JP6648933B2 - Image processing device, image forming device, image processing method, and program. - Google Patents

Description

  The present invention relates to a technique for improving jaggies at edge portions in image data.

  2. Description of the Related Art Conventionally, in an image forming apparatus, there have been proposed some techniques for improving rattling occurring at an edge portion of a character or the like called jaggy. There are various reasons for the occurrence of jaggies, but the main reasons include rattling of pixels by a low-resolution printer and rattling due to screen processing.

  As a technique for improving the backlash caused by the screen processing, for example, there is a technique of generating correction data from image data before the screen processing and adding the correction data to an edge portion of the image data after the screen processing so as to border the correction data (for example, And Patent Document 1). This is done by judging whether or not the edge portion should be subjected to the edge correction process, and in the case of the edge portion, comparing the correction data with the image data after the screen processing, and outputting the data having the larger value. , Has improved jaggies due to screen processing. Furthermore, with this technique, the rattling of pixels by the former low-resolution printer is also improved at the same time, despite the simple configuration.

JP 2006-295877 A

  The method disclosed in Japanese Patent Application Laid-Open No. H11-157100 combines and outputs correction data composed of half-dot centers with respect to edge portions and image data after screen processing composed of full-dot centers. For this reason, the edge portion is composed of different dots of a half dot and a full dot, and the half dot may be absorbed by a nearby full dot, and the degree of improvement in jaggies at the edge portion may be small. This is because, in the case of the electrophotographic method, at the time of development, a latent image having a shallow correction data is attracted to a deeper latent image at a location adjacent to a deeper latent image of image data after screen processing. It is. As a result, development is difficult in a shallow latent image portion, and the degree of improvement of jaggies in an edge portion is small in some cases.

  An image processing apparatus according to the present invention includes: a determination unit configured to determine an edge of an object from input image data; a unit configured to generate edge correction data that corrects a pixel value of a pixel configuring the edge; Means for generating screen data by performing processing, and image combining means for generating output image data in which screen dots existing at the edge of the screen data are shifted inside the object. .

  ADVANTAGE OF THE INVENTION According to this invention, the jaggy of an edge part can be improved, minimizing image degradation.

FIG. 2 is a schematic block diagram illustrating a configuration of the MFP according to the first embodiment. FIG. 3 is a diagram illustrating details of a printer unit. FIG. 3 is a block diagram illustrating an internal configuration of an image processing unit according to the first embodiment. FIG. 3 is a block diagram illustrating an internal configuration of an edge information generation unit. It is a flowchart which shows the flow of a mixed data conversion process. It is a flowchart which shows the flow of an edge determination process. It is a flowchart which shows the detail of edge determination processing of three or more values. A specific example of the edge determination processing will be described. FIG. 5 is a diagram illustrating a positional relationship between a target pixel and four pixels referred to by an edge direction determination unit. It is a flowchart which shows the flow of an edge direction determination process. 9 shows an example of a look-up table used in edge correction data generation processing. 6 is a flowchart illustrating a flow of screen processing performed by a screen processing unit. FIG. 3 is a diagram illustrating a dither matrix with a phase shifted. It is a figure which compares and explains the screen processing result in a screen processing part, and the screen processing result in a phase shift screen processing part. 6 is a flowchart illustrating a flow of an image combining process according to the first embodiment. FIG. 4 is a diagram illustrating a reference pixel. FIG. 9 is a diagram illustrating an example of a case where edge correction processing is performed without using a result of the phase shift screen processing. FIG. 7 is a diagram illustrating an example of a case where an edge correction process is performed by applying the first embodiment. FIG. 4 is a diagram illustrating an exposure amount of each pixel in an image after the edge correction processing according to the first embodiment. FIG. 13 is a block diagram illustrating an internal configuration of an image processing unit according to a second embodiment. FIG. 13 is a block diagram illustrating an internal configuration of an edge information generation unit according to the second embodiment. 9 is a flowchart illustrating a flow of an image combining process according to the second embodiment. FIG. 13 is a diagram illustrating an example of a case where an edge correction process is performed using the second embodiment. FIG. 7 is a diagram illustrating an example of characteristics of a correction rate table when a predetermined peripheral area is an area of 5 × 5 pixels. FIG. 14 is a diagram illustrating an example of a result of an edge correction process according to the second embodiment.

  Hereinafter, embodiments for carrying out the present invention will be described with reference to the drawings. The configurations shown in the following embodiments are merely examples, and the present invention is not limited to the illustrated configurations.

  In this embodiment, an electrophotographic digital multifunction peripheral (hereinafter, MFP) having a plurality of functions such as copy, print, and facsimile will be described as an example of an image forming apparatus. However, the present embodiment is not limited to this, and can be applied to an apparatus using another process such as an inkjet method.

  FIG. 1 is a schematic block diagram illustrating a configuration of the MFP 100 according to the present embodiment. The MFP 100 includes a scanner unit 101, a controller 102, a printer unit 103, and an operation unit 104.

  The scanner unit 101 optically reads an image of a document and acquires the image as image data.

  The controller 102 includes a CPU, a ROM, and a RAM, and performs predetermined image processing on image data and the like read by the scanner unit 101. The image data subjected to the image processing is stored in the RAM in the controller 102.

  The printer unit 103 forms an image on recording paper by electrophotography according to designated print setting conditions for the image data on which image processing has been performed. In the printer unit 103 of this embodiment, it is assumed that the exposure amount of the laser can be adjusted by PWM, and image data of 4 bits per pixel is input.

  The operation unit 104 is a user interface for the user to perform various operations. The user sets various printing conditions for image data to be printed via the operation unit 104.

  A server 108 that manages image data via a network 106, a personal computer (PC) 107 that instructs the MFP 100 to execute printing, and the like are connected to the MFP 100. When print execution is instructed from the server 108 or the PC 107, the controller 102 rasterizes the image data transmitted from the server 108 or the PC 107, converts the rasterized image data into image data (bitmap data) corresponding to the printer unit 103, and To be stored.

  FIG. 2 is a diagram showing details of the printer unit 103, and shows a mechanism of a color four-color drum for forming an image on a recording medium by an electrophotographic method.

  The bitmap data and the attribute data once stored in the RAM in the controller 102 are transferred to the printer unit 103 after image processing for printing described later is performed again in the controller 102. In the printer unit 103, the pulse signal is converted into a pulse signal by the PWM control in the printer unit 103, and recording laser light of four colors of cyan (C), magenta (M), yellow (Y), and black (K) in the laser recording unit. Is converted to Then, the recording laser light is applied to the photoconductor 201 of each color to form an electrostatic latent image on each photoconductor.

  The printer unit 103 performs toner development on each photoconductor with toner supplied from the toner cartridge 202, and the toner image visualized on each photoconductor is primarily transferred to the intermediate transfer belt 203. The intermediate transfer belt 203 rotates clockwise in FIG. 2, and when the recording paper fed from the paper cassette 204 through the paper feed path 205 arrives at the secondary transfer position 206, the recording is started from the intermediate transfer belt 203. The toner image is transferred to the paper. The recording paper on which the image has been transferred is fixed with toner by pressure and heat in a fixing device 207, and is conveyed through a paper discharge conveyance path. Then, the recording paper is discharged to a face-down center tray 208 or a face-up side tray 209. Is done.

  Next, image processing for printing executed in the controller 102 will be described. FIG. 3 is a block diagram illustrating an internal configuration of an image processing unit as a functional unit that performs image processing. That is, each process in the image processing unit 300 described below is realized by the CPU in the controller 102 expanding the control program stored in the ROM into the RAM and executing the control program. The image processing unit 300 according to the present embodiment includes a color correction unit 301, an edge information generation unit 302, a gamma correction unit 303, an edge correction data generation unit 304, a screen processing unit 305, a phase shift screen processing unit 306, and an image synthesis unit 307. It consists of.

  The color correction unit 301 performs a color correction process on image data (bitmap data) acquired from the RAM in the controller 102. Specifically, the image data is converted into image data in a CMYK color space in which the density is expressed by four types of colors (image signals) of CMYK by a color conversion LUT or matrix operation. The converted image data has an 8-bit (0-255) value for each pixel for each color.

  The edge information generation unit 302 generates edge information for each color of CMYK. The edge information is data having a 3-bit value for each pixel. The upper one bit indicates whether or not the pixel is a pixel forming an edge (hereinafter, an edge pixel), and is “1” when the pixel is an edge pixel and “0” when the pixel is a non-edge pixel. The lower two bits indicate the direction of the edge (Edge Direction). The upper part is represented by “00”, the lower part by “01”, the right part by “10”, and the left part by “11”. The generated edge information is referred to when switching processing in an image combining unit 307 described below.

  The gamma correction unit 303 corrects the input CMYK image data using a one-dimensional look-up table so as to have a desired density characteristic when the image is transferred to recording paper (gamma correction processing). ). The image data that has been subjected to the gamma correction processing is sent to the edge correction data generation unit 304, the screen processing unit 305, the phase shift screen processing unit 306, and the image synthesis unit 307.

  The edge correction data generation unit 304 generates correction data of an edge portion (hereinafter, edge correction data) from the input image data. The generated edge correction data is sent to the image synthesis unit 307.

  The screen processing unit 305 performs screen processing on the input image data to generate screen data (halftone image data). The generated screen data is sent to the image combining unit 307.

  The phase shift screen processing unit 306 generates screen data having a phase shift with respect to the input image data by using a dither matrix in which the phase of the dither matrix used in the screen processing unit 305 is shifted. The generated phase shift screen data is sent to the image synthesizing unit 307.

  The image synthesizing unit 307 performs an image synthesizing process described later based on the edge information received from the edge information generating unit 302 and the image data after gamma correction.

<Edge information generation processing>
Next, the edge information generation processing in the edge information generation unit 302 will be described in detail with reference to FIGS.

  FIG. 4 is a block diagram showing the internal configuration of the edge information generation unit 302. The edge information generation unit 302 includes a mixed data conversion unit 401, an edge determination unit 402, an inside determination unit 403, an edge direction determination unit 404, and a determination result integration unit 405. When the image data converted into the CMYK color space is input to the edge information generation unit 302, it is sent to the mixed data conversion unit 401.

  The mixed data conversion unit 401 performs a mixed data conversion process on the input image data in units of a predetermined reference area (here, an area of 5 × 5 pixels) centering on the target pixel, and outputs the CMYK four colors. Is generated at an arbitrary ratio (hereinafter, mixed data). Each pixel in the image data has an 8-bit value. However, in the following processing, up to 8-bit data is not necessary. In this embodiment, the lower 2 bits are discarded and this conversion processing is performed. I have. As a result, the circuit scale can be reduced. The details of the mixed data conversion process will be described later. The generated mixed data is sent to the edge determination unit 402, the inside determination unit 403, and the edge direction determination unit 404.

  The edge determination unit 402 performs a process (edge determination process) of determining whether the target pixel is an edge pixel to which the above-described edge correction data is to be applied. The details of this edge determination processing will be described later. The result of the edge determination processing is sent to the determination result integration unit 405 as an edge determination signal (for example, a 1-bit signal that is “1” for an edge pixel and “0” for a non-edge pixel).

  The inside determination unit 403 performs a process of determining whether the target pixel is inside (inside determination process). Specifically, the maximum value of the pixel values of eight pixels around the target pixel is obtained, and when the pixel value of the target pixel is equal to or larger than the maximum value, the target pixel is determined to be inside. The result of the inner determination processing is sent to the determination result integration unit 405 as an inner determination signal (for example, a 1-bit signal that is “1” when inside, and “0” when not inside).

  The edge direction determination unit 404 performs processing (edge direction determination processing) for determining the direction of an edge based on the target pixel. Specifically, the pixel value of the pixel of interest is compared with the pixel values of the upper, lower, left, and right pixels, and if there is a difference between them, it is determined that there is an edge in that direction. The result of the edge direction determination processing is sent to the determination result integration unit 405 as an edge direction signal.

  The determination result integration unit 405 generates the above-described edge information based on the input edge determination signal, inside determination signal, and edge direction signal. Specifically, first, the upper 1 bit of the edge information is set to “1” when it is determined to be an edge pixel in the edge determination process and is determined to be inside by the inside determination process, and otherwise, Set “0”. Further, for the lower two bits of the edge information, the result of the edge direction determination processing is directly substituted into the lower two bits. Thus, 3-bit edge information is generated for each pixel.

<Mixed data conversion processing>
The details of the mixed data conversion process performed by the mixed data conversion unit 401 will be described.

  FIG. 5 is a flowchart showing the flow of the mixed data conversion process. This mixed data conversion processing is executed for each of all 25 pixels constituting the above-described reference area (area of 5 × 5 pixels) in the CMYK image data.

In step 501, the mixed data conversion unit 401 obtains the product of the K pixel value D _K and the K mixing ratio MR _K to obtain a K pixel value D _K ′. In this case, the mixing ratio MR _K is a value arbitrarily set in the range of 0 to 15. Thus, it is determined whether the mixed data is generated only with the CMY chromatic colors or the mixed data including the achromatic K is generated.

In step 502, the mixed data conversion unit 401 determines whether the K pixel value D _K ′ obtained in step 501 is larger than the _C pixel value DC. If the pixel value D _K ′ is larger than the pixel value D _C , the process proceeds to step 503. On the other hand, if the pixel value D _K ′ is not larger than the pixel value D _C , the process proceeds to step 504.

In step 503, the mixed data conversion unit 401 obtains the product of the K pixel value D _K ′ obtained in step 501 and the C mixing ratio MR _C to obtain a C pixel value D _C ′. Here, the mixture ratio MR _C is also a value arbitrarily set in the range of 0 to 15 similarly to the above-described mixture ratio MR _K, and the mixture data generated by changing the value of the mixture ratio MR _C Can be controlled.

In step 504, the mixed data conversion unit 401 obtains the product of the pixel value DC of _C and the above-described mixing ratio MR _C to obtain the pixel value DC 'of _C.

In step 505, the mixed data conversion unit 401 determines whether the K pixel value _DK ′ obtained in step 501 is larger than the _M pixel value DM. If the pixel value D _K ′ is larger than the pixel value D _M , the process proceeds to step 506. On the other hand, if the pixel value D _K ′ is not larger than the pixel value D _M , the process proceeds to step 507.

In step 506, mixed data conversion unit 401 'obtains the product of the mixing ratio MR _M of M, the pixel value D _M of M' pixel value D _K of K obtained at step 501 obtain. Here, the mixing ratio MR _M also, as in the mixing ratio MR _K above, a value is arbitrarily set in a range of 0 to 15, by changing the value of the mixing ratio MR _M, mixed data generated Can be controlled.

In step 507, mixed data conversion unit 401 obtains a product with mixing ratio MR _M pixel value D _M as M K obtained in step 501, obtaining a pixel value D _M 'of M.

In step 508, the mixed data conversion unit 401 determines whether the K pixel value _DK ′ obtained in step 501 is larger than the _Y pixel value DY. If the pixel value D _K ′ is larger than the pixel value D _Y , the process proceeds to step 509. On the other hand, if the pixel value D _K ′ is not larger than the pixel value D _Y , the process proceeds to step 510.

In step 509, the mixed data conversion unit 401 obtains the product of the K pixel value D _K ′ obtained in step 501 and the Y mixing ratio MR _Y to obtain a Y pixel value D _Y ′. Here, the mixing ratio MR _Y also, similarly to the mixing ratio MR _K above, a value is arbitrarily set in a range of 0 to 15, by changing the mixing ratio MR _Y, Y for the mixed data generated Can be controlled.

In step 510, mixed data conversion unit 401 obtains a product of the mixing ratio MR _Y pixel values D _Y and Y K obtained in step 501 to obtain the pixel values D _Y of Y '.

In step 511, mixed data conversion unit 401, D _C was determined by the processing up to here ', D _M', to find the total value of D _Y ', the number of bits specified a total value obtained by the bit shift amount BS By right shifting by the amount, the mixed data [MIX] for each pixel is obtained.

  In step 512, the mixed data conversion unit 401 determines whether the value of the mixed data [MIX] obtained in step 511 is larger than 63, which is the maximum value of 6 bits. If the value of the mixed data [MIX] obtained in step 511 is larger than 63, the process proceeds to step 513. On the other hand, if the value of the mixed data [MIX] obtained in step 511 is not larger than 63, the process ends. That is, the value obtained in step 511 is determined as the mixed data [MIX].

In step 513, the mixed data conversion unit 401 changes the value of the mixed data [MIX] to 63, which is the upper limit of 6 bits (clipping processing). Normally, MR _C, MRM _, MRY _, MRK _{, and the} bit shift amount BS are set so that [MIX] does not exceed 63. However, even if an incorrect value is set, operation is guaranteed. Such a clipping process is performed.

The above is the content of the mixed data conversion processing. Here, in MR _C is 4, MR _M is 6, MR _Y is 6, MR _K is 0, BS is 4, D _C is 5, D _M is 8, D _Y is 10, D _K is 2, is A specific example of the mixed data conversion process in the case is shown.

First, since MR _K is 0, D _K ′ is 0 (step 501).

Next, since D _K ′ is 0 and D _C is 5, the process proceeds to step 504 (No in step 502).

At step 504, D _C is MR _C in 5 since 4, D _C 'is 20.

Then, since D _K ′ is 0 and D _M is 8, the process proceeds to step S507 (No in step 505).

At step 507, D _M is MR _M in 8 since 6, D _M 'becomes 48.

Then, D _K 'is D _Y is so 10, the process proceeds to step 510 at 0 (No in step 508).

At step 510, D _Y is MR _Y at 10 because 6, D _Y 'is 60.

In step 511, first D _C ', D _M', calculated total value of D _Y '(20 + 48 + 60 = 128), shifts the sum 128 to 4 bits to the right in accordance with BS = 4. As a result, 8 is obtained as the value of the mixed data [MIX].

  Since the value (8) of the mixed data [MIX] obtained in Step 511 is smaller than 63 (No in Step 512), the final value of the mixed data [MIX] is 8.

  Thus, mixed data of 6 bits for each pixel in which each color of CMYK is mixed at an arbitrary ratio is generated. Then, by using this mixed data in the subsequent processing, it becomes possible to perform an edge correction process independent of the color to be processed, and as a result, it is possible to suppress a false color occurring at an edge portion.

  Note that the mixed data conversion process is not limited to the above method, and may be any process that generates mixed data using a plurality of colors.

  In step 511 described above, the mixed data [MIX] is obtained by shifting the total value of the product of the pixel value D of each color and the predetermined mixing ratio MR to the right by the bit shift amount BS. Alternatively, mixed data [MIX] may be obtained by using division instead of right shift.

<Edge determination processing>
Next, details of the edge determination processing performed by the edge determination unit 402 will be described.

  FIG. 6 is a flowchart illustrating the flow of the edge determination process.

  In step 601, the edge determination unit 402 sets the mixed data generated by the mixed data conversion unit 401 out of a total of 9 pixels of 3 pixels in width and 3 pixels in height around the pixel of interest in the above-described reference area. , The largest pixel value (maximum value [MAX]).

  In step 602, the edge determination unit 402 determines, for the mixed data generated by the mixed data conversion unit 401, out of a total of 9 pixels having a width of 3 pixels and a height of 3 pixels centered on the target pixel in the above-described reference area. , The smallest pixel value (minimum value [MIN]).

  In step 603, the edge determination unit 402 obtains a contrast value [CONT] by subtracting the minimum value [MIN] obtained in step 602 from the maximum value [MAX] obtained in step 601.

  In step 604, the edge determination unit 402 compares the contrast value [CONT] obtained in step 603 with the edge determination value [Sub], and determines whether the contrast value [CONT] is larger. Here, the edge determination value [Sub] is a threshold value for determining the edge portion of the object, and an arbitrary value that can determine whether the edge portion is a character or a line edge is set, for example. If the result of determination is that the contrast value [CONT] is greater than the edge determination value [Sub], the flow proceeds to step 605. On the other hand, if the contrast value [CONT] is not larger than the edge determination value [Sub], the process proceeds to step 617.

  In step 605, the edge determination unit 402 adds the maximum value [MAX] obtained in step 601 and the minimum value [MIN] obtained in step 602, and divides the obtained addition value by 2 to obtain an average value. Ask [AVE].

  In step 606, the edge determination unit 402 searches for a maximum-minimum value [MAX_MIN] from a total of 9 pixels of 3 pixels in width and 3 pixels in height around the target pixel in the above-described reference area. Here, the maximum-minimum value [MAX_MIN] is the minimum value among the pixel values equal to or more than the average value [AVE] excluding the maximum value [MAX].

  In step 607, the edge determination unit 402 determines whether the maximum-minimum value [MAX_MIN] is found in the search in step 606. If the maximum-minimum value [MAX_MIN] is found, the process proceeds to step 608. On the other hand, if the maximum-minimum value [MAX_MIN] is not found, the process proceeds to step 609.

  In step 608, the edge determination unit 402 subtracts the maximum-minimum value [MAX_MIN] found in step 606 from the maximum value [MAX] found in step 601 to find a difference value [MAX_DIFF_MIN] from the maximum value.

  In step 609, since the maximum-minimum value [MAX_MIN] does not exist, the edge determination unit 402 sets the difference value [MAX_DIFF_MIN] from the maximum value to “0”.

  In step 610, the edge determination unit 402 searches for a minimum-maximum value [MIN_MAX] from a total of 9 pixels of 3 pixels in width and 3 pixels in height around the pixel of interest in the above-described reference area. Here, the minimum-maximum value [MIN_MAX] is the maximum value among the pixel values less than the average value [AVE] excluding the minimum value [MIN].

  In step 611, the edge determination unit 402 determines whether the minimum-maximum value [MIN_MAX] is found in the search in step 610. If the minimum-maximum value [MIN_MAX] is found, the process proceeds to step 612. On the other hand, if the minimum-maximum value [MIN_MAX] is not found, the process proceeds to step 613.

  In step 612, the edge determination unit 402 obtains a difference value [MIN_DIFF_MAX] from the minimum value by subtracting the minimum value [MIN] obtained in step 602 from the minimum-maximum value [MIN_MAX] found in step 611.

  In step 613, since the minimum-maximum value [MIN_MAX] does not exist, the edge determination unit 402 sets the difference value [MIN_DIFF_MAX] from the minimum value to “0”.

  In step 614, the edge determination unit 402 determines whether none of the above-described maximum-minimum value [MAX_MIN] and minimum-maximum value [MIN_MAX] has been found. If there is neither, the process proceeds to step 615. On the other hand, if there is either, the process proceeds to step 616.

  At step 615, neither the maximum-minimum value [MAX_MIN] nor the minimum-maximum value [MIN_MAX] is found (Yes in step 614), and the contrast value [CONT] is sufficiently large (Yes in step 604). It turns out that. Therefore, the edge determination unit 402 determines that the region is a region where the pixel value changes abruptly (that is, an edge pixel requiring edge correction), and sets the edge determination signal to “1 (ON)”. This process ends.

  In step 616, when there are three or more pixel values in the above-described reference area, the edge determination unit 402 determines whether or not the edge needs to be corrected (three or more edge determination processing). Do. The edge determination process of three or more values is a process for detecting an edge existing in a character, a natural image, or the like deteriorated by a compression process or the like. FIG. 7 is a flowchart showing the details of the ternary or higher edge determination processing. The details will be described below.

  In step 701, the edge determination unit 402 subtracts the difference value [MAX_DIFF_MIN] between the maximum value and the difference value [MIN_DIFF_MAX] from the contrast value [CONT] obtained in step 603, and obtains a value near the average value. The variance [DIFF] indicating the distribution of is obtained.

  In step 702, the edge determination unit 402 determines whether the variance value [DIFF] obtained in step 701 is larger than a first threshold value [DiffTh_1]. The first threshold is a threshold for determining whether the difference between the maximum minimum value and the minimum maximum value is sufficiently large. Therefore, a value representing a signal value difference to be detected as an edge is set as the first threshold value. It is desirable that a value slightly smaller than the edge determination value [Sub] be set as the first threshold. If the variance value [DIFF] is larger than the first threshold value [DiffTh_1], the process proceeds to step 704. On the other hand, if the variance value [DIFF] is not larger than the first threshold value [DiffTh_1], the process proceeds to step 703.

  In step 703, the edge determination unit 402 determines that the area does not have a sharp change in the pixel value because the variance [DIFF] obtained in step 701 is not a sufficiently large value. That is, it is determined that the pixel is a non-edge pixel that does not need edge correction, the edge determination signal is set to “0 (OFF)”, and the process ends.

  In step 704, the edge determination unit 402 determines whether the difference value [MAX_DIFF_MIN] from the above-described maximum value is smaller than a second threshold value [DiffTh_2]. The second threshold value and a third threshold value to be described later are threshold values for determining whether the difference value [MAX_DIFF_MIN] from the maximum value and the difference value [MIN_DIFF_MAX] from the minimum value are sufficiently small. The fact that the difference value [MAX_DIFF_MIN] from the maximum value and the difference value [MIN_DIFF_MAX] from the minimum value are large means that there is a change in a gradation portion or the like. Since a portion where the gradient of the gradation change is steep is not desired to be detected as an edge, the determination is made using the second and third threshold values. It is desirable that the second and third threshold values be sufficiently smaller than the first threshold value. This is because, as the second and third threshold values become larger, a portion having contrast and a variation in signal value is determined as an edge. If the difference value [MAX_DIFF_MIN] from the maximum value is smaller than the second threshold value [DiffTh_2], the process proceeds to step 705. On the other hand, if the difference value [MAX_DIFF_MIN] from the maximum value is not smaller than the second threshold value [DiffTh_2], the process proceeds to step 707.

  In step 705, the edge determination unit 402 determines whether the difference value [MIN_DIFF_MAX] from the minimum value is smaller than a second threshold value [DiffTh_2]. If the difference value [MIN_DIFF_MAX] from the minimum value is smaller than the second threshold value [DiffTh_2], the process proceeds to step 706. On the other hand, if the difference value [MIN_DIFF_MAX] from the minimum value is not smaller than the second threshold value [DiffTh_2], the process proceeds to step 707.

  In step 706, the edge determination unit 402 determines that the pixel value has an abrupt change because both the difference value [MAX_DIFF_MIN] from the maximum value and the difference value [MIN_DIFF_MAX] from the minimum value are sufficiently small. to decide. That is, it is determined that the pixel is an edge pixel that needs edge correction, and the edge determination signal is set to “1 (ON)”.

  In step 707, the edge determination unit 402 determines that the pixel value does not have a sharp change, that is, determines that the pixel value is a non-edge pixel that does not require edge correction, and sets the edge determination signal to “0 (OFF). ", And the process proceeds to step 708.

  In step 708, the edge determination unit 402 determines whether the difference value [MAX_DIFF_MIN] from the maximum value is smaller than a third threshold value [DiffTh_3]. If the difference value [MAX_DIFF_MIN] from the maximum value is smaller than the third threshold value [DiffTh_3], the process proceeds to step 710. On the other hand, if the difference value [MAX_DIFF_MIN] from the maximum value is not smaller than the third threshold value [DiffTh_3], the process proceeds to step 709.

  In step 709, the edge determination unit 402 determines whether the difference value [MIN_DIFF_MAX] from the minimum value is smaller than a third threshold value [DiffTh_3]. If the difference value [MIN_DIFF_MAX] from the minimum value is smaller than the third threshold value [DiffTh_3], the process proceeds to step 710. On the other hand, if the difference value [MIN_DIFF_MAX] from the minimum value is not smaller than the third threshold value [DiffTh_3], the process ends.

  In step 710, the edge determination unit 402 determines that the region is one in which either the difference value [MAX_DIFF_MIN] from the maximum value or the difference value [MIN_DIFF_MAX] from the minimum value is sufficiently small and the pixel value changes sharply. That is, it is determined that the pixel is an edge pixel that needs edge correction, the edge determination signal is set to “1 (ON)”, and the process ends.

  In the above description, each of the edge determination value and the first to third threshold values is described as being one, but a plurality of these values may be prepared and used as needed. . For example, by further referring to the attribute data, if the attribute of the target pixel is an image, the edge determination value and the first threshold are set to be larger than the other attributes, and the second and third thresholds are set to be larger. Switch to each smaller value. Further, when the attribute of the pixel of interest is a character or a line, the edge determination value and the first threshold are switched to smaller values and the second and third thresholds are switched to higher values than the other attributes. Good. By performing such switching, correction is difficult to be performed on image attributes whose colors and shapes are likely to be complicated, and correction is more easily performed on character attributes and line attributes whose colors and shapes are likely to be uniform. Thus, precise control is possible.

  The above is the content of the edge determination processing with three or more values.

  Returning to the description of the flowchart of FIG.

  In step 617, since the contrast value [CONT] is small (No in step 604), the edge determination unit 402 determines that the reference region is a region where there is no sharp change in the pixel value. That is, it is determined that the reference area is a non-edge pixel that does not require edge correction, the edge determination signal is set to “0 (OFF)”, and the process ends.

  The above is the content of the edge determination processing. Here, a specific example of the edge determination processing will be described with reference to FIG. FIG. 8 shows a 3 × 3 pixel image 801 to be detected as an edge and a 3 × 3 pixel image 802 not to be detected as an edge. The images 801 and 802 have the same contrast.

  First, the maximum value [MAX] of both images 801 and 802 is "40" (step 601), and the minimum value [MIN] is both "0" (step 602). Then, the contrast [CONT] of both images 801 and 802 both becomes “40” (step 603).

  If the edge determination value [Sub] is “20”, the process proceeds to step 605 for any image (Yes in step 604), and both “20” are derived as the average value [AVE]. .

  In a subsequent step, in the case of the image 801, there is no maximum-minimum value [MAX_MIN] (No in step 607), so that a difference value [MAX_DIFF_MIN] from the maximum value is “0” (step 609). On the other hand, in the case of the image 802, since the maximum-minimum value [MAX_MIN] exists (Yes in Step 607), “10 (= 40−30)” is obtained as the difference value [MAX_DIFF_MIN] from the maximum value ( Step 608).

  Then, in the case of the image 801, there is no minimum-maximum value [MIN_MAX] (No in step 611), and the difference value [MIN_DIFF_MAX] from the minimum value also becomes “0” (step 613). On the other hand, in the case of the image 802, since the minimum-maximum value [MIN_MAX] exists (Yes in step 611), “18 (= 18−0)” is obtained as the difference value [MIN_DIFF_MAX] from the minimum value ( Step 612).

  By the processing up to this point, since neither the maximum-minimum value [MAX_MIN] nor the minimum-maximum value [MIN_MAX] exists for the image 801 (Yes in step 614), the edge determination signal is set to ON. (Step 615). That is, the image 801 is determined to be an edge portion. On the other hand, for the image 802, since both the maximum-minimum value [MAX_MIN] and the minimum-maximum value [MIN_MAX] exist, the process proceeds to the edge determination processing of three or more values (step 616).

  In the edge determination processing of three or more values for the image 802, first, a variance value [DIFF] is derived (step 701). Since the contrast [CONT] is “40”, the difference value [MAX_DIFF_MIN] from the maximum value is “10”, and the difference value [MIN_DIFF_MAX] from the minimum value is “18”, the variance value [DIFF] is “12”. Become.

  Then, if the first threshold value [DiffTh_1] in step 702 is “16” (No in step 702), the edge determination signal is set to OFF (step 703). That is, it is determined that the image 802 is not an edge portion.

  As described above, in the flowchart of FIG. 6, if the reference area is a simple area having less than three pixel values, it is determined whether or not the edge part needs correction using only the contrast value [CONT]. Like that. However, it is not limited to such an embodiment. For example, regardless of the state in the reference area, it may be determined whether or not the edge needs edge correction based only on whether or not the contrast value in the reference area is sufficiently large. The point is that it suffices if it is possible to appropriately determine whether or not edge correction processing should be performed on the edge portion.

(Edge direction determination processing)
Next, details of the edge direction determination processing performed by the edge direction determination unit 404 will be described.

  FIG. 9 is a diagram illustrating a positional relationship between a target pixel and four pixels referred to by the edge direction determination unit 404. In FIG. 9, the pixel of interest is [tr_pix], the pixel to the left of the pixel of interest is [left_pix], the pixel to the right of the pixel of interest is [right_pix], the pixel immediately above the pixel of interest is [up_pix], and the pixel below the pixel of interest is Neighboring pixels are indicated by [down_pix]. FIG. 10 is a flowchart illustrating the flow of the edge direction determination process.

  In step 1001, the edge direction determination unit 404 compares the pixel value of the target pixel with the pixel value of the pixel immediately above the target pixel, and determines whether the pixel value of the target pixel is larger. If the pixel value of the target pixel is larger, the process proceeds to step 1002. On the other hand, if the pixel value of the target pixel is not larger, the process proceeds to step 1003.

  In step 1002, the edge direction determination unit 404 generates an edge direction signal [Edge_Dir] indicating that the edge is above the pixel of interest. In the present embodiment, the edge direction signal [Edge_Dir] is a 2-bit signal, and “00” is set as a value indicating that the edge is above the pixel of interest. The generated edge direction signal [Edge_Dir] is sent to the determination result integration unit 405, and this processing ends.

  In step 1003, the edge direction determination unit 404 compares the pixel value of the target pixel with the pixel value of the lower neighboring pixel of the target pixel, and determines whether the pixel value of the target pixel is larger. If the pixel value of the target pixel is larger, the process proceeds to step 1004. On the other hand, if the pixel value of the target pixel is not larger, the process proceeds to step 1005.

  In step 1004, the edge direction determination unit 404 generates an edge direction signal [Edge_Dir] indicating that the edge is below the pixel of interest. Specifically, “01” is set in the edge direction signal [Edge_Dir] as a value indicating that the edge is below the pixel of interest. The generated edge direction signal [Edge_Dir] is sent to the determination result integration unit 405, and this processing ends.

  In step 1005, the edge direction determination unit 404 compares the pixel value of the target pixel with the pixel value of the pixel on the right of the target pixel, and determines whether the pixel value of the target pixel is larger. If the pixel value of the target pixel is larger, the process proceeds to step 1006. On the other hand, if the pixel value of the target pixel is not larger, the process proceeds to step 1007.

  In step 1006, the edge direction determination unit 404 generates an edge direction signal [Edge_Dir] indicating that the edge is to the right of the pixel of interest. Specifically, “10” is set in the edge direction signal [Edge_Dir] as a value indicating that the edge is to the right of the target pixel. The generated edge direction signal [Edge_Dir] is sent to the determination result integration unit 405, and this processing ends.

  In step 1007, the edge direction determination unit 404 generates an edge direction signal [Edge_Dir] indicating that the edge is to the left of the pixel of interest. Specifically, “11” is set in the edge direction signal [Edge_Dir] as a value indicating that the edge is to the left of the pixel of interest. The generated edge direction signal [Edge_Dir] is sent to the determination result integration unit 405, and this processing ends.

  The above is the content of the edge direction determination processing. From the edge direction signal, it is possible to determine in which direction the target pixel exists with respect to the edge.

<Edge correction data generation processing>
Next, details of the edge correction data generation processing in the edge correction data generation unit 304 will be described.

  The edge correction data generation unit 304 generates edge correction data with reference to a table or the like (for example, a one-dimensional lookup table: LUT) prepared in advance. Specifically, an output value corresponding to a predetermined input value is determined (converted) with reference to the LUT, and is set as edge correction data. In this embodiment, the edge correction data is generated after converting the input image data from 8 bits to 4 bits. FIG. 11 shows an example of a look-up table used in the edge correction data generation processing. In each of the LUTs shown in FIGS. 11A to 11C, 4-bit (0 to 15) input values and output values are associated with each other on a one-to-one basis. FIG. 11A shows an LUT having a linear characteristic of outputting an input value as it is. FIGS. 11B and 11C show LUTs having nonlinear characteristics according to the characteristics of the printer. FIG. 11B shows an LUT in which the output value is smaller than the input value, and is used when the gradation characteristics of the printer are darker than usual or when the edge correction effect is weakened. FIG. 11C shows an LUT in which the output value is larger than the input value, and is used when the gradation characteristics of the printer are thinner than usual or when the edge correction effect is enhanced.

<Screen processing>
Next, screen processing in the screen processing unit 305 will be described.

  FIG. 12 is a flowchart illustrating the flow of the screen processing performed by the screen processing unit 305. The screen process is a process of converting input image data into 4-bit image data that can be printed by the printer unit 103 by using a preset dither matrix.

  In step 1201, the screen processing unit 305 converts image data into 4-bit screen data by a multivalued dither method using a predetermined dither matrix. The multi-valued dither method using the dither matrix is a well-known technique, and will not be described.

  In step 1202, the screen processing unit 305 outputs the screen data obtained in step 1201 to the image combining unit 307.

  The above is the contents of the screen processing in the screen processing unit 305.

<Phase shift screen processing>
Next, the phase-shifted screen processing performed by the phase-shift screen processing unit 306 will be described in detail. This processing makes it possible to shift the screen dots at the edge to the inside of the object in the subsequent processing in the image synthesis unit 307. Since the flow of the screen processing in the phase shift screen processing unit 306 is the same as the flow of the screen processing in the above-described screen processing unit 305, only the differences will be described below.

  In step 1201 described above, the phase shift screen processing unit 306 converts the image data into 4-bit screen data by a multi-value dither method using the phase shifted dither matrix. Specifically, the phase of the dither matrix used in the screen processing unit 305 is converted into screen data using a dither matrix shifted by one pixel in each of up, down, left, and right. FIG. 13 is a diagram illustrating a dither matrix with a phase shifted. A dither matrix 1301 of levels 1 to 15 shown in FIG. 13A is a dither matrix used in the screen processing unit 305. The dither matrices 1302 to 1305 composed of levels 1 to 15 shown in FIGS. 13B to 13E are dither matrices used in the phase shift screen processing unit 306, and shift the phase of the dither matrix 1301 up, down, left, and right. Each is shifted by one square. That is, FIG. 13B shows a dither matrix shifted by one cell upward, FIG. 13C shows a dither matrix shifted by one cell downward, FIG. 13D shows a dither matrix shifted one cell left, and FIG. This is a dither matrix shifted one cell to the right.

  FIG. 14 is a diagram illustrating a comparison between the screen processing result of the screen processing unit 305 and the screen processing result of the phase shift screen processing unit 306. FIG. 14A shows image data input to both screen processing units, in which a rectangular object 1401 of 12 × 7 pixels exists.

  FIG. 14B shows screen data generated by the screen processing in the screen processing unit 305. The pixel group 1402 is a pixel group converted into a halftone dot.

  FIGS. 14C to 14F show screen data generated by the screen processing in which the phase is shifted by one square in the vertical and horizontal directions in the phase shift screen processing unit 306. 14C to 14F, pixel groups 1403 to 1406 are pixel groups converted into halftone dots, and correspond to the pixel groups 1402 in FIG. 14B, respectively. The pixel group 1403 in FIG. 14C has a phase shifted upward by one cell compared to the pixel group 1402. Similarly, the pixel group 1404 in FIG. 14D has a phase shifted downward by one cell compared to the pixel group 1402. The pixel group 1405 in FIG. 14E has a phase shifted to the left by one cell compared to the pixel group 1402, and the pixel group 1406 in FIG. 14F has a phase shifted to the right by one cell compared to the pixel group 1402. I have.

<Image synthesis processing>
Next, an image combining process performed by the image combining unit 307 will be described. FIG. 15 is a flowchart illustrating a flow of the image combining process performed by the image combining unit 307. The following processing is performed on all the pixels in the input image data (image data after gamma correction).

  In step 1501, based on the edge information input from the edge information generation unit 302, the image synthesis unit 307 determines whether the pixel of interest is an edge pixel. Specifically, referring to the upper 1-bit data of the edge information, if the value is “1”, it is determined that the pixel is an edge pixel, and the process proceeds to step 1502. On the other hand, if the value of the upper 1 bit is “0”, it is determined that the pixel is a non-edge pixel, and the process proceeds to step 1503.

  In step 1502, the image synthesizing unit 307 outputs the pixel value of the pixel of interest in the edge correction data input from the edge correction data generation unit 304 to the printer unit 103 as output image data.

  In step 1503, the image synthesizing unit 307 refers to the edge information input from the edge information generating unit 302, and determines whether an edge pixel exists in a peripheral pixel of the target pixel (eight pixels surrounding the target pixel). I do. If an edge pixel exists in the peripheral pixels, the process proceeds to step 1504. On the other hand, if there is no edge pixel in the peripheral pixels, the process proceeds to step 1513.

In step 1504, the image processing unit 307 compares the pixel value of the target pixel with the pixel value of the predetermined reference pixel, and determines whether the target pixel is a pixel inscribed in the edge pixel (whether the target pixel is a pixel inside the object). judge. Here, the reference pixel is a pixel adjacent to the target pixel and has a positional relationship as shown in FIG. In FIG. 16, a central pixel 1601 indicates a target pixel, and other pixels 1602 to 1609 indicate reference pixels. Then, it is determined according to the following conditions whether the target pixel is a pixel inscribed in the edge pixel.
When the upper one bit of the edge information of the reference pixel 1602 is “1”, if the value obtained by subtracting the pixel value of the reference pixel 1606 from the pixel value of the target pixel 1601 is larger than “0”, the target pixel is an edge pixel Is determined to be a pixel inscribed in.
When the upper one bit of the edge information of the reference pixel 1603 is “1”, if the value obtained by subtracting the pixel value of the reference pixel 1607 from the pixel value of the target pixel 1601 is larger than “0”, the target pixel is an edge pixel Is determined to be a pixel inscribed in.
When the upper one bit of the edge information of the reference pixel 1604 is “1”, if the value obtained by subtracting the pixel value of the reference pixel 1608 from the pixel value of the target pixel 1601 is larger than “0”, the target pixel is an edge pixel Is determined to be a pixel inscribed in.
When the upper one bit of the edge information of the reference pixel 1605 is “1”, if the value obtained by subtracting the pixel value of the reference pixel 1608 from the pixel value of the target pixel 1601 is larger than “0”, the target pixel is an edge pixel Is determined to be a pixel inscribed in.
If the above four conditions are not met, it is determined that the target pixel is not a pixel inscribed in the edge pixel.

  If it is determined that the target pixel is a pixel inscribed in the edge pixel, the process proceeds to step 1505. On the other hand, if it is determined that the target pixel is not a pixel inscribed in the edge pixel, the process proceeds to step 1513.

  In step 1505, the image synthesizing unit 307 refers to the edge information and determines whether the edge is located above the pixel of interest (whether the value of the lower two bits of the edge information is “00”). If the value of the lower two bits is “00”, it is determined that the edge is above the pixel of interest, and the process proceeds to step 1506. On the other hand, if the value of the lower two bits is not “00”, the process proceeds to step 1507.

  In step 1506, the image combining unit 307 combines the output results of the screen processing unit 305 and the phase shift screen processing unit 306 to generate combined data [combData]. In this step, in order to move the screen dot to the inside of the object, screen data whose phase is shifted in the direction opposite to the direction of the edge (here, above) is synthesized. Specifically, the output result [Sc] of the screen processing unit 305 is added with the screen data [PSdownSc] shifted in phase from the output result of the phase shift screen processing unit 306, and a predetermined combination coefficient β To obtain composite data [combData].

  In step 1507, the image synthesizing unit 307 refers to the edge information and determines whether the edge is below the pixel of interest (whether the value of the lower two bits of the edge information is "01"). If the value of the lower two bits is “01”, it is determined that the edge is below the pixel of interest, and the process proceeds to step 1508. On the other hand, if the value of the lower two bits is not “01”, the process proceeds to step 1509.

  In step 1508, the image combining unit 307 combines the output results of the screen processing unit 305 and the phase shift screen processing unit 306 to generate combined data [combData]. In this step, in order to move the screen dot to the inside of the object, screen data whose phase is shifted in the direction opposite to the direction of the edge (here, below) is synthesized. Specifically, the output result [Sc] of the screen processing unit 305 is added with the screen data [PSupSc] whose phase has been shifted upward from the output result of the phase shift screen processing unit 306, and a predetermined combination coefficient β To obtain composite data [combData].

  In step 1509, the image combining unit 307 determines whether or not the edge is to the left of the target pixel (whether the value of the lower two bits of the edge information is “10”) by referring to the edge information. If the value of the lower two bits is “10”, it is determined that the edge is to the left of the pixel of interest, and the flow proceeds to step 1510. On the other hand, if the value of the lower two bits is not “10”, it is determined that the edge is to the right of the pixel of interest, and the flow proceeds to step 1511.

  In step 1510, the image combining unit 307 combines the output results of the screen processing unit 305 and the phase shift screen processing unit 306, and generates combined data [combData]. In this step, in order to move the screen dot to the inside of the object, the screen data whose phase is shifted in the direction opposite to the direction of the edge (here, left) is synthesized. Specifically, the output result [Sc] of the screen processing unit 305 is added with the screen data [PSrightSc] whose phase is shifted to the right from the output result of the phase shift screen processing unit 306, and a predetermined combination coefficient β To obtain composite data [combData].

  In step 1511, the image combining unit 307 combines the output results of the screen processing unit 305 and the phase shift screen processing unit 306 to generate combined data [combData]. In this step, in order to move the screen dot to the inside of the object, the screen data whose phase is shifted in the direction opposite to the direction of the edge (here, right) is synthesized. Specifically, the output result [Sc] of the screen processing unit 305 is added with the screen data [PSleftSc] whose phase has been shifted to the left from the output result of the phase shift screen processing unit 306, and a predetermined combination coefficient β To obtain composite data [combData].

  In step 1512, the image combining unit 307 outputs the combined data [combData] generated for the target pixel to the printer unit 103 as output image data.

  In step 1513, the image synthesizing unit 307 outputs the pixel value of the target pixel in the screen data [Sc], which is the output result of the screen processing unit 305, to the printer unit 103 as output image data.

  The above process is repeated until there is no unprocessed pixel in the input image data. The above is the content of the image synthesis processing. As described above, the output image data (the pixel value of the target pixel) is appropriately switched to any of the edge correction data, the composite data, and the screen data according to the edge information and the image data after the gamma correction. .

  In the present embodiment, the composite data is generated by multiplying the composite coefficient β, but the present invention is not limited to this. For example, the combined value of the processing results of the screen processing unit 305 and the phase shift screen processing unit 306 may be input to a one-dimensional lookup table to generate composite data.

  Lastly, the difference between the case where the present embodiment is applied and the case where the present embodiment is not applied will be described with reference to FIGS.

  FIG. 17 is a diagram illustrating an example of a case where the edge correction processing is performed without using the result of the phase shift screen processing.

  FIG. 17A shows image data input to the screen processing unit 305. As shown in FIG. 14A, there is a rectangular object 1701 of 12 pixels × 7 pixels.

  FIG. 17B is a diagram in which an edge pixel whose upper one bit is “1” is indicated by oblique lines based on edge information (input to the image synthesis unit 307). Edge correction data is applied to the edge pixel 1702, and screen data is applied to the pixel 1703 in the area inside the edge pixel 1702.

  FIG. 17C is a diagram illustrating screen data generated by the screen processing unit 305. The pixel group 1704 is a pixel group converted into a halftone dot.

  FIG. 17D is a diagram illustrating edge correction data input to the image combining unit 307. The pixel value of a pixel indicated by gray is the pixel value of the edge pixel 1702 illustrated in FIG.

  FIG. 17E is a diagram illustrating an example of a result of the edge correction processing according to the related art. In this case, the edge pixel 1702 is compared with the screen data shown in FIG. 17C and the edge correction data shown in FIG. 17D, and the larger value is used as the output value. I have.

  FIG. 17F is a diagram illustrating an example of a result of the edge correction processing according to another conventional technique. In this case, a synthesizing process using the edge correction data shown in FIG.

  FIG. 18 is a diagram illustrating an example of a case where edge correction processing is performed by applying the present embodiment.

  FIG. 18A shows image data input to the screen processing unit 305. As in FIG. 17A, a rectangular object 1801 of 12 pixels × 7 pixels exists.

  FIG. 18B shows an edge pixel 1802 whose upper 1 bit of the edge information (input to the image synthesis unit 307) is “1” by oblique lines, and an inner edge pixel 1803 adjacent to the edge pixel by oblique lattice. FIG. Edge correction data is used for the edge pixel 1802, composite data is used for the inner edge pixel 1803, and screen data is used for a pixel 1804 in the area inside the inner edge pixel 1803.

  FIG. 18C is a diagram illustrating screen data generated by the screen processing unit 305. The pixel group 1805 is an image group converted into a halftone dot.

  FIG. 18D is a diagram illustrating edge correction data input to the image combining unit 307. The pixel value of a pixel indicated by gray is the pixel value of the edge pixel 1802 illustrated in FIG. 18B.

  FIG. 18E is a diagram illustrating combined data generated by the image combining unit 307. The pixel values of the pixels indicated by gray at a plurality of levels correspond to the pixel values of the inner edge pixels 1803 illustrated in FIG. Become.

  FIG. 18F is a diagram illustrating output image data output from the image combining unit 307 as a result of the edge correction processing according to the present embodiment. In this case, the edge correction data of FIG. 18D for the edge pixel 1802, the composite data of FIG. 18E for the inner edge pixel 1803, and FIG. 18 for the pixel 1804 in the inner area of the inner edge pixel. The screen data of (c) is used respectively.

  FIG. 19 is a diagram illustrating the exposure amount of each pixel in the image after the edge correction processing according to the present embodiment.

  FIG. 19A shows the exposure amount with respect to the signal value of the pixel. The pixel value of the pixel 1901 is “4”, which indicates that the exposure amount is small in light gray. The pixel 1902 has a pixel value of “8”, has a larger exposure than the pixel 1901, and is shown in a slightly darker gray. The pixel 1903 has a pixel value of “15”, has a larger exposure than the pixel 1902, and is shown in black.

  FIG. 19B is a diagram illustrating the exposure amount of each pixel in the edge correction processing result (1) according to the related art illustrated in FIG. 17E. In FIG. 19B, the pixel value of the edge portion differs depending on the pixel, and for example, a step occurs between the pixel 1904 (gray) and the pixel 1905 (black). Further, since the half dots are attracted to the full dots due to the characteristics of electrophotography, the steps become more remarkable at the time of forming a latent image, and jaggies cannot be completely removed.

  FIG. 19C is a diagram showing the exposure amount of each pixel in the edge correction processing result (2) according to another conventional technique shown in FIG. 17F. In FIG. 19C, since the pixel values at the edge portion are all the same pixel value “8”, the problem of the step at the edge portion does not occur. However, the screen dots (full dots) existing at the pixel 1905 in FIG. 19B are replaced with half dots of the pixel value “8” in FIG. 19C, and the continuity of the screen dots is lost. Would. For example, the influence of the exposure of the pixel 1907 on the pixel 1908 is lost. That is, by replacing the full dots existing in the edge portion with the half dots by the edge correction data, the continuity of the screen dot is lost just before the edge, and becomes unnatural.

  As is clear from FIGS. 19B and 19C, when the results of the screen processing with the phases shifted from each other are not combined in the vicinity of the edge portion, jaggies cannot be completely removed or an image defect occurs. You can see that.

  FIG. 19D is a diagram illustrating the exposure amount of each pixel in the result of the edge correction processing according to the present embodiment illustrated in FIG. 18F. In FIG. 19D, the edge portions are all the same half dots as in FIG. 19C, and there is no variation in pixel values. In addition, by synthesizing the screen result with the phase shifted with respect to the pixel (inner edge pixel) adjacent to the edge portion inside the object, the screen dot existing at the edge portion can be shifted to the inside of the object. ing. For example, by adding a dot to the pixel 1910, it is possible to make it appear as if a screen dot exists around the pixel 1909. Thereby, the continuity of the screen dots is maintained even at the edge portion.

  According to this embodiment, the results of the screen processing with the phase shifted for the pixels inside the object adjacent to the edge are synthesized, and the screen dots of the replaced edge are rearranged inside the object. This makes it possible to improve jaggies at the edges while maintaining the continuity of the screen dots up to the vicinity of the edges.

  Next, a mode in which screen dots existing in the edge portion are moved to the inside of the object instead of synthesizing the screen result with the phase shifted for pixels adjacent to the edge portion inside the object will be described as a second embodiment. . The description of the parts common to the first embodiment will be omitted or simplified, and the following description will focus on the differences.

  FIG. 20 is a block diagram illustrating an internal configuration of an image processing unit as a functional unit that performs image processing according to the present embodiment. The image processing unit 2000 according to the present embodiment includes a color correction unit 301, an edge information generation unit 2001, a gamma correction unit 303, an edge correction data generation unit 304, a screen processing unit 305, and an image synthesis unit 2002. Among them, a color correction unit 301, a gamma correction unit 303, an edge correction data generation unit 304, and a screen processing unit 305 are the same as those in the first embodiment.

  The edge information generation unit 2001 generates edge information for each color of CMYK. FIG. 21 is a block diagram illustrating the internal configuration of the edge information generation unit 2001 according to the present embodiment. As compared with FIG. 4 of the first embodiment, it can be seen that the edge direction determination unit 404 does not exist. Therefore, the edge information in this embodiment does not include information indicating the direction of the edge. That is, the edge information in the present embodiment is data having a 1-bit value for each pixel, and information “1” when the pixel is an edge pixel and information “0” when the pixel is a non-edge pixel. Become. The generated edge information is referred to when the image combining unit 2002 switches the processing.

  The image synthesizing unit 2002 performs an image synthesizing process described later based on the edge information received from the edge information generating unit 2001 and the image data after gamma correction.

<Image synthesis processing>
Next, details of the image synthesis processing performed by the image synthesis unit 2002 will be described. FIG. 22 is a flowchart illustrating the flow of the image combining process performed by the image combining unit 2002. FIG. 23 is a diagram corresponding to FIG. 18 of the first embodiment, and is a diagram illustrating an example of a case where edge correction processing is performed using the present embodiment. FIG. 23A shows image data input to the screen processing unit 305. As shown in FIG. 18A, a rectangular object 2301 of 12 pixels × 7 pixels exists. FIG. 23B is a diagram in which an edge pixel 2302 whose edge information is “1” (input to the image synthesis unit 2002) is indicated by oblique lines, and an inner edge pixel 2303 adjacent to the edge pixel is indicated by oblique lattice. FIG. 23C is a diagram illustrating screen data generated by the screen processing unit 305. The pixel group 2304 is an image group converted into a halftone dot. FIG. 23D is a diagram illustrating edge correction data input to the image combining unit 2002. The pixel value of a pixel indicated by gray is the pixel value of the edge pixel 2302 illustrated in FIG. FIG. 23E is a diagram showing dot movement data generated by the image composition unit 2002. FIG. 23F is a diagram illustrating the dot correction data generated by the image composition unit 2002. FIG. 23G is a diagram illustrating output image data output from the image combining unit 2002 as a result of the edge correction processing according to the present embodiment.

  In step 2201, the image synthesis unit 2002 determines whether or not the pixel of interest is an edge pixel based on the edge information input from the edge information generation unit 2001. Specifically, when the value of the edge information is “1”, it is determined that the pixel is an edge pixel, and the process proceeds to step 2202. On the other hand, when the value of the edge information is “0”, it is determined that the pixel is a non-edge pixel, and the process proceeds to step 2203.

  In step 2202, the image synthesis unit 2002 outputs the pixel value of the target pixel in the edge correction data input from the edge correction data generation unit 304 to the printer unit 103 as output image data.

  In step 2203, the image synthesis unit 2002 refers to the edge information input from the edge information generation unit 2001, and determines whether or not an edge pixel exists in a peripheral pixel of the target pixel (eight pixels surrounding the target pixel). I do. If an edge pixel exists in the peripheral pixels, the process proceeds to step 2204. On the other hand, if there is no edge pixel in the peripheral pixels, the process proceeds to step 2211.

In step 2204, the image synthesizing unit 2002 compares the pixel value of the target pixel with the pixel value of the above-described reference pixel (both after gamma correction) to determine whether the target pixel is a pixel inscribed in the edge pixel (inside the object). Is determined). Specifically, the determination is made according to the following conditions (see FIG. 16 described above).
When the value of the edge information of the reference pixel 1602 is “1”, if the value obtained by subtracting the pixel value of the reference pixel 1606 from the pixel value of the target pixel 1601 is larger than “0”, the target pixel is included in the edge pixels. It is determined that the pixels are in contact with each other.
When the value of the edge information of the reference pixel 1603 is “1”, if the value obtained by subtracting the pixel value of the reference pixel 1607 from the pixel value of the target pixel 1601 is larger than “0”, the target pixel is included in the edge pixels. It is determined that the pixels are in contact with each other.
When the value of the edge information of the reference pixel 1604 is “1”, if the value obtained by subtracting the pixel value of the reference pixel 1608 from the pixel value of the target pixel 1601 is larger than “0”, the target pixel is included in the edge pixels. It is determined that the pixels are in contact with each other.
When the value of the edge information of the reference pixel 1605 is “1”, if the value obtained by subtracting the pixel value of the reference pixel 1608 from the pixel value of the target pixel 1601 is larger than “0”, the target pixel is included in the edge pixels. It is determined that the pixels are in contact with each other.
If the above four conditions are not met, it is determined that the target pixel is not a pixel inscribed in the edge pixel.

  If it is determined that the target pixel is a pixel inscribed in the edge pixel, the process proceeds to step 2205. On the other hand, if it is determined that the target pixel is not a pixel inscribed in the edge pixel, the process proceeds to step 2211.

  In step 2205, the image synthesis unit 2002 determines whether the upper, lower, left, and right pixels adjacent to the pixel of interest are edge pixels and part of screen dots (whether the pixels constitute screen dots). I do. Specifically, whether the value of the edge information (1 bit) of the upper, lower, left and right pixels adjacent to the target pixel is “1” and whether the pixel has a value of “1” or more in the screen data Is determined. If the result of determination is that the upper, lower, left, and right pixels adjacent to the pixel of interest are edge pixels and pixels that form screen dots (hereinafter, edge dot pixels), the flow proceeds to step 2206. On the other hand, if the upper, lower, left, and right pixels adjacent to the target pixel are not edge dot pixels, the process proceeds to step 2208.

In step 2206, the image composition unit 2002 generates dot movement data. Specifically, the pixel value in the edge correction data input from the edge correction data generation unit 304 is subtracted from the pixel value in the screen data of the edge dot pixel to obtain dot movement data. That is, the dot movement data [DotShiftVal] is obtained by the following equation (1) from the screen data [DotVal] and the edge correction data [EdgeVal] of the edge dot pixel.
DotShiftVal = DotVal-EdgeVAl ・ ・ ・ Equation (1)

  Here, the case where the pixel of interest is the pixel 2305 (one of the pixels shown in gray in the dot movement data) of FIG. The pixel 2306 in FIG. 23A, the pixel 2307 in FIG. 23B, the pixel 2308 in FIG. 23C, and the pixel 2309 in FIG. 23D are located at the same position as the pixel 2305 in FIG. The corresponding pixels are shown.

  First, since the target pixel is not an edge, the process proceeds to step 2203 (No in step 2201; see pixel 2307 in FIG. 23B).

  Next, since there is a pixel 2310 in FIG. 23B as a pixel which becomes an edge around the target pixel, the process proceeds to step 2204 (Yes in step 2203).

  Then, since the pixel value of the pixel 2306 in FIG. 23A is larger than “0”, it is determined that the pixel is inside the edge, and the process proceeds to Step 2205 (Yes in Step 2204).

  Then, the pixel on the left of the pixel of interest is an edge pixel and a screen dot (see pixel 2310 in FIG. 23B and pixel 2311 in FIG. 23C). Generated. Since the pixel value of the pixel 2311 is “15”, DotVal is “15”. Since the pixel value of the pixel 2312 is “8” (see FIG. 23D), the edge correction data [EdgeVal] becomes “8”. Therefore, the dot movement data [DotShiftVal] is “7” which is a value obtained by subtracting “8” from “15”.

    In this way, dot movement data is generated. Alternatively, the dot movement data may be obtained by referring to the screen data around the target pixel and determining a pixel value corresponding to the distance from the peripheral screen dot to the target pixel.

  Returning to the description of the flowchart in FIG.

  In step 2207, the image synthesizing unit 2002 outputs the dot movement data generated for the pixel of interest to the printer unit 103 as output image data.

  In step 2208, the image synthesizing unit 2002 determines whether the pixel of interest is a part of a screen dot. Specifically, it is determined whether or not the target pixel is a pixel having a value of “1” or more in the screen data input from the screen processing unit 305 (a pixel forming a screen dot). If the result of determination is that the pixel of interest is a pixel that constitutes a screen dot, the flow proceeds to step 2209. On the other hand, if the target pixel is not a pixel constituting the screen dot, the process proceeds to step 2211.

In step 2209, the image synthesizing unit 2002 generates dot correction data. The dot correction data [DotFixVal] includes a total value [DotSum] of screen data in a predetermined area (for example, 3 × 3 pixels) including the target pixel, a total value [EdgeSum] of edge correction data in the predetermined area, and From the number of pixels [DotNum] to which the dot correction data is applied, it is obtained by the following equation (2). (Truncation after decimal point)
DotFixVal = (DotSum-EdgeSum) / DotNum ... Equation (2)

  Here, the case where the pixel of interest is the pixel 2313 (one of the pixels shown in gray in the dot correction data) of FIG. A pixel 2314 in FIG. 23A, a pixel 2315 in FIG. 23B, and a pixel 2316 in FIG. 23C indicate corresponding pixels located at the same position as the pixel 2313 in FIG.

  First, since the target pixel is not an edge, the process proceeds to step 2203 (No in step 2201; see pixel 2315 in FIG. 23B).

  Next, since there is a pixel 2317 shown in FIG. 23B at a pixel which becomes an edge around the target pixel, the process proceeds to step 2204 (Yes in step 2203).

  Then, since the pixel value of the pixel 2313 in FIG. 23F is a value larger than “0”, it is determined that the pixel is inside the edge, and the process proceeds to Step 2205 (Yes in Step 2204).

  Then, since there is no pixel that is an edge pixel and forms a screen dot in the upper, lower, left and right pixels adjacent to the target pixel, the process proceeds to step 2208 (No in step 2205).

  Then, since the target pixel forms a screen dot (see the pixel 2316 in FIG. 23C), the process proceeds to step 2209, where dot correction data is generated. Here, when the pixel 2313 (= pixel 2316) is set as the target pixel, the total value [DotSum] of the 3 × 3 pixel screen data is three pixels having the pixel value “15” and the pixel having the pixel value “8”. Is one, so it is “53”. Similarly, the total value [EdgeSum] of the edge correction data in the 3 × 3 pixel is “24” because there are three pixels having the pixel value “8”. Further, the number [DotNum] of the pixels to which the dot correction data in the 3 × 3 pixels is applied (pixels having pixel values other than 0 and not forming an edge in the 3 × 3 pixels) is three (FIG. 23). (F)). Therefore, the dot correction data [DotFixVal] in this case is (53-24) /3=9.66/10.

  Note that “the difference between the total value of the screen data in the predetermined area including the target pixel and the total value of the edge correction data in the predetermined area” is “the number of pixels to which the dot correction data is applied in the predetermined area”. The reason for dividing by is to preserve the density of 3 × 3 pixels centered on the target pixel. Originally, the screen data in FIG. 23C is in a state where the density is stored for the input image in FIG. However, since the edge correction data of FIG. 23E is added to the edge portion, the density increases. Since the edge portion after the image synthesizing process is composed of the edge correction data and the dot correction data, the dot correction data can be calculated by subtracting the edge correction data from the result of the density-preserved screen data. In the present embodiment, the predetermined area is a 3 × 3 pixel area, but may be a 5 × 5 pixel or a 7 × 7 pixel.

  In step 2210, the image synthesis unit 2002 outputs the dot correction data generated for the target pixel to the printer unit 103 as output image data.

  In step 2211, the image synthesis unit 2102 outputs the pixel value of the target pixel in the screen data [Sc], which is the output result of the screen processing unit 305, to the printer unit 103 as output image data.

  The above process is repeated until there is no unprocessed pixel in the input image data. The above is the content of the image synthesis processing according to the present embodiment.

  The method of generating the dot movement data and the dot correction data is not limited to the above. For example, an LUT in which the pixel value of the target pixel after gamma correction is set as an input value, the dot movement data corresponding to each input value and the value of the dot correction data as an output value are prepared in advance, and the dot movement data Or, dot correction data may be generated.

<Modification>
For objects with dense edges, such as small point characters, if dot movement data is generated using the above-described method, the density difference between the densely populated edges and the surrounding edges that are not densely populated. Occurs, causing image degradation. Therefore, a modified example will be described in which the dot movement data is generated based on the above-described second embodiment and also considering the number of edge pixels around the target pixel.

  Since only the content of the dot movement data generation processing in step 2206 is different from that of the second embodiment, only the method of generating dot movement data in the present modification will be described below.

  In step 2206, the image synthesizing unit 2002 subtracts the pixel value of the edge correction data from the pixel value of the screen data of the edge dot pixel, multiplies the value obtained by the subtraction by a predetermined edge correction rate, and performs dot shift. Get the data. Here, the edge correction rate is a value determined by the number of edge pixels in a predetermined peripheral area (for example, a 5 × 5 pixel area) of the target pixel, and is determined by, for example, a one-dimensional LUT (correction rate table). Is done. FIG. 24 is a diagram illustrating an example of the characteristics of the correction rate table when the predetermined peripheral area is an area of 5 × 5 pixels. In the correction rate table having the characteristics shown in FIG. 24, for example, when there are seven edge pixels around the target pixel, the edge correction rate is 80%, and the edge correction rate decreases as the number of edge pixels increases. It turns out that it becomes. That is, in the present modification, the value of the generated dot movement data decreases as the number of edges around the target pixel increases. As a result, it is possible to suppress the occurrence of a large density difference between the portion where the edges are dense and the periphery thereof.

  FIG. 25 is a diagram illustrating an example of a result of the edge correction process in the present embodiment (including the modification).

  FIG. 25A shows image data input to the screen processing unit 305, in which a rectangular object 2501 of 6 × 7 pixels and an irregularly shaped object 2502 slightly smaller than the rectangular object 2501 exist at close positions. are doing. The objects 2501 and 2502 have the same density.

  FIG. 25B is a diagram in which the edge pixels whose edge information value is “1” are indicated by oblique lines. In FIG. 25B, a dashed rectangle 2504 indicates a peripheral region (5 × 5 pixel region) when the pixel 2503 is set as a target pixel. At this time, the number of edge pixels is seven. A dashed rectangle 2506 indicates a peripheral region (5 × 5 pixel region) when the pixel 2505 is set as a target pixel. At this time, the number of edge pixels is 15.

  FIG. 25C is a diagram illustrating screen data generated by the screen processing unit 305. The pixel group 2507 is a pixel group converted into a halftone dot.

  FIG. 25D is a diagram illustrating the edge correction data input to the image synthesis unit 2002, which is a pixel value of an edge pixel indicated by oblique lines in FIG. 25B.

  FIG. 25E is a diagram illustrating dot movement data input to the image composition unit 2002.

  FIG. 25F is a diagram illustrating an example of an edge correction processing result when the second embodiment is applied. Since the pixel 2508 is an edge pixel forming an edge and has a screen dot, a dot is added to the pixel 2509 by the image synthesis processing.

  FIG. 25G is a diagram illustrating an example of the edge correction processing result when the modification is applied. The pixel 2510 is an edge pixel forming an edge and is a screen dot. In the second embodiment, the pixel 2511 to which a dot is added has 15 edge pixels in the peripheral area when the pixel is set as a target pixel, and thus has an edge correction rate of 0% (see FIG. 24). (See the correction rate table.) As a result, the pixel value is “0 (white pixel)”.

  When comparing FIG. 25 (f) with FIG. 25 (g), the density of the entire object 2502 is higher in FIG. 25 (f) than in FIG. 25 (g). As a result, the two objects 2501 and 2502, which should be originally expressed with the same density, have a large density difference in FIG. 25F, but the pixel 2511 is a white pixel in FIG. Therefore, the density difference is small.

  As described above, by generating dot movement data according to the number of edge pixels, it is possible to realize edge correction processing that does not cause image degradation even when edges are dense.

  Further, the edge correction data and the dot correction data may be generated according to the number of edge pixels, similarly to the dot movement data.

  As described above, according to the present embodiment, by moving the screen dots to the inside of the object, it is possible to improve the jaggies at the edges while maintaining the continuity of the screen dots near the edges. Further, since the phase shift screen processing is not performed, it is not necessary to hold a circuit for the screen processing, so that it can be realized at lower cost than in the first embodiment.

(Other Examples)
The present invention is also realized by executing the following processing. That is, software (program) for realizing the functions of the above-described embodiments is supplied to a system or apparatus via a network or various storage media, and a computer (or CPU, MPU, or the like) of the system or apparatus reads the program and reads the program. This is the process to be performed.

Claims (13)

Detecting means for detecting an edge portion of an object included in the input image;
By performing halftone processing on the input image, a halftone processing unit that generates a halftone image having a smaller number of tones than the input image,
The pixel of interest is inscribed in the edge portion, constitutes a screen dot, and constitutes the edge portion in peripheral pixels thereof, and when there is no pixel constituting the screen dot, the pixel in the halftone image Output means for correcting and outputting so that the density of the pixel of interest is reduced,
An image processing apparatus comprising:   2. The output unit according to claim 1, wherein, when the pixel of interest forms the edge portion, edge correction data having a constant density is output as the density of the pixel of interest in the halftone image. The image processing apparatus according to any one of the preceding claims.   The image processing apparatus according to claim 2, wherein the correction is a process of subtracting the constant density of the edge correction data from the density of the target pixel in the halftone image.   The image processing apparatus according to claim 3, wherein the constant density of the edge correction data is determined based on a density in the input image. The output means is based on a value obtained by subtracting a total density value of the predetermined area in the edge correction data from a total density value of a predetermined area including the target pixel and the peripheral pixels in the halftone image. 3. The image processing apparatus according to claim 2 , wherein the correction is performed so that the density in the halftone image is preserved. 6. The image processing apparatus according to claim 1, further comprising an image forming unit configured to form an electrophotographic image based on the corrected halftone image output by the output unit. 7. apparatus. A detecting step of detecting an edge portion of an object included in the input image;
By performing a halftone process on the input image, a halftone process step of generating a halftone image having a smaller number of tones than the input image,
The pixel of interest is inscribed in the edge portion, constitutes a screen dot, and constitutes the edge portion in peripheral pixels thereof, and when there is no pixel constituting the screen dot, the pixel in the halftone image An output step of correcting and outputting so that the density of the pixel of interest is reduced,
An image processing method comprising:   In the output step, when the target pixel forms the edge portion, edge correction data having a constant density is output as the density of the target pixel in the halftone image. 8. The image processing method according to 7.   The image processing method according to claim 8, wherein the correction is a process of subtracting the constant density of the edge correction data from the density of the target pixel in the halftone image.   The image processing method according to claim 9, wherein the constant density of the edge correction data is determined based on a density in the input image. Wherein in the output step, from said density sum of a predetermined area consisting of the pixel of interest and the peripheral pixels in the halftone image, on the basis of the value obtained by subtracting the density sum of the predetermined region at an edge correction data 11. The image processing method according to claim 8 , wherein the correction is performed so that the density in the halftone image is preserved. 12. The image according to claim 7, further comprising an image forming step of forming an electrophotographic image based on the corrected halftone image output in the output step. Processing method.   A program for causing a computer to function as the image processing device according to any one of claims 1 to 6.