JP2018094879A - Image processing device, image processing method and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To convert image data into data constituted of bit patterns including a bit pattern for a first image area to be reproduced with high resolution and a bit pattern for a second image area to be reproduced with low resolution, without losing color information.SOLUTION: An image processing device, which converts image data of bit map form into data constituted of bit patterns including a first bit pattern and a second bit pattern, comprises: first creating means that creates the first bit pattern storing color information on a first image area on the basis of pixel values of pixel in the first image area to be reproduced with first resolution; and second creating means that creates the second bit pattern storing color information on a second image area having more data amounts than the color information on the first image area, on the basis of pixel values of pixel in the second image area to be reproduced with second resolution lower than the first resolution.SELECTED DRAWING: Figure 4

Description

  The present invention relates to an image processing technique in a printer or the like.

  There is a need for a technique for efficiently holding information on bitmap-format image data obtained by rendering processing for PDL data using data having a smaller data amount than the image data. As such a technique, for example, Patent Document 1 develops intermediate language data on a memory as multi-value data when gradation is important, while intermediate language data is binary data when importance is given to resolution. Disclosed is a technique for developing on a memory.

JP-A-10-240482

  However, Patent Document 1 does not disclose that color information of image data is held when the resolution is important. For example, in a business document, as shown in FIG. 1, an expression that emphasizes a character by arranging a thin intermediate color rectangle on the background of the character is generally used. When the method of Patent Document 1 is applied to such an image, the emphasized expression is lost as a result of the lack of the background rectangle in the region to be emphasized (the region where the character string “problem” exists in FIG. 1). As a result, a document that does not reflect the user's intention is printed.

  Therefore, in view of the above-described problems, the present invention provides a bit pattern for a first image area that is reproduced at a high resolution and a bit pattern for a second image area that is reproduced at a low resolution without losing color information. It is intended to convert the data into a bit pattern including

  The present invention is an image processing apparatus for converting image data in a bitmap format into data composed of a bit pattern including a first bit pattern and a second bit pattern, which is reproduced at a first resolution. First creation means for creating the first bit pattern storing color information of the first image area based on pixel values of pixels in the image area, and a second lower than the first resolution. The second bit pattern in which the color information of the second image area having a larger data amount than the color information of the first image area is stored based on the pixel value of the pixel in the second image area to be reproduced at the resolution. And a second creating means for creating the image processing apparatus.

  According to the present invention, the image data is composed of a bit pattern including a bit pattern for the first image area reproduced at high resolution and a bit pattern for the second image area reproduced at low resolution without losing color information. It can be converted into data.

The figure explaining the subject of this invention 1 is a block diagram of a system including an image forming apparatus according to a first embodiment. Flowchart of processing executed by printer in embodiment 1 Flowchart of conversion processing in the first and second embodiments FIG. 6 is a diagram for explaining edge detection processing in the first embodiment. The figure explaining the bit pattern of the edge area | region in Example 1 The figure explaining the bit pattern of the solid area in Example 1 A diagram for explaining how image information is arranged in a memory after completion of conversion processing on image data Flowchart of image processing in Embodiment 1 Flowchart of recording data creation processing in Embodiment 1 FIG. 3 is a diagram illustrating a mechanism related to image formation of a printer in Embodiment 1. Bit pattern of edge region in embodiment 2 Bit pattern of solid area in embodiment 2 Diagram showing VTF characteristics FIG. 10 is a diagram for explaining a color reduction process executed using the dither method according to the third embodiment. Diagram explaining binarization processing Diagram explaining color information derivation method

  Hereinafter, exemplary embodiments of the invention will be described by way of example with reference to the drawings. However, the contents, relative arrangement, and the like of the constituent elements described below are not intended to limit the scope of the invention only to those unless otherwise specified.

[Example 1]
FIG. 2 is a block diagram of a system including the image forming apparatus 200 in the present embodiment. In this system, an image forming apparatus 200 and a host 210 are connected.

<About image forming apparatus>
Hereinafter, the configuration of the image forming apparatus 200 shown in FIG. 2 will be described. The CPU 201 executes a program stored in the ROM 202 using the RAM 203 as a work memory, and comprehensively controls each component of the image forming apparatus 200 via the bus 207. Thereby, various processes described later are executed. The ROM 202 stores programs and tables for performing print processing, an embedded operating system (OS) program, and the like. In the present embodiment, the program stored in the ROM 202 performs software control such as scheduling and task switching under the management of the embedded OS stored in the ROM 202. Further, a program for executing the flow in this embodiment such as FIG. The RAM 203 is configured by an SRAM (static RAM) or the like, and a work area is provided when the program is executed.

  The head 204 forms an image on the recording medium by discharging the color material in synchronization with the conveyance of the recording medium (paper or the like). An interface (hereinafter referred to as IF) 205 connects the image forming apparatus 200 and an external apparatus (host 210 in this embodiment) of the image forming apparatus 200 by wire or wirelessly. Although not shown, the image forming apparatus 200 includes mechanisms such as a motor that drives the head 204 and a motor that transports the recording medium. Further, the image forming apparatus 200 may include a DSP (Digital Signal Processor) 206 as a hardware device for performing high-load processing such as image processing.

  These components are electrically connected to each other via a bus 207. In addition, all or some of these components may be mounted as a single LSI and may be componentized as an ASIC.

  Hereinafter, a case where the image forming apparatus 200 is an inkjet printer including one CPU, ROM, and RAM will be described. However, the image forming apparatus 200 may include a plurality of CPUs, ROMs, and RAMs, and may be a printer of an arbitrary system such as an electrophotographic system. In addition, the image forming apparatus in the present embodiment is not limited to a dedicated machine specialized for the printing function, but is a multi-function machine that combines the printing function and other functions, or a manufacturing apparatus that forms an image or pattern on a recording medium. Etc. are also included.

<About the host>
The image forming apparatus 200 is connected to the host 210 via the I / F 205. The host is a computer having a processing capability, and is assumed to be a general PC, a mobile phone, a smartphone, a tablet, a digital camera, and other portable and stationary terminals. Although the breakdown of devices included in the host varies depending on the main purpose of each device, the host generally includes a CPU, a ROM, a RAM, an IF, an input device, and an output device.

  Hereinafter, the configuration of the host 210 shown in FIG. 2 will be described. The CPU 211 executes a program stored in the ROM 212 using the RAM 213 as a work memory, and comprehensively controls each component of the host 210 via the bus 217. The IF 214 connects the host 210 and the image forming apparatus 200 by wire or wirelessly. These components are electrically connected to each other via a bus 217. In addition, the host 210 includes an input device 215 that is connected via the IF 214 and allows the user to input an instruction to the host, and an output device 216 that outputs (presents) information to the user.

<Data transmission processing from the host to the image forming apparatus>
Hereinafter, data transmission processing from the host to the image forming apparatus will be described by taking as an example a case where a document is printed from a smartphone using an inkjet printer.

  When a user viewing a document on a smartphone wants to print the document being viewed, the user operates the smartphone to input a print instruction. When a print instruction is input, the OS on the smartphone converts the document into some form of image data, and then creates a print job according to the print protocol (for example, converts the document to be printed into PDF data, Create a print job in XML format that includes PDF data). The smartphone transmits the created print job to the inkjet printer.

<About processing executed by inkjet printer>
Hereinafter, the processing executed by the ink jet printer will be described with reference to FIG.

  In step S300, the CPU 201 analyzes a print job transmitted from a host such as the above-described smartphone and obtains print data. Specifically, the CPU 201 analyzes a print job using an XML parser, acquires setting information including information such as paper to be used and print quality, and data (PDL data) described in PDL, and The acquired setting information and PDL data are stored in the RAM 203.

  In step S310, the CPU 201 interprets the PDL data derived in step S300 and creates an intermediate file called a display list. Then, a rendering process is executed on the display list to create a 2400 dpi × 2400 dpi bitmap image in which each pixel has a value of 16 bits for each channel of R (red), G (green), and B (blue). This bitmap image is abbreviated as “2400 dpi × 2400 dpi, 16-bit, 3ch bitmap image”, and the other bitmap images are abbreviated in the same manner. The bitmap image created in this step is not limited to a 16-bit bitmap image, and a bitmap image (for example, a low-bit bitmap image such as 8 bits or 3ch) corresponding to the processing capability of the printer is created. You can do it. Also, the creation of the display list and the creation of the bitmap image may be performed collectively for the entire page. In order to reduce the amount of use of the RAM 203, in response to a request from the subsequent step S330, in units of bands. May be executed. Further, this step may be executed by the DSP 206 instead of the CPU 201.

  In step S320, the CPU 201 executes a conversion process on the bitmap image derived in step S310. In the conversion process, a 4 × 4 pixel area in a 2400 dpi × 2400 dpi image is handled as one tile, and it is determined for each tile whether it is an edge area or a solid area. Each tile is represented by a 32-bit bit pattern (FIG. 6A) corresponding to the determination result. Specifically, a tile determined to be an edge region is represented by a 32-bit bit pattern shown in FIG. 6B, while a tile determined to be a solid region is a 32-bit bit shown in FIG. 7A. Expressed with a pattern. As described above, in this embodiment, information for all pixels in each image area is expressed by a 32-bit bit pattern in both the edge area and the solid area. The data amount (32 bits) of this bit pattern is smaller than the data amount (768 bits) of all the pixels constituting the edge region or the solid region. Here, in the most significant bit b31 of the bit pattern, an identification flag indicating whether it is an edge region or a solid region is stored, and 0 is stored in the case of the edge region, and 1 is stored in the case of the solid region. On the other hand, data corresponding to the information of each tile is stored in the lower bits excluding the most significant bit. The conversion process in this step will be described later with reference to FIG.

  In step S330, the CPU 201 executes image processing based on the data derived by the conversion processing in step S320. Through this step, image data in which each pixel has an ink color value is derived and developed on the RAM 203. This image data is, for example, 1-bit, 4-channel image data in which each pixel has a value of 1 bit for each channel of C (cyan), M (magenta), Y (yellow), and K (black). The image processing in this step will be described later with reference to FIG.

  In step S340, the CPU 201 creates recording data based on the image data derived in step S330. This recording data is, for example, binary image data in which each pixel has a value of 0 or 1 for each channel of ink color (CMYK, etc.). The recording data created in this step is developed on the RAM 203 so as to form an image in the correct direction on the paper when transmitted to the head 204. The recording data creation process in this step will be described later with reference to FIG.

  In step S350, the CPU 201 transmits the recording data derived in step S340 to the head 204, drives the head 204 and the feeder based on the transmitted recording data, and performs an image forming process for actually forming an image on a sheet. Run. The image forming process in this step will be described later with reference to FIG.

<About conversion processing>
Hereinafter, the conversion process (step S320 in FIG. 3) in the present embodiment will be described with reference to FIG.

  In step S400, the CPU 201 executes an edge detection process that applies an edge filter to the bitmap image derived in step S310. In this embodiment, a convolution operation using a 3 × 3 filter shown in FIG. For example, when the filter shown in FIG. 5A is convoluted with the image shown in FIG. 5B, the edge image shown in FIG. 5C is derived.

  In step S410, the CPU 201 initializes n, that is, sets 1 to the value of n. Here, n is a parameter that represents a target tile that is a region to be processed, that is, a region of a predetermined size in the image data (a region of 4 × 4 pixels in an image of 2400 dpi × 2400 dpi in this embodiment). Since the subsequent steps S420 to S460 are sequentially executed for each tile, initialization is performed in this step.

  In step S420, the CPU 201 derives an edge amount of an area (attention area) corresponding to the target tile in the edge image derived in step S400. Here, the edge amount is a parameter used as an index when determining whether a target tile is an edge region (that is, a region to be reproduced at a high resolution), and is a total pixel value of each pixel in the target region. In the present embodiment, when the edge amount of an arbitrary region of interest is larger than a predetermined threshold Th1, the tile of interest corresponding to the region of interest is determined to be an edge region.

  In step S430, the CPU 201 determines whether the edge amount of the attention area is larger than the threshold value Th1. If the result of the determination in step S430 is true, the target tile is determined to be an edge region, and the process proceeds to step S431. On the other hand, if the result of the determination in step S430 is false, the target tile is determined to be a solid region, and the process proceeds to step S435. Note that the algorithm for edge detection to edge region determination is not limited to the above-described method, and other methods may be used or a known technique may be used in combination. For example, it may be determined whether the object is text based on the display list derived in step S310, and the determination result may be used to determine whether the object is an edge region or a solid region.

  Hereinafter, a case where the tile of interest is determined to be an edge region (YES in step S430) will be described. In this case, in step S431, the CPU 201 performs a quantization process on the target tile in the full-color image derived in step S310, and derives a bitmap pattern expressed by only two colors, the foreground color and the background color. In the derivation of the bitmap pattern in this step, an arbitrary method such as adaptive binarization processing may be used. Here, an example of adaptive binarization processing will be described with reference to FIG. First, the pixel values of each pixel constituting the two-dimensional image on the two-dimensional XY plane as indicated by reference numeral 1601 in the figure are plotted in the RGB space as indicated by reference numeral 1602. On the other hand, using a clustering method (for example, k-means method), colors are divided into two groups as indicated by reference numeral 1603, one of the divided groups is a background color (light color), and the other is a foreground color. Decide (dark color). When the result indicated by reference numeral 1603 is fed back to the two-dimensional XY plane, as shown by reference numeral 1604, a two-dimensional image composed of two types of pixels, foreground pixels and background pixels, is obtained. Note that the pixel values of the foreground pixels and the background pixels are the average values of the colors of each group as indicated by reference numeral 1605. By this step, shape information (16-bit data) indicating which pixel corresponds to the foreground pixel and which pixel corresponds to the background pixel among the 4 × 4 pixels constituting the target tile is acquired. Specifically, shape information that takes a value of 1 when the pixel of interest corresponds to a foreground pixel and takes a value of 0 when the pixel of interest corresponds to a background pixel is a 32-bit bit pattern shown in FIG. Are stored in the last 16-bit area “map — 4 × 4”. Thus, in this step, the CPU 201 functions as shape information deriving means.

  In step S432, the CPU 201 executes a color reduction process for reducing the number of gradations for each of the two colors (16 bits and 3ch, respectively) used in the bitmap pattern derived in step S431. Specifically, the foreground color is represented by 1 bit, 3 ch 3 bit data having 1 bit value for each RGB channel, while the background color is a 4 bit, 3 ch value having 4 bit values for each RGB channel. Expressed with 12-bit data. That is, any one of eight types of pure colors of white, black, red, blue, green, cyan, magenta, and yellow can be designated as the foreground color. The RGB value of the foreground color after the color reduction process is stored in the 3 bit area “fg_rgb” of the bit pattern shown in FIG. 6B, and the RGB value of the background color after the color reduction process is the 12 bit area “bg_rgb”. Stored in Thus, in this step, the CPU 201 functions as color information deriving means. Here, a method for deriving color information will be specifically described with reference to FIG. In this embodiment, quantization processing using a dither table as indicated by reference numeral 1702 is performed for each of the background color average value and foreground color average value as indicated by reference numeral 1701 described with reference to FIG. . The dither table has a threshold value for each tile in units of 4 × 4 pixels (= 1 tile). This quantization processing may be performed using an already known algorithm. For example, when the background color average value is (0, 10, 230) and the foreground color average value is (0, 0, 5), if the quantization process is performed using the dither table denoted by reference numeral 1702, the background color quantization is performed. (0, 1, 13) is obtained as the value, and (0, 0, 0) is obtained as the foreground color quantization value. After the quantization process, the quantized values obtained for the background color and the foreground color are stored in the corresponding portions of the bit pattern. Each bit pattern may be used as a palette color having a dictionary for each page. When the aforementioned bit pattern is used as the palette color, any one of the eight colors can be designated as the foreground color and any one of 2 ^ 12 = 4096 colors can be designated as the background color. By switching the palette to be used according to the contents of each page, it is possible to express colors according to the contents of each page.

  Next, a case where the target tile is determined to be a solid area (NO in step S430) will be described. In this case, in step S435, the CPU 201 calculates the sum of the pixel values of each pixel in the target tile in the full color image derived in step S310. As described above, in the present embodiment, the target tile indicates a 4 × 4 pixel area in a 2400 dpi × 2400 dpi image, and each pixel constituting the target tile has 16 bits and 3 ch information. Therefore, in this step, as a result of totaling 16 pixel values of pixels having values in the range of 0 to 65535 per channel in the original image, the target tile has a value in the range of 0 to 1048560 per channel. Thus, such a value can be expressed by 20 bits.

  In step S436, the CPU 201 performs a color reduction process for reducing the number of gradations for each of the RGB values (20 bits each) derived in step S435, and derives 10-bit RGB values. The R value derived in this step is stored in the 10-bit area “r10” of the bit pattern shown in FIG. 7A, and similarly, the G value is stored in the 10-bit area “g10”, and the B value is It is stored in the area “b10” for 10 bits. The information of 10 bits and 3ch derived in steps S435 to S436 is handled as information of one pixel in a 600 dpi × 600 dpi image in a subsequent step.

  As already described, the data size (bit length) of the bit pattern representing the edge region (FIG. 6B) and the bit pattern representing the solid region (FIG. 7A) are both 32 bits. Thus, by making the data formats of different image areas have the same data size, it is possible to easily perform random access in units of assigned formats and improve the efficiency of data processing.

  Returning to the description of the flow of FIG. In step S440, the CPU 201 arranges the bit pattern on the memory. FIG. 8 is a diagram for explaining how image information is arranged on the memory after the conversion process on the image data is completed. Reference numeral 801 in FIG. 8 indicates image data including two types of image areas, an edge area and a solid area, and reference numeral 802 indicates how image information is arranged on the memory after the conversion process is completed. . In FIG. 8, the bit pattern of the edge area shown in FIG. 6B is expressed in the abbreviated form shown in FIG. 6C, and the bit pattern of the solid area shown in FIG. 7A is shown in FIG. It is expressed in the abbreviation format shown in.

  In step S450, the CPU 201 increments n.

  In step S460, the CPU 201 determines whether n is larger than the total number of tiles. If the result of the determination in step S460 is true, the series of processing ends, while if the result of the determination is false, the process returns to step S420. The above is the content of the conversion process in the present embodiment.

  In the conversion process of the present embodiment, the edge region is represented by a high resolution (one pixel in an image of 2400 dpi × 2400 dpi) in exchange for expressing the foreground color and the background color with low bits (with low gradation). Expressed in units). On the other hand, for the solid area, the color is expressed with high bits (with high gradation) in exchange for expressing the shape with low resolution (in units of one pixel in an image of 600 dpi × 600 dpi).

  Here, the color reproduction capability of an image is determined based on the number of colors that can be used simultaneously in the image (three colors of R, G, and B in this embodiment) and the number of gradations that can be expressed by each color. On the other hand, the shape reproducibility in an image depends on how finely the signal value of each pixel can be expressed (that is, at what resolution). In other words, in this embodiment, the shape reproduction ability is given priority over the color reproduction ability in the edge area, the color reproduction ability is given priority over the shape reproduction ability for the solid area, and the edge area and the solid area have the same data size. Expressed in bit pattern.

<About image processing>
Hereinafter, the image processing (step S330 in FIG. 3) in the present embodiment will be described with reference to FIG. This image processing is executed for all bit patterns on the memory arranged as shown in FIG.

  In step S900, the CPU 201 initializes the value of m, that is, sets 1 to the value of m. Here, m is a parameter representing a bit pattern to be processed (hereinafter referred to as a target bit pattern). The subsequent processes in steps S910 to S960 are sequentially executed for each 32-bit bit pattern corresponding to one tile on the memory arranged as shown in FIG. 8, and therefore initialization is performed in this step. As described above, both the edge area of one tile and the solid area of one tile are expressed by a 32-bit bit pattern, so that processing for each bit pattern is performed at high speed by, for example, burst transfer using a DMA controller. Can be executed sequentially.

  In step S910, the CPU 201 determines whether the target bit pattern represents an edge area or a solid area. Specifically, it is determined whether the most significant bit (b31) of the bit pattern of interest is 0. If the result of the determination is true, it is determined that the bit pattern of interest represents an edge region, and step S921 Proceed to On the other hand, if it is determined that b31 is not 0, it is determined that the target bit pattern represents a solid area, and the process proceeds to step S931.

  Hereinafter, a case where it is determined that the bit pattern of interest represents an edge region (YES in step S910) will be described. In this case, in step S921, the CPU 201 executes a color conversion process for converting each of the foreground color and the background color expressed by the bit pattern of interest into a color in the recording color space of the image processing apparatus in the printer. . As the recording color space, a color space representing a color gamut (gamut) that can be printed by the printer is used. Through the color conversion processing in this step, the input color outside the printer gamut is converted to a color within the printer gamut (gamut compression). In step S432, the foreground color in the edge area is reduced to 3 bits (8 colors) and the background color is reduced to 12 bits (4096 colors). Therefore, a color conversion LUT that directly corresponds to these colors is prepared in advance. Thus, the color conversion process of this step can be easily executed. That is, it is possible to omit the linear interpolation calculation process after referring to the LUT, which occurs during the normal color conversion process.

  In step S922, the CPU 201 executes color separation processing for converting the multi-value information of the color derived in step S921 into multi-value information of the color material used in the printer, for each of the foreground color and the background color. For example, in the case of an ink jet printer (hereinafter referred to as CMYK ink jet printer) in which four colors of inks of C, M, Y, and K are used, RGB values are converted into CMYK values. For example, RGB (0, 0, 0) indicating black is converted to CMYK (0, 0, 0, 255), and RGB (255, 0, 0) indicating red is CMYK (0, 255, 255, 0). Is converted to

  In step S923, the CPU 201 executes gradation correction processing for correcting the gradation of the signal value of each ink color after the color separation processing derived in step S922 for each of the foreground color and the background color. The purpose of executing the gradation correction process in this step is as follows. In general, in an ink jet printer, the greater the amount of ink discharged per unit area, the stronger the ink color appears on the paper. However, since there is a non-linear relationship between the ink ejection amount and the color development on the paper (spectral reflectance, Lab value, etc.), tone correction processing is performed in this step to correct this color development. To do. Since the ink discharge amount may change due to the manufacturing intersection of the head 204, this error may be absorbed by the gradation correction processing in this step. In that case, as a method for estimating the ejection amount of the head, any known technique such as a method of causing the user to print a test pattern can be used.

  In step S924, the CPU 201 executes a quantization process for binarizing the signal value of each ink color derived in step S923. Here, the processing in steps S921 to S923 in the preceding stage is executed for each target bit pattern, whereas the quantization processing in this step is performed in units of pixels in the tile corresponding to the target bit pattern, that is, 2400 dpi × 2400 dpi. Need to run at resolution. Specifically, based on the shape information stored in the “map — 4 × 4” bit pattern shown in FIG. 6B, it is determined whether each pixel is a foreground pixel or a background pixel. Then, in accordance with the result of this determination, a value obtained by binarizing the signal value of each ink color (foreground color or background color) derived in step S923 is assigned to each pixel. For example, in the case of a CMYK inkjet printer, a total of 64 bits of output results corresponding to 16 pixels having 1 bit and 4 ch values (ie, one tile) are obtained in this step.

  Next, a case where it is determined that the target bit pattern represents a solid area (NO in step S910) will be described. In this case, in step S931, the CPU 201 changes the color represented by the values of r10, g10, and b10 (FIG. 7A) in the bit pattern of interest to colors in the recording color space of the image processing apparatus in the printer. Execute color conversion processing to convert.

  In step S932, the CPU 201 executes color separation processing for converting the multi-value information of the color derived in step S931 into multi-value information of the color material of the printer.

  In step S933, the CPU 201 executes a gradation correction process for correcting the gradation of the signal value of each ink color after the color separation process derived in step S932.

  In step S934, the CPU 201 executes a quantization process for converting the signal value of each ink color derived in step S933 into an ejection level that defines the ink ejection amount per unit area. Here, the discharge level takes 16 levels (0 to 15). In this step, only the discharge level is derived, and in the subsequent index development process (step S935), the presence / absence of discharge and the number of shots are determined for each nozzle. By the quantization process in this step, a value for each pixel (any of 0 to 15) in an image of 600 dpi × 600 dpi is obtained.

  In step S935, the CPU 201 executes an index expansion process based on the ejection level derived in step S934. The index expansion process is a process of expanding one pixel in a 600 dpi × 600 dpi image into a 4 × 4 pixel bitmap pattern in a 2400 dpi × 2400 dpi image. Specifically, a bitmap pattern is created by determining the pixel value of 4 × 4 pixels in an image of 2400 dpi × 2400 dpi based on the value of the ejection level of each ink color possessed by one pixel in an image of 600 dpi × 600 dpi. . The index expansion process may be executed using a known technique. For example, the dot arrangement corresponding to the ejection level may be stored in advance as a table, and the dot arrangement may be determined using the table corresponding to the ejection level derived in step S934. By the index development processing in this step, the final dot placement destination on the paper surface is determined. For example, when the head 204 can arrange dots with a resolution of 2400 dpi × 2400 dpi on the paper surface, it is determined whether or not to arrange dots for each coordinate obtained by dividing the paper surface into a grid of 2400 dpi × 2400 dpi. By this step, an output result of a total of 64 bits corresponding to 16 pixels (that is, 1 tile) having 1 bit and 4 ch values is obtained.

  In step S940, the CPU 201 stores the 64-bit output result obtained in step S924 or step S935 in the buffer on the RAM 203. When stored in the buffer in this step, both the edge area data and the solid area data are information for controlling whether or not ink is ejected to each nozzle of the 2400 dpi head 204 (On / Off). It has the same meaning. Therefore, for the processing after the recording data creation processing in the subsequent step S340, the same processing as that in the conventional printer may be executed.

  In step S950, the CPU 201 increments m.

  In step S960, the CPU 201 determines whether m is larger than the total number of bit patterns. If the result of the determination in step S960 is true, the series of processing ends, while if the result of the determination is false, the process returns to step S910. The above is the content of the image processing in the present embodiment.

  In step S923 or step S933, the same gradation correction may be realized in all printer models by executing the gradation correction process using the correction amount according to the printer model.

  In the above-described example, it is determined which of the edge area and the solid area corresponds to each target tile (target bit pattern), and different processing is executed according to the determination result. However, for all tiles (all bit patterns), processing for the edge region and processing for the solid region are executed separately, and the results of these processing are used to determine the image area ( The determination may be made based on the result of the determination of whether the region is an edge region or a solid region. In this case, since it is not necessary to selectively execute image processing, the printer circuit can be prevented from becoming complicated, and the processing time can be shortened because the image processing is performed without waiting for the result of the image area determination. .

<Recording data creation process>
Hereinafter, the recording data creation process (step S340 in FIG. 3) in the present embodiment will be described with reference to FIG.

  In step 1000, the CPU 201 executes a path decomposition process. In the serial head type ink jet printer, when the dot landing accuracy at the time of image formation is low, color unevenness and streaks due to dot landing position deviation cause deterioration in image quality. In order to avoid this, it is effective to form an image by a plurality of head scans (multi-pass method) instead of forming an image by a single head scan (single-pass method). Accordingly, in this step, recording data for printing by the multipass method is created. Any known technique may be used as the path decomposition method.

  Conventional image processing is performed by scanning an image in the raster direction and placing the processing result on the RAM in the same direction. However, when recording data is transmitted to the head 204, it is necessary to rearrange the images in the direction (for example, the column direction) received by the head 204. Accordingly, in step 1010, the CPU 201 executes HV conversion processing for converting an image arranged in the raster direction into an image arranged in the column direction. Note that the HV conversion process may be performed by memory access to the RAM 203. Alternatively, the image arranged in the raster direction is directly input by burst transfer via dedicated hardware in order to speed up the processing, rearranged by the SRAM on the hardware, and the rearranged result is directly used as the head 204. You may make it transfer directly to. The above is the content of the recording data creation process in the present embodiment.

<About processing to form an image>
Hereinafter, the process of forming an image in the present embodiment (step S350 in FIG. 3) will be described with reference to FIG. FIG. 11 is a schematic view of a mechanism for forming an image by ejecting ink droplets onto a sheet in a general serial head type ink jet printer. In FIG. 11, the x direction represents the scanning direction of the carriage 1105, and the y direction represents the conveyance direction of the paper 1107.

  As shown in the figure, the inkjet printer includes a paper feed shaft 1101 that feeds paper as a paper feed mechanism and a paper discharge shaft 1102 that discharges paper, and the paper 1107 is rotated by rotating these shafts. Transport at a constant speed.

  An ink tank 1104 that stores inks of CMYK colors is installed on a carriage 1105, and the ink is supplied to the ejection head 1106 through a flow path in the carriage 1105. The carriage 1105 is mounted on the transport rail 1103 and can move along the transport rail 1103. The ejection head 1106 is connected to the printer bus 207 via a communication cable (not shown), and receives print data derived by the above-described HV conversion and head ejection control information for controlling the timing of ink ejection. . An image is formed on the paper by ejecting ink based on the recording data and the head ejection control information. Note that either a thermal method or a piezo method may be employed as a method by which the discharge head discharges ink droplets onto the paper.

  On the head, for example, 512 nozzles are arranged in a row at 600 dpi intervals corresponding to each ink color, and the ink jet printer can independently control ON / OFF of ejection for each nozzle.

  The serial head type ink jet printer controls both the movement of the carriage and the conveyance of the paper, thereby moving the head to an arbitrary position on the paper and ejecting ink. For example, when printing on A4 size (8.5 inches wide × 11 inches long) paper by the single pass method, by ejecting ink while moving the carriage in the x direction, the width is 8.5 in one head scan. An image is formed on an area of inch × length 0.85 inch. Next, the sheet is conveyed 0.85 inches in the y direction and then the head scanning is performed again, thereby forming an image on the next area of 8.5 inches wide × 0.85 inches long. As described above, by repeating the head scanning and the sheet conveyance, the image forming process for the A4 size sheet is completed by the 13-time head scanning in the single pass method. In the case of printing in the multi-pass format, image formation is performed by increasing the number of head scans by shortening the sheet conveyance distance for each head scan, compared to the case of image formation in the single-pass format. For example, when printing on A4 size paper in four passes, the paper transport distance for each head scan is set to 0.21 inches, and the image forming process is completed by 52 head scans. The above is the content of the process for forming an image in the present embodiment.

<About the effects of this embodiment>
In this embodiment, multivalued color information is expressed by a bit pattern of a predetermined format for both the edge region and the solid region in the image (FIG. 6B, FIG. 7A). This bit pattern does not depend on the characteristics of the printer (such as the ink configuration used in the printer or the output characteristics of the head), so there is no need to consider the output destination printer when creating the bit pattern. Useful. In particular, in commercial printing, it is necessary to maintain a high operation rate of the apparatus by connecting a plurality of printer engines in parallel via a network and allocating print jobs evenly. For that purpose, it is important whether the rendering process (step S310) and the conversion process (step S320) can be executed regardless of the printer.

  Further, RIP processing for creating a raster image by rendering processing based on PDL data is generally performed using a dedicated RIP server. By using the RIP server as a common resource for all printers without being linked to a specific printer, it is possible to always execute processing with the maximum performance regardless of the operating status of individual printers.

  Furthermore, when the printer designated as the output destination cannot be used due to a failure or the like, in order to minimize downtime, it is necessary to immediately transmit the data after RIP processing to another usable printer. It is done. Even in such a system, by adopting the format of the present embodiment as the data format after the RIP processing, it becomes possible to share the data after the RIP processing in each printer, and the productivity can be improved.

  In the present embodiment, the case has been described in which the image area that prioritizes the shape reproduction ability over the color reproduction ability is an edge area that specifies the foreground color and the background color. However, a bit pattern corresponding to such an image area is used. The number of colors specified by is not limited to two. When the shape information of the bit pattern designates which of the three or more pixels each pixel is, the number of colors designated by the color information of the bit pattern is 3 or more. That is, in the bit pattern, the number of colors corresponding to the type of pixel specified by the shape information is specified, but this number may be smaller than the total number of pixels in the image area, which is a unit of shape information, and larger than one.

  In the present embodiment, the case has been described in which the number of gradations of both the foreground color and the background color specified by the bit pattern of the edge region is smaller than the number of gradations of the color specified by the bit pattern of the solid region. However, it is sufficient that the number of gradations of at least one of the foreground color and the background color specified by the bit pattern of the edge region is smaller than the number of gradations of the color specified by the bit pattern of the solid region.

  Furthermore, in the present embodiment, the case has been described in which the image area where the resolution is more important than the gradation is the edge area, and the image area where the gradation is more important than the resolution is the solid area. However, the combination of the image area in which the resolution is more important than the gradation and the image area in which the gradation is more important than the resolution is not limited to the combination of the edge area and the solid area.

[Example 2]
In the conversion process according to the first embodiment, the foreground color is expressed by 3 bit data of 1 bit and 3 ch and the background color is expressed by 12 bit data of 4 bit and 3 ch by the color reduction process for the edge region (step S432) (FIG. 6B). )).

  On the other hand, in this embodiment, the size of a tile to be processed is enlarged, and a conversion process is performed in which an 8 × 8 pixel area in an image having a resolution of 2400 dpi × 2400 dpi is handled as one tile. By the conversion process of this embodiment, both the edge area of one tile and the solid area of one tile are expressed by a 128-bit bit pattern. As a result, the foreground color and the background color of the edge region can be expressed in full color (30 bits). In the following, differences from the first embodiment will be mainly described, and the description of the same configuration and the same processing as in the first embodiment will be omitted as appropriate.

<About conversion processing>
Hereinafter, the conversion process in the present embodiment will be described with reference to FIG.

  In step S400, the CPU 201 executes an edge detection process that applies an edge filter to the bitmap image derived in step S310. In this embodiment, as in the first embodiment, a 3 × 3 filter shown in FIG.

  In step S410, the CPU 201 initializes n, that is, sets 1 to the value of n. Here, n is a parameter representing a target tile that is a processing target area (an area of 8 × 8 pixels in an image of 2400 dpi × 2400 dpi in this embodiment).

  In step S420, the CPU 201 derives the edge amount of the attention area in the edge image derived in step S400. In the present embodiment, when the edge amount of an arbitrary region of interest is larger than a predetermined threshold Th2, the tile of interest is determined to be an edge region.

  In step S430, the CPU 201 determines whether the edge amount of the attention area is larger than the threshold value Th2. If the result of the determination in step S430 is true, the target tile is determined to be an edge region, and the process proceeds to step S431. On the other hand, if the result of the determination in step S430 is false, the target tile is determined to be a solid region, and the process proceeds to step S435.

  Hereinafter, a case where the tile of interest is determined to be an edge region (YES in step S430) will be described. In step S431, the CPU 201 performs a quantization process on the target tile in the full-color image derived in step S310, and derives a bitmap pattern expressed by only two colors, the foreground color and the background color. In the derivation of the bitmap pattern in this step, an arbitrary method such as adaptive binarization processing may be used. By this step, shape information (64-bit data) indicating which of the 8 × 8 pixels constituting the target tile corresponds to the foreground pixel and which of the 8 × 8 pixels corresponds to the background pixel is acquired. This shape information is stored in the last 64-bit area “map — 8 × 8” of the 128-bit bit pattern shown in FIG.

  In step S432, the CPU 201 executes a color reduction process for reducing the number of gradations for each of the two colors (16 bits and 3ch, respectively) used in the bitmap pattern derived in step S431. Specifically, both the foreground color and the background color are represented by 10 bits and 3 channels of 30 bit data in which each RGB channel has a value of 10 bits. The RGB values of the foreground color after the color reduction processing are stored in the bit pattern areas “fg_r10”, “fg_g10”, and “fg_b10” shown in FIG. Stored in “bg_g10” and “bg_b10”. As described above, in the first embodiment, the foreground color is expressed by 3 bit data and the background color is expressed by 12 bit data by the color reduction processing on the input 16 bit, 3 ch color information. Therefore, in the first embodiment, the background color can be expressed in full color, but the foreground color cannot be expressed in full color. On the other hand, in this embodiment, both the foreground color and the background color after the color reduction process can be expressed in full color (in 30-bit data).

  Next, a case where the target tile is determined to be a solid area (NO in step S430) will be described. In this case, in step S435, the CPU 201 sums the pixel value of each pixel for each 4 × 4 pixel region (subtile) obtained by dividing the target tile (8 × 8 pixel region) in the full color image derived in step S310. To obtain a total value for each subtile. As a result of summing up 16 pixel values of pixels having a value in the range of 0 to 65535 per channel in the original image, the subtile has a value in the range of 0 to 1048560 per channel. As a result of this step, data for one subtile that can be expressed in 20 bits and 3 channels is acquired for four subtiles. This is synonymous with acquiring a 20-bit, 3-channel value as a pixel value corresponding to each pixel in a 2 × 2 pixel region in a 600 dpi × 600 dpi image.

  In step S436, the CPU 201 performs a color reduction process for reducing the number of gradations for each of the RGB values (each 20 bits) for each subtile derived in step S435, and derives 10-bit RGB values. The R values of the four subtiles derived in this step are stored in 10-bit areas “a11_r10”, “a10_r10”, “a01_r10”, and “a00_r10” of the bit pattern shown in FIG. Similarly, the G values of the four subtiles are stored in 10-bit areas “a11_g10”, “a10_g10”, “a01_g10”, and “a00_g10” of the bit pattern shown in FIG. Similarly, the B values of the four subtiles are stored in 10-bit areas “a11_b10”, “a10_b10”, “a01_b10”, and “a00_b10” of the bit pattern shown in FIG. The four sets of 10-bit, 3-channel information derived in steps S435 to S436 are handled as information of 2 × 2 pixels in a 600 dpi × 600 dpi image in subsequent steps. The above is the content of the conversion process in the present embodiment.

  As described above, the data size of the bit pattern representing the edge area (FIG. 12) and the bit pattern representing the solid area (FIG. 13) are both 128 bits. By having different information with the same data size in this way, it is possible to easily perform random access in the assigned format unit and improve the efficiency of data processing.

  In the conversion process of this embodiment, the color of the edge region is expressed with a low resolution in exchange for expressing the shape with a high resolution. Specifically, foreground and background color information is shared in units of one pixel in a 300 dpi × 300 dpi image. On the other hand, the solid color can be expressed with a higher resolution than the edge region (in units of one pixel in a 600 × 600 dpi image) in exchange for expressing the shape with a low resolution. That is, in the present embodiment as well as in the first embodiment, the shape reproduction capability is given priority over the color reproduction capability in the edge region, the color reproduction capability is given priority over the shape reproduction capability for the solid region, and the edge region and the solid region are separated. It is expressed as a bit pattern with the same data size.

  In this embodiment, the case where both the foreground pixel color and the background pixel color are expressed by 30-bit data has been described. However, if the foreground pixel color and the background pixel color are each expressed in full color. Well, the data used is not limited to 30-bit data. That is, the bit length of the bit area allocated to the foreground pixel color and the background pixel color may be appropriately changed according to the bit length of the bit pattern to be used.

[Example 3]
In the first embodiment, in the conversion process, the edge area and the solid area are expressed by bit patterns having different formats (FIGS. 6B and 7A). On the other hand, in the present embodiment, in the conversion process, the solid area is expressed by the same bit pattern (FIG. 6B) as the edge area. That is, the image data is expressed by only the bit pattern shown in FIG. 6B without using the bit pattern shown in FIG.

  Specifically, a 4 × 4 pixel region in a 2400 dpi × 2400 dpi image is handled as one tile. When it is determined that the tile to be processed (target tile) is a solid area, the solid color is expressed in 10 bits and 3 ch in the first embodiment (FIG. 7A), but in this embodiment, 4 bits. , Expressed in 3ch. This is synonymous with expressing a solid color with 4 bits and 3 ch (RGB each ch with 16 gradations) in a 1 × 1 pixel area in a 600 dpi × 600 dpi image. In this embodiment, the gradation property is ensured by executing a color reduction process (color reduction process to 16 gradations) for deriving a solid color of 4 bits and 3 channels using a dither method. In the following, differences from the first embodiment will be mainly described, and the description of the same configuration and the same processing as in the first embodiment will be omitted as appropriate.

  Regarding human visual characteristics against periodic changes in luminance (Visual Transfer Function), it is known that a slight luminance change can be identified in a low-frequency image, whereas a high luminance image can be identified only by a large luminance change. Yes.

  FIG. 14 shows the relationship (VTF curve) between sensitivity and frequency, which is the ability to discriminate human gradation changes. If the number of gradations in the image exceeds this VTF curve, it can be said that the number of gradations is sufficient.

  Hereinafter, a case where 600 dpi × 600 dpi image data (each pixel is expressed by 16 gradations) is created by dithering will be described as an example. Here, one pixel in the image data of 300 dpi × 300 dpi corresponds to 2 × 2 pixels in the image data of 600 dpi × 600 dpi. One pixel in 150 dpi × 150 dpi image data corresponds to 4 × 4 pixels in 600 dpi × 600 dpi image data. Since each pixel in the image data of 600 dpi × 600 dpi can represent 16 values of 0 to 15, one pixel in the image data of 300 dpi × 300 dpi uses 0 to 60 (= 15 ×) using four pixels in the image data of 600 dpi × 600 dpi. 4) 61 values can be expressed. Similarly, one pixel in 150 dpi × 150 dpi image data can represent 241 values from 0 to 240 (= 15 × 16) using 16 pixels in 600 dpi × 600 dpi image data. To generalize the above-mentioned contents, when 600 dpi is used as a reference, Xdpi can express 15 × (600 / X) ^ 2 + 1 gradations. In other words, in the area gradation, a finer gradation expression can be achieved in exchange for a decrease in resolution.

  Based on the above, FIG. 15 is a plot of the relationship between the area gradation value by dither and the frequency in FIG. As shown in FIG. 15, the value of the area gradation by dither exceeds the number of gradations that can be identified by humans at any resolution (the number of gradations indicated by the VTF curve). It can be seen that sufficient gradation can be obtained by using. By utilizing this fact, only the format shown in FIG. 6B is used in this embodiment.

<About conversion processing>
Hereinafter, the conversion process (step S320 in FIG. 3) in the present embodiment will be described with reference to FIG. In the present embodiment, only the color reduction processing in step S436 is different from the first embodiment.

  In step S436, the CPU 201 performs a color reduction process for reducing the number of gradations for each of the RGB values (20 bits each) derived in step S435, and derives 4-bit RGB values. Here, in this embodiment, the color reduction process is executed using a dither method. Instead of executing the color reduction process using the dither method, the color reduction process may be executed using the error diffusion method. The reason is as follows.

  In the first embodiment, the color of the solid area is expressed with a sufficient number of bits (10 bits, 3 ch) (FIG. 7A). On the other hand, in this embodiment, the color of the solid area is expressed by 4 bits and 3 ch (FIG. 6B). If the color reduction process for deriving the color information of the solid area is executed by using the dither method, there is a possibility that an adverse effect may occur. For example, when the quantization process in the subsequent step S924 is executed using the dither method, the dithering process is executed a plurality of times for the same region, so that the graininess is caused by interference of the dither pattern. Gets worse. Therefore, the occurrence of interference can be avoided by executing the color reduction process using the error diffusion method. Alternatively, when the dither method is used, the pattern may be set in advance so that the dither pattern used in the color reduction process and the dither pattern used in the quantization process do not interfere with each other.

  The 4-bit and 3-channel color information derived by the color reduction processing is stored in a 12-bit area “bg_rgb” area of the bit pattern shown in FIG. Further, 0 is stored as a value indicating that the pixel is a background pixel for all the bits of the region “map — 4 × 4”. The area “fg_rgb” stores foreground color information in the first embodiment, but is not referred to in the present embodiment, and may store an arbitrary value such as an indefinite value.

  According to the present exemplary embodiment, by creating a bit pattern similar to the edge region for the solid region, it is not necessary to execute the processing in steps S931 to S935 in FIG. 9, and thus the configuration of the image processing apparatus in the printer is simplified. Can be

[Other Examples]
The present invention supplies a program that realizes one or more functions of the above-described embodiments to a system or apparatus via a network or a storage medium, and one or more processors in a computer of the system or apparatus read and execute the program This process can be realized. It can also be realized by a circuit (for example, ASIC) that realizes one or more functions.

200 Image forming apparatus 201 CPU

Claims (13)

An image processing apparatus for converting image data in a bitmap format into data composed of a bit pattern including a first bit pattern and a second bit pattern,
First creation means for creating the first bit pattern in which color information of the first image area is stored based on a pixel value of a pixel in the first image area to be reproduced at a first resolution;
The color information of the second image area having a data amount larger than the color information of the first image area based on the pixel value of the pixel in the second image area reproduced with the second resolution lower than the first resolution. An image processing apparatus comprising: a second creation unit that creates the second bit pattern in which is stored.   The image processing apparatus according to claim 1, wherein a bit length of the first bit pattern is equal to a bit length of the second bit pattern.   The data amount of the data composed of a bit pattern including the first bit pattern and the second bit pattern is smaller than the data amount of the image data in the bitmap format. An image processing apparatus according to 1.   4. The image processing apparatus according to claim 1, wherein the first image area is an edge area, and the second image area is a solid area. 5. .   The image processing apparatus according to claim 4, further comprising a determination unit that determines which of the edge area and the solid area is for each image area of a predetermined size in the image data. In the first bit pattern, shape information indicating whether each pixel in the first image area corresponds to a foreground pixel or a background pixel is further stored.
In the color information of the first image area, the color of the foreground pixel and the color of the background pixel are designated,
6. The image processing apparatus according to claim 4, wherein a solid color of the solid area is designated by color information of the second image area.   The image processing apparatus according to claim 6, further comprising a shape information deriving unit that performs a quantization process on a pixel value of each pixel in the edge region and derives the shape information. First color information deriving means for deriving color information of the edge region;
The image processing apparatus according to claim 7, further comprising: second color information deriving means for deriving color information of the solid area. The quantization process is a binarization process,
The image processing apparatus according to claim 8, wherein the first color information deriving unit executes a color reduction process for reducing the number of color gradations.   The number of gradations of the solid color is greater than at least one of the number of gradations of the color of the foreground pixel and the number of gradations of the color of the background pixel. The image processing apparatus according to item. In the second bit pattern, a solid color for each image area obtained by dividing the second image area is designated,
The image processing apparatus according to claim 6, wherein the color of the foreground pixel and the color of the background pixel are expressed in full color. An image processing method for converting image data in a bitmap format into data composed of a bit pattern including a first bit pattern and a second bit pattern,
Creating the first bit pattern storing color information of the first image area based on pixel values of pixels in the first image area to be reproduced at a first resolution;
The color information of the second image area having a data amount larger than the color information of the first image area based on the pixel value of the pixel in the second image area reproduced with the second resolution lower than the first resolution. Creating the second bit pattern in which is stored. An image processing method comprising:   A program for causing a computer to function as the image processing apparatus according to any one of claims 1 to 11.