle. It is a flowchart with which it uses for description of the correction | amendment process of the error diffusion process which concerns on this invention. It is explanatory drawing which shows an example of a case where a defect | deletion nozzle is not correct | amended and a defect | deletion nozzle is correct | amended by the process which concerns on this invention. It is explanatory drawing explaining the creation example of a non-ejection nozzle management table. It is explanatory drawing explaining an example of the chart for non-ejection nozzle inspection. It is explanatory drawing explaining the other example of the chart for non-ejection nozzle inspection. It is explanatory drawing with which it uses for description of a binarization process and a value reduction process.

Explanation of symbols

DESCRIPTION OF SYMBOLS 2 ... Image formation part 3 ... Paper 5 ... Conveyance mechanism part 14 ... Recording head 33 ... Conveyance belt 90 ... Host (image processing apparatus)
91: Printer driver 135: Halftone processing unit

Claims (10)

  Converts multi-value data into dot patterns in an image processing method for processing image data for an image forming apparatus that mounts a recording head having a plurality of nozzles that eject recording liquid droplets and forms an image on a recording medium. An image processing method comprising: performing correction processing to compensate for defects in a recorded image caused by nozzles that cannot perform normal recording in halftone processing.   The image processing method according to claim 1, wherein an error diffusion process is used as the halftone process, and the correction process uses a recording sequence table indicating correspondence between nozzles and pixel positions reflecting non-ejection nozzle data as an error in dot generation. An image processing method, which is reflected in diffusion processing.   3. The image processing method according to claim 2, wherein when the recording sequence table reflecting the non-ejection nozzle data is reflected in the error diffusion process as a dot generation condition, the recording position by the non-ejection nozzle is forcibly turned off while the quantum is turned off. An image processing method characterized in that the conversion error itself is developed as it is to surrounding pixels.   4. The image processing method according to claim 2, wherein the error diffusion process is a reduction error diffusion process corresponding to n values (n ≧ 2), and correction processing is performed for each dot size that can be expressed by each nozzle. An image processing method characterized by being applied.   A program for causing a computer to execute the image processing method according to claim 1.   An image processing apparatus comprising the program according to claim 5.   In an image forming apparatus that mounts a recording head having a plurality of nozzles that eject recording liquid droplets to form an image on a recording medium, normal recording can be performed by halftone processing that converts multi-value data into a dot pattern. An image forming apparatus that performs a correction process to compensate for defects in a recorded image due to a missing nozzle.   8. The image forming apparatus according to claim 7, wherein an error diffusion process is used as the halftone process, and the correction process uses a recording sequence table indicating correspondence between nozzles and pixel positions reflecting non-ejection nozzle data as an error in generating dots. An image forming apparatus that is reflected in diffusion processing.   9. The image forming apparatus according to claim 8, wherein when the recording sequence table reflecting the non-ejection nozzle data is reflected in the error diffusion process as the dot generation condition, the recording position by the non-ejection nozzle is forcibly turned off while the quantum is turned off. An image forming apparatus characterized in that the conversion error itself is developed as it is to surrounding pixels. 10. The image forming apparatus according to claim 8, wherein the error diffusion processing is a reduction error diffusion processing corresponding to n values (n ≧ 2), and correction processing is performed for each dot size that can be expressed by each nozzle. An image forming apparatus characterized by being applied.