JP2016092553A - Image processing apparatus, image processing method, and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To solve a problem that when detecting an edge included image data, since it is required to refer a target pixel and a wide range pixel in its periphery, a lot of memory is required.SOLUTION: An image processing apparatus: converts a first resolution image into a second resolution image having a resolution lower than that of the first resolution; determines whether or not a target pixel is an edge part with reference to the target pixel in the second resolution image and a peripheral pixel of the target pixel; and determines a value of the target pixel determined as an edge part and the pixel of edge part in the first resolution image corresponding to the target pixel based on the ratio of the first resolution and the second resolution.SELECTED DRAWING: Figure 13

Description

  The present invention relates to an image processing apparatus, and more particularly to a technique for detecting an edge portion included in an image.

  In an electrophotographic image forming apparatus, a developing roller to which toner is attached is brought into contact with a photosensitive drum on which a latent image portion has been formed by exposure, so that toner adheres to the latent image portion from the developing roller. A toner image is formed on the body drum. At this time, a phenomenon called an edge effect occurs. The edge effect is a phenomenon in which toner adheres excessively to the edge of a high-density image region when an electric field concentrates on the boundary between a latent image portion and a non-latent image portion formed on a photosensitive drum.

  In order to suppress an increase in toner consumption due to such an edge effect, the image forming apparatus detects an edge based on the density difference between the target pixel of the image data and its surrounding pixels, and the density near the detected edge Is corrected (see Patent Document 1).

JP 2009-118378 A

However, in the image forming apparatus disclosed in Patent Document 1, when performing edge detection included in image data, it is necessary to refer to the pixel of interest and a wide range of pixels around it, and a large amount of memory is required.
An object of the present invention is to detect an edge portion included in image data with high accuracy with a small amount of memory.

  The image processing apparatus of the present invention that solves the above problems has the following configuration.

  Resolution conversion means for converting an image of the first resolution into an image of a second resolution lower than the first resolution, a pixel of interest in the image of the second resolution, and a peripheral pixel of the pixel of interest Determining means for determining whether or not the target pixel is an edge portion; a value of the target pixel determined to be an edge portion by the determination means; and the first resolution and the second resolution Determining means for determining a pixel of an edge portion in the image of the first resolution corresponding to the target pixel based on the ratio.

  According to the present invention, an edge portion included in image data can be detected with high accuracy with a small amount of memory.

Schematic sectional view of the image forming apparatus 101 (A) A diagram for explaining a non-contact development method, (b) a diagram for explaining a contact development method. The figure explaining the mode of the electric field in an electrostatic latent image when an image carrier and a developer carrier are in a non-contact state (A) The figure which shows the toner image in which the edge effect generate | occur | produced, (b) The figure which shows the toner image in which sweeping occurred Diagram for explaining image analysis processing Diagram explaining the mechanism of sweeping Diagram explaining the exposure process The figure which shows that the density of a toner image changes by changing the exposure method The figure which shows an example of the function implement | achieved by ASIC Flow chart showing correction processing Diagram showing edge effect correction parameters Diagram showing edge effect correction parameters Block diagram of the image analysis unit 901 Flow chart of image analysis processing Diagram for explaining resolution conversion processing The figure for demonstrating the pixel for correction Flowchart of correction target edge prediction processing The figure for demonstrating correction object edge prediction processing

[Example 1]
<Outline of image forming apparatus>
The operation of the image forming apparatus 101 will be described with reference to FIG. The image forming apparatus 101 includes a drum-shaped electrophotographic photosensitive member (hereinafter referred to as “photosensitive drum”) 1 that is an image carrier. A charging device 2 such as a charging roller which is a charging unit uniformly charges the surface of the photosensitive drum 1. An exposure device 7 such as a laser beam scanner device or a surface light emitting element as exposure means irradiates the uniformly charged photosensitive drum 1 with an exposure amount of light based on image data. Thus, exposure is performed by a laser beam. An electrostatic latent image is formed on the surface of the photosensitive drum 1 by exposure.

  The exposure device 7 receives the drive signal 71 output from the image calculation unit (image processing device) 9 and irradiates the photosensitive drum 1 with optical information 72 according to the drive signal 71 to form an electrostatic latent image. The exposure control unit 19 outputs a light amount adjustment signal 73 for adjusting the target light amount at the time of exposure to the exposure apparatus 7. As a result, a constant amount of current is supplied to the exposure apparatus 7 and the exposure intensity is controlled to be constant. By adjusting the light amount for each pixel with the target light amount as a reference, or adjusting the light emission time by pulse width modulation, the gradation expression of the image is realized.

  The image calculation unit 9 executes a correction process for reducing the toner consumption according to the physical parameter detected by the detection device 12. In the present embodiment, toner consumption is reduced by suppressing excessive toner adhesion due to edge effects and sweeping. The image calculation unit 9 receives raster data (image data) transmitted from the image scanner or the host computer 8 and executes correction processing so that the toner consumption is reduced. Here, the edge effect refers to a phenomenon in which the developer adheres excessively at the boundary (edge) between the exposed exposed area and the unexposed area on the surface of the photosensitive drum 1. Since the surface potential of the exposed region is different from the surface potential of the non-exposed region, an electric field wraps around these boundaries, and the developer adheres excessively. Further, the sweeping is a phenomenon in which the developer adheres excessively at the rear end portion in the transport direction of the electrostatic latent image. Such excessive adhesion of the developer not only lowers the reproducibility of the image density with respect to the original density but also causes excessive consumption of the developer. Therefore, if excessive consumption of the developer can be suppressed, the developer can be saved.

  The CPU 10 is a control unit that comprehensively controls the entire image forming apparatus 101. For example, the CPU 10 corrects a pixel value of a pixel that may cause a developer edge effect or sweeping out of a plurality of pixels constituting the image data, thereby reducing the developer edge effect or sweeping. Function as. Further, the CPU 10 may function as a specifying unit that specifies a pixel in which the developer becomes excessive due to the edge effect or sweeping of the developer among the plurality of pixels constituting the image data. A part or all of the CPU 10 described below may be executed by the ASIC 18. The storage device 11 has an image memory 111 and stores an LUT 112. For example, the exposure control unit 19 executes APC (automatic light quantity control) on the light source of the exposure apparatus 7 to set a target light quantity. The image memory 111 (RAM) is a storage area (page memory, line memory, etc.) in which image data to be subjected to image formation is expanded. The LUT 112 is a look-up table, and stores exposure amount correction values for reducing edge effects and sweeping. For example, the correction value corresponding to the physical parameter detected by the detection device 12 is read from the LUT 112. As described above, the LUT 112 of the storage device 11 functions as a storage unit that stores the physical parameter and the correction amount of the pixel value in association with each other. However, the LUT 112 stores the physical parameter and the number of pixels in association with each other. Also good. This makes it easier for the CPU 10 to specify a pixel to be corrected.

  The developing device 3 as developing means includes a toner container for storing and storing a developer (hereinafter referred to as “toner”) 13 and a developing roller 14 as a developer carrying member. Here, a non-magnetic one-component toner is used as the toner 13, but a two-component toner or a magnetic toner may be used. The layer thickness of the toner 13 supplied to the developing roller 14 is regulated by a regulating blade 15 that functions as a toner layer thickness regulating member. The regulating blade 15 may be configured to apply a charge to the toner 13. Then, the toner 13 regulated to a predetermined layer thickness and given a predetermined amount of charge is conveyed to the developing region 16 by the developing roller 14. The developing area 16 is an area where the developing roller 14 and the photosensitive drum 1 are close to or in contact with each other, and is an area where toner adhesion is performed. The electrostatic latent image formed on the surface of the photosensitive drum 1 is developed with the toner 13 and converted into a toner image. The toner image formed on the surface of the photosensitive drum 1 is transferred onto the transfer material P by the transfer device 4 at the transfer position T. The toner image transferred onto the transfer material P is conveyed to the fixing device 6. The fixing device 6 fixes the toner image on the transfer material P by applying heat and pressure to the toner image and the transfer material P.

<Development method>
The developing method will be described with reference to FIGS. 2 (a) and 2 (b). Development methods mainly include a jumping development method and a contact development method. The jumping development method is a development region 16 which is the closest part between the developing roller 14 and the photosensitive drum 1 maintained in a non-contact state, and the development applied between the developing roller 14 and the photosensitive drum 1. This is a developing method using voltage. The development voltage is an AC bias voltage on which a DC bias is superimposed. FIG. 2A shows an example of the developing device 3 using the jumping development method. The developing device 3 adopting the jumping developing system has a gap 17 between the developing roller 14 and the photosensitive drum 1 at the developing position. If the gap 17 is too small, leakage from the developing roller 14 to the photosensitive drum 1 is likely to occur, and it becomes difficult to develop the latent image. If the gap 17 is too large, it becomes difficult for the toner 13 to fly to the photosensitive drum 1. Therefore, the gap 17 may be maintained in an appropriate size by the abutment roller rotatably supported on the shaft of the developing roller 14.

  The contact development method is a developing region 16 that is the closest part between the photosensitive drum 1 and the developing roller 14 in contact with each other, and a developing voltage (DC) applied between the developing roller 14 and the photosensitive drum 1. In this method, the toner 13 is developed by bias). FIG. 2B shows an example of the developing device 3 using the contact developing method.

  The photosensitive drum 1 and the developing roller 14 rotate in the forward direction at different peripheral speeds. A DC voltage is applied as a developing voltage between the photosensitive drum 1 and the developing roller 14, and the polarity of the developing voltage is set to the same polarity as the charging potential of the surface of the photosensitive drum 1. The toner 13 thinned on the developing roller 14 is conveyed to the developing region 16 and develops the electrostatic latent image formed on the surface of the photosensitive drum 1.

<Principle of edge effect generation>
The edge effect is that the electric field concentrates on the boundary between the exposed portion (electrostatic latent image) and the non-exposed portion (charged portion) formed on the photosensitive drum 1, so that the toner 13 is excessive on the edge of the image. It is a phenomenon that adheres. According to FIG. 3, since the electric lines of force from the non-exposure parts 301 and 302 around the exposure unit 300 wrap around the edge of the exposure unit 300, the electric field at the edge rather than the center of the exposure unit 300. Strength increases. That is, more toner than the center of the exposure unit 300 adheres to the edge.

  FIG. 4A shows an example of a toner image having an edge effect. An arrow A indicates the toner image conveyance direction (the rotation direction of the photosensitive drum 1, the so-called sub-scanning direction). In the image data that is the source of the toner image 400, the toner image 400 is an image having a uniform density. When the edge effect occurs, the toner 13 concentrates and adheres to the edge portion 402a of the toner image 400. As a result, the image density of the edge portion 402a becomes higher than that of the non-edge portion 401a.

  As described above, in the jumping development method, the edge effect occurs due to the concentration of the electric field on the edge portion. On the other hand, in the contact development method, since the gap 17 is extremely short, an electric field is generated from the photosensitive drum 1 toward the developing roller 14, so that the electric field concentration on the edge portion is alleviated and the edge effect is hardly generated.

<Sweeping principle>
The sweeping that occurs in the contact development method will be described. As shown in FIG. 4B, the sweeping is a phenomenon in which the toner 13 concentrates on the edge of the rear end portion of the image on the photosensitive drum 1. The rear end portion is a rear end portion in the toner image conveyance direction (the rotation direction of the photosensitive drum 1) indicated by an arrow A in the toner image. When sweeping occurs, as shown in FIG. 4B, the density of the trailing edge portion 402b of the toner image 410 becomes higher than the density of the non-edge portion 401b, and the consumption amount of the toner 13 increases. In the contact development method, as shown in FIG. 6, the peripheral speed of the developing roller 14 is set to be the peripheral speed of the photosensitive drum 1 so that the toner on the photosensitive drum 1 has a predetermined height. Is faster than. As a result, the toner 13 can be stably supplied to the photosensitive drum 1, and the image density is maintained at a target density. In the developing area 16, the electrostatic latent image is developed by the toner 13 conveyed by the developing roller 14. Further, since the developing roller 14 rotates faster with respect to the photosensitive drum 1, the positional relationship on the surface of both of them constantly deviates. When the rear end portion of the electrostatic latent image 600 enters the developing area 16, the toner 13 a on the developing roller 14 is more rotated than the rear end portion 13 b of the electrostatic latent image 600 in the rotation direction from the starting position of the developing area 16. Located on the back side. Thereafter, the toner 13a on the developing roller 14 passes the rear end portion 13b of the electrostatic latent image 600 until the rear end portion 13b of the electrostatic latent image 600 exits the development region 16. Since the toner 13a is supplied to the rear end portion 13b of the electrostatic latent image 600, the development amount of the rear end portion 13b increases. This is the generation mechanism of sweeping.

<Control method of exposure apparatus>
A method for controlling the exposure apparatus 7 will be described with reference to FIG. The exposure control unit 19 includes an IC 2003 having a built-in 8-bit DA converter 2021 and a regulator 2022, and generates and sends a signal for controlling the exposure apparatus 7. The exposure apparatus 7 includes a VI conversion circuit 2306 that converts a voltage into a current, a laser driver IC 2009, and a semiconductor laser LD. The IC 2003 adjusts the voltage VrefH output from the regulator 2022 based on the light amount adjustment signal 73 indicating the drive current of the semiconductor laser LD set by the CPU 10. The voltage VrefH is a reference voltage for the DA converter 2021. When the IC 2003 sets the input data 2020 of the DA converter 2021, the DA converter 2021 outputs the light amount correction analog voltage Va. The VI conversion circuit 2306 converts the light amount correction analog voltage Va into a current value Id and outputs it to the laser driver IC 2009. In FIG. 7, the IC 2003 mounted on the exposure control unit 19 outputs the light amount correction analog voltage Va. However, the DA converter 2021 may be mounted on the exposure apparatus 7 and the light amount correction analog voltage Va may be generated in the vicinity of the laser driver IC 2009.

  The laser driver IC 2009 switches the switch SW according to the drive signal 71 output from the image calculation unit 9. The switch SW performs ON / OFF control of light emission of the semiconductor laser LD by switching whether the current IL is supplied to the semiconductor laser LD or the dummy resistor R1.

<Exposure amount correction method>
FIG. 8A shows an image formed by exposing one pixel with a light amount of 100% with respect to a target light amount. FIG. 8B shows an image formed by exposing the light amount of one pixel to 50% of the target light amount. This can be realized by reducing the exposure intensity to 50% or halving the density (tone value) of the toner image. FIG. 8C shows an image formed by dividing one pixel into N (N is a natural number of 2 or more) subpixels and thinning out some subpixels. This can be realized, for example, by PWM (pulse width modulation) of 100% of the target light amount. For example, it is realized by dividing one pixel into 16 subpixels and driving the semiconductor laser LD so as to expose only odd-numbered subpixels.

<Edge effect correction procedure>
Here, an embodiment will be described in which image data for forming an electrostatic latent image is corrected to reduce the edge effect and reduce the consumption amount of the toner 13. A relationship between a condition such as a physical parameter correlated with the edge effect and a correction value of the exposure amount for reducing the edge effect is obtained in advance through experiments and simulations and stored in the LUT 112.

  A processing method for correcting the edge effect will be described with reference to FIG. In order to reduce the edge effect, the correction process is executed by the CPU 10 or the ASIC 18 of the image calculation unit 9. Here, an example in which the CPU 10 executes the correction process will be described. The exposure intensity is corrected in order to reduce the edge effect and sweeping, but there are two methods for correcting the exposure intensity. The first is a method of correcting the drive signal 71 of the exposure apparatus 7, and the second is a method of correcting the light amount adjustment signal 73. The edge effect correction process is a process of reducing the pixel value of a pixel that can cause a developer edge effect or sweeping out of a plurality of pixels constituting the image data, and this process reduces the edge effect of the developer. can do. The correction process may include, for example, a step of identifying a pixel in which the developer is excessive due to the edge effect of the developer among a plurality of pixels constituting the image data. Further, an image area composed of pixels having a pixel value greater than or equal to a predetermined value is obtained from a plurality of pixels constituting the image data, and pixels within a predetermined number of pixels from the pixels located at the edge of the image area are developed by the edge effect A step of identifying a pixel in which the agent is excessive may be included.

  Image data transmitted from the host computer 8 is stored in the image memory 111.

  Based on the correction width parameter L set by the parameter setting unit 902, the image analysis unit 901 identifies a pixel that can generate an edge effect from a plurality of pixels that form the image data 906 on the image memory 111. . Here, the correction width parameter L is the number of continuous black pixels that need edge effect correction or sweep correction in a high-resolution image. The correction width parameter L is determined in advance regardless of the content of the input image. With reference to the look-up table in FIG. 11, the correction signal ID 907 is output according to the distance from the edge of the pixel region to the pixel where the edge effect can occur.

  By correcting the exposure intensity of the identified pixel, the edge effect is reduced, and the consumption amount of the toner 13 is reduced. The correction width parameter L indicates the number of pixels from the edge of the image area composed of pixels having a predetermined value or more. For example, if the correction width parameter L is 8, each pixel within 8 pixels from the edge of the image area is to be corrected. In the present embodiment, the correction for the edge effect and the correction for sweeping are not performed for the image area having the number of black pixels shorter than the value of the correction width parameter L.

  The image data 904 and the image data 906 are the same data stored on the image memory 111, but the timing at which the image analysis processing is performed by the image analysis unit 901 and the exposure amount correction processing are performed by the exposure amount correction unit 903. The timing is different. Accordingly, the CPU 10 synchronizes and transfers the image so that the pixel position of the correction signal ID output from the image analysis unit 907 matches the pixel position input to the exposure amount correction unit 903.

<Explanation of exposure correction unit>
The exposure amount correction unit 903 corrects the pixel value (exposure amount) for each pixel according to the image data 904 on the image memory 111, the correction signal ID 907, and the exposure amount adjustment parameter 909. Then, based on the corrected pixel value, a drive signal 71 is generated and output to the exposure device 7.

  At this time, the exposure amount correction unit 903 performs correction according to the correction signal ID 907 with reference to a lookup table shown in FIG. Further, when correcting the light amount adjustment signal 73, the light amount correction is performed using the exposure amount adjustment parameter 909 of the exposure amount reduction ratio [light amount] in FIG. 12, and when the drive signal 71 is corrected, the exposure amount reduction in FIG. Correction is performed using the exposure amount adjustment parameter 909 of the ratio [PWM].

  When the drive signal 71 is corrected, as shown in FIG. 8C, the exposure interval is corrected, and the toner amount per pixel is reduced. When the light amount adjustment signal 73 is corrected, as shown in FIG. 8B, the exposure amount is reduced so that the image density per pixel is lowered.

  FIG. 10 is a flowchart showing each step of the correction process executed by the CPU 10. When receiving an instruction to start printing from the host computer 8, the CPU 10 starts processing according to this flowchart. In step S <b> 1201, the CPU 10 acquires the exposure amount adjustment parameter 909 stored in the LUT 112.

  In step S <b> 1202, the CPU 10 specifies a correction target pixel in the image data based on the exposure adjustment parameter 909 acquired from the LUT 112. Also, the distance from the edge portion of the area of the image having a predetermined density or higher is calculated. As a result, the number of pixels to be corrected from the edge of the image area and the exposure adjustment parameter 909 for those pixels to be corrected are determined.

  In step S1203, the CPU 10 corrects the exposure amount of the correction target pixel by correcting the pixel value of the specified correction target pixel using the exposure amount adjustment parameter 909. Here, the example in which the CPU 10 performs the correction process has been described, but the ASIC 18 or the host computer 8 may execute the correction process. As described above, the CPU 10, the ASIC 18, and the host computer 8 function as an image processing apparatus.

<Detailed description of image analysis processing>
Next, image analysis processing performed by the image analysis unit 901 in the present embodiment will be described in detail with reference to FIGS. 13, 14, and 15.

  FIG. 13 is a diagram illustrating a configuration of the image analysis unit 901. FIG. 14 is a flowchart of image analysis processing performed by the image analysis unit 901. FIG. 15 is a diagram for explaining the resolution conversion process. The block width S is a unit for resolution conversion. That is, by converting the pixel value of the block composed of S × S pixels in the input image (high resolution image) 906 into multi-value data of one pixel, the resolution of the input image (high resolution image) 906 is converted into a low resolution image. I do. The block width S is equal to the ratio between the resolution of the high resolution image and the resolution of the low resolution image. For example, when the resolution of the high resolution image is 1200 dpi and the resolution of the low resolution image is 400 dpi, the block width S is “3”.

  First, in step S1401, the resolution conversion unit 1301 performs resolution conversion for reducing the resolution of the high-resolution image (first resolution image) 906, and the low-resolution image after the resolution conversion (second resolution image). Image) is stored in the low-resolution memory unit 1302. The resolution conversion unit 1301 counts the number of black in the binary input image shown in FIG. 15A and converts it to a low-resolution multi-valued image (Mbit) shown in FIG.

  For example, FIG. 15C shows a case where the block width S is 3 and three black pixels exist in the S × S pixel region of the high resolution image. At this time, the resolution conversion unit 1301 counts the number of black pixels in the S × S pixel area, and converts the number into one pixel of a low-resolution image having a pixel value of 3. In order to express an input image of 3 × 3 pixels, a data amount of 9 bits is necessary. However, if one pixel of low-resolution multi-value data is used, it can be expressed by a data amount of 4 bits. That is, the data amount can be suppressed to 4/9.

  Next, in step S1402, the correction target edge prediction unit 1303 predicts an edge in the high resolution image with respect to the low resolution image stored in the low resolution memory unit 1302, and converts the predicted edge signal into a correction signal ID generation unit. 1304.

  Finally, in step S1403, the correction signal ID generation unit 1304 generates a correction signal according to the distance from the edge of the correction target region to the target pixel, and outputs the correction signal to the exposure amount correction unit 903.

<Detailed Description of Correction Target Edge Prediction Unit 1303>
Next, the correction target edge prediction process performed by the correction target edge prediction unit 1303 in the present embodiment will be described in detail with reference to FIGS. 17 and 18.

  FIG. 17 is a flowchart of the correction target edge prediction process performed in the edge detection process. FIG. 18 is a diagram for explaining the correction target edge prediction process performed by the image analysis unit 901. FIG. 18 shows a low-resolution image input to the correction target edge prediction unit 1303. The upper pixel of the image of interest i is defined as a pixel uj. j represents the distance from the pixel of interest. The pixel one pixel above the target pixel i is the pixel u1. The pixel one pixel below the target pixel i is the pixel d1.

  First, a pixel to be corrected will be described with reference to FIGS. 4 and 16. As shown in FIGS. 4A and 4B, the pixels in the image area that are within the predetermined pixels from the edge of the image area having a predetermined density or more are according to the distance from the edge of the image. The exposure amount is corrected. Further, the correction amount may be changed depending on the size of the white area adjacent to the edge of the image area having a predetermined density or more. This is because when another black pixel group is present near the black pixel group to be corrected, it is difficult for toner to adhere excessively.

  Accordingly, the image analysis unit 901 sets the black pixels as the target pixels for correcting the edge effect as shown in FIGS. 16A, 16B, 16C, and 16D. Extract the pixels to be used. As the target pixel for correcting the sweeping phenomenon, as shown in FIG. 16A, pixels in which black pixels are continuous for the length of the correction width parameter L are extracted. Then, the correction amount of the correction target pixel is determined according to the distance from the edge of the image area having a predetermined density or more.

  First, in step S1701, the correction target edge prediction unit 1303 divides the correction width parameter L by the block width S and calculates the reference pixel number n to be referred to in the low resolution image by rounding up the fractional part. Here, ROUNDUP means to round up after the decimal point.

  In step S1702, 1 is substituted for the reference position j and 0 is substituted for the correction stop flag (flag) in order to initialize the variables. From step S1702 to step S1706, the low-resolution pixel above the target pixel i in the low resolution is referred to, and the target pixel i is a pixel including the edge to be corrected in terms of continuity of the black pixel in the sub-scanning direction. Determine whether. In step S1703, it is determined whether the value of the reference pixel (peripheral pixel) uj matches S × S. That is, in step S1703, it is determined whether the value of the reference pixel uj is black. Here, when all the pixels of the S × S pixel block in the high resolution are black, the pixel value of the low resolution pixel is S × S. Whether the reference pixel uj is a black pixel is determined by comparing with S × S. If the reference pixel uj matches S × S, it is determined that all the pixels in the pixel group at the position of the reference pixel uj before the resolution conversion process are black, and the process proceeds to step S1705. If the reference pixel uj does not match S × S, it is determined that the pixel group is not subject to correction because white pixels are mixed in the pixel group at the position of the reference pixel uj before the resolution conversion process, and the process proceeds to step S1704. .

  In step S1704, since the target pixel i is not a correction target, 1 is substituted for the correction stop flag flag, and the process proceeds to step S1705.

  In step S1705, 1 is added to the reference position j in order to move the reference pixel, and the process proceeds to step S1706.

  If the reference position j is larger than the reference pixel number n in step S1706, the process proceeds to step S1707, and if it is smaller, the process proceeds to step S1703. When the reference position j is larger than the reference pixel number n, it is determined that it is out of the reference range.

  In step S1707, the correction target edge prediction unit 1303 proceeds to step S1717 when the correction stop flag flag is 1, and proceeds to step S1708 when the correction stop flag flag is 0. When the correction stop flag flag is 1, it is determined that the target pixel is not a pixel including an edge to be corrected, and the determination is shifted to the next pixel. When the correction stop flag flag is 0, it is determined that the target pixel is a pixel including the edge to be corrected, and the lower part of the target pixel is determined.

  In step S1708, in order to perform initial setting of variables, the correction target edge prediction unit 1303 substitutes 1 for the reference position j, and substitutes 0 for the correction stop flag flag. From step S1708 to step S1712, the low-resolution pixel below the target pixel i in the low resolution is referenced, and the target pixel i is a pixel including the edge to be corrected from the viewpoint of continuity of the white pixel in the sub-scanning direction. Determine whether.

  In step S1709, it is determined whether the reference pixel dj matches zero. If the reference pixel dj matches 0, it is determined that all the pixels in the pixel group at the position of the reference pixel dj before the resolution conversion process are white, and the process proceeds to step S1711. When the reference pixel dj does not match 0, it is determined that the pixel group is not subject to correction because the black pixel is mixed in the pixel group at the position of the reference pixel dj before the resolution conversion process, and the process proceeds to step S1710.

  In step S1710, since the target pixel i is not a correction target, 1 is substituted for the correction stop flag flag, and the process proceeds to step S1711.

  In step S1711, the correction target edge prediction unit 1303 adds 1 to the reference position j in order to move the reference pixel, and the process proceeds to step S1712.

  In step S1712, the correction target edge prediction unit 1303 proceeds to step S1713 if the reference position j is larger than the reference pixel number n, and proceeds to step S1709 if it is smaller. When the reference position j is larger than the reference pixel number n, it is determined that it is out of the reference range.

  In step S1713, the correction target edge prediction unit 1303 proceeds to step S1717 when the correction suspension flag flag is 1, and proceeds to step S1714 when the correction suspension flag flag is 0. When the correction stop flag flag is 1, it is determined that the target pixel is not a pixel including an edge to be corrected, and the determination is shifted to the next pixel. When the correction stop flag flag is 0, it is determined that the target pixel is a pixel including the edge to be corrected.

  In step S1714, it is determined whether or not the following conditional expression is satisfied.

  The left side is the sum of the pixel values of the target pixel i and the reference pixels u1 to u1. That is, it is the sum of pixels of high resolution conversion. The right side is the number of pixels in the high resolution of the block of pixels having the block width S in the main scanning direction and the correction width parameter L in the sub scanning direction. The right side is a block of black pixels to be corrected. If the left side is greater than or equal to the right side, it can be determined that the reference pixels u1 to un including the target pixel are black pixel blocks suitable for correction. Here, it is determined whether black pixels are L pixels continuous at high resolution from the pixel of interest i of the low resolution image and the pixel values of the reference pixels u1 to un. If the conditional expression matches, the process proceeds to step S1715, and if not, the process proceeds to step S1717. In step S1715, the correction target edge prediction unit 1303 determines whether or not the following conditional expression is satisfied.

  The left side is the sum of the pixel values of the target pixel i and the reference pixels d1 to dn. That is, it is the sum of pixels of high resolution conversion. The right side is the number of pixels in the high resolution of the block of pixels having the block width S in the main scanning direction and the correction width parameter L in the sub scanning direction. The right side is a lump of white pixels to be corrected. If the left side is less than or equal to the right side, it can be determined that the reference pixels d1 to dn including the target pixel are white pixel blocks suitable for correction. Here, it is determined whether white pixels are continuous in L pixels at high resolution from the pixel of interest i of the low resolution image and the pixel values of the reference pixels d1 to dn. If the conditional expression matches, the process proceeds to step S1716, and if not, the process proceeds to step S1717. From step S1702 to step S1715, it is determined that a high resolution edge exists in the low resolution pixel of interest i.

  If the determination in S1713 is No without performing the processes in steps S1714 and S1715, the process may proceed to S1716. By performing the processing of step S1714 and step S1715, it is possible to more accurately determine whether or not the pixel of interest i of the low resolution image and the pixels from the reference pixels u1 to un are continuous by L pixels.

  In step S1716, the correction target edge prediction unit 1303 divides the pixel value of the low-resolution target pixel i by the block width S, thereby determining the edge pixel in the high-resolution image of the low-resolution target pixel i, The process proceeds to step S1717. According to the determinations so far, it is known that black pixels are continuous above the pixel of interest i and white pixels are continuous below. Therefore, the pixel value of the pixel of interest i is calculated from the top of the block S × S at a high resolution to calculate the number of pixels from the top on the basis of the upper side of the pixel of interest i.

  In step S1717, the correction target edge prediction unit 1303 determines whether correction target edge prediction processing has been performed for all pixels. If processing has not been performed for all pixels, the process proceeds to step S1702. . If the process has been performed on all pixels, the process ends.

  In the present embodiment, as shown in FIG. 16A, the edge direction in which black pixels are continuous above the target pixel and white pixels are continuous below the target pixel is described as one direction. In addition to FIG. 16A, the image patterns shown in FIGS. 16B, 16C, and 16D are also determined in order to determine the pixel that corrects the edge effect shown in FIG. In FIG. 16A, the continuous black pixels are above the target pixel. However, when the determination of FIG. 16B is performed, the continuous black pixels are below the target pixel. Processing is performed by changing the direction of determination. When the determination of FIG. 16C is performed, the continuation of the black pixels is the right pixel with respect to the target pixel, and therefore, the process is performed by changing the determination direction. That is, from step 1702 to step S1706, it is determined whether or not black pixels are continuous in the main scanning direction with reference to the pixel to the right of the target pixel. Since the processing contents are repeated, the description is omitted.

<Explanation with specific examples>
With reference to FIG. 5, a processing example of pixel analysis that may cause sweeping in the image analysis unit 901 will be described in detail. In the present embodiment, a case where the correction width parameter L is 8 and the block width S is 3 will be described.

  FIG. 5A shows high-resolution binary image data 906 input to the image analysis unit 901. The pixel group 1901 is a cluster of black pixels of 3 pixels horizontally and 8 pixels vertically. The pixel group 1902 is a cluster of black pixels of 3 pixels horizontally and 10 pixels vertically. The pixel group 1903 is a cluster of black pixels of 3 pixels horizontally and 4 pixels vertically.

  FIG. 5B shows image data after being converted into a low resolution image by the resolution conversion unit 1301. Since the number of black pixels at the pixel position 1904 is three, the pixel value of the low resolution pixel 1906 is three. Further, since the number of black pixels at the pixel position 1905 is nine, the pixel value of the low resolution pixel 1907 is nine.

  FIG. 5C shows a correction target edge predicted by the correction target edge prediction unit 1303. When the target pixel is the low-resolution pixel 1908, the reference pixel number n is ROUNDUP (L / S) = ROUNDUP (8/3) = ROUNDUP (2.67) = 3 (step S1701). When the target pixel is the low resolution pixel 1908, the reference pixel u1 is the low resolution pixel 1909, and the reference pixel u2 is the low resolution pixel 1910, which matches S × S = 3 × 3 = 9 (step S1703). The reference pixel d1 is a low resolution pixel 1911 and the reference pixel d2 is a low resolution pixel 1912, which is equal to 0 (step S1709).

  When the pixel of interest is the low resolution pixel 1908, the left side is (un + i) + S × S × (n−1) = (3 + 3) + 3 × 3 × (3-1) = 6 + 18 = 24. The right side is L × S = 8 × 3 = 24. Accordingly, since step S1714 matches, the process proceeds to step S1715. When the target pixel is the low resolution pixel 1908, the left side is (dn + i) + S × S × (n−1) = (0 + 3) + 3 × 3 × (3-1) = 3 + 18 = 21. The right side is L × S = 8 × 3 = 24.

  Accordingly, in step S1715, since the right side is larger than the left side, the process proceeds to step S1716. The position of the predicted edge e is i / S = 3/3 = 1, and it is predicted that the first pixel from the upper end of the target pixel is an edge (step S1716). Therefore, the pixel group of the correction target edge is determined to be the pixel group 1913 from the low resolution pixels 1906, 1908, 1909, 1910, 1911, 1912 in FIG. Similarly, the correction target edge of the pixel group adjacent to the low resolution pixel 1907 is the pixel group 1914. This is because the position of the predicted edge e is i / S = 9/3 = 3, and the third pixel from the upper end of the target pixel is predicted to be an edge. In step S1703, the pixel group adjacent to the low resolution pixel 1915 is determined not to be S × S because the pixel u2 is the pixel 1916, and the process proceeds to step S1704. That is, it is determined that the pixel group adjacent to the low resolution pixel 1915 does not include the correction target edge.

  FIG. 5D shows correction signal ID data generated by the correction signal ID generation unit 1304. The correction signal ID generation unit 1304 receives the image shown in FIG. 5C output from the correction target edge prediction unit 1303, and shows the image shown in FIG. 5D according to the distance from the target pixel and the correction target edge. Output an image. The pixel 1917 outputs 0 as the correction amount signal ID because the distance from the correction target edge is 0 pixel. Since the distance from the correction target edge is 8 pixels, the pixel 1918 outputs 8 as the correction amount signal ID. The pixel 1919 has a distance of 9 pixels from the correction target edge, but since the correction width parameter L is 8, it outputs 0 as the correction signal ID. In FIG. 5D, the correction signal ID is 0 for the pixels on which no number is written. 5C and 5D are image data having the same resolution as the input image (high resolution image) 906. FIG.

  As described above, in order to determine the correction target pixel in the input high-resolution image, a circuit that refers to a wide area is required in the determination process, resulting in high cost. Further, when the resolution of the input high-resolution image is reduced and the correction target pixel is determined for the low-resolution image, the position of the edge in the high-resolution image cannot be accurately specified.

  In the present embodiment, the input high resolution image is reduced in resolution, the correction target pixel is determined in the low resolution image, and the correction target edge in the input high resolution image is predicted, thereby reducing the data amount. It becomes possible to perform edge determination with high accuracy.

  In the present embodiment, the method for predicting the edge position from the pixel value of the pixel of interest i has been described, but it goes without saying that other methods may be used. For example, the edge shape in the high resolution image corresponding to the target pixel i of the low resolution image may be predicted by performing pattern matching with the pixel values around the target pixel i.

<Summary>
According to this embodiment, when the toner consumption reduction process is performed, the input image is reduced in resolution, and the edge of the original input image is predicted from the low-resolution image while determining the correction target in the low-resolution image. . By the above method, it is possible to determine the correction target pixel with high accuracy while reducing the memory, and to obtain a good image.

  The image analysis unit 901 specifies a pixel in which the developer becomes excessive due to the edge effect or sweeping of the developer among a plurality of pixels constituting the image data, and the exposure amount correction unit 903 corrects the value of the specified pixel. May be. The image analysis unit 901 obtains an image area having a pixel value greater than or equal to a predetermined value from among a plurality of pixels constituting the image data, and determines pixels within a predetermined number of pixels from pixels located at the edge of the image area. It may be specified as a pixel in which the developer becomes excessive due to the edge effect. The edge effect and sweeping are easily visible when the optical density of the pixel is greater than a certain value. Further, the edge effect occurs at the edge of the pixel region, and the sweeping occurs at the rear end of the pixel region. Therefore, if the correction target pixel is determined in consideration of these, the edge effect and sweeping can be efficiently reduced.

  In the present embodiment, as an example of the correction process for the pixel value of the edge part, the process of correcting the edge effect and sweeping is described as an example. But you can. For example, processing that suppresses scattering of toner at the edge portion and smoothing processing that suppresses rattling of pixels at the edge portion may be used.

  In the present exemplary embodiment, the correction processing is described as being performed in the image processing apparatus (image calculation unit 9) in the image forming apparatus 101. However, the image processing apparatus is a computer installed outside the image forming apparatus 101. Also good. Further, the CPU 10 may execute the image processing method shown in FIG. 10 by executing a program stored in the storage device 11 to realize the function shown in FIG.

(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.

Claims (8)

Resolution conversion means for converting an image with a first resolution into an image with a second resolution lower than the first resolution;
A determination unit that refers to a target pixel in the image of the second resolution and a peripheral pixel of the target pixel, and determines whether the target pixel is an edge;
The image of the first resolution corresponding to the target pixel based on the value of the target pixel determined to be an edge portion by the determination unit and the ratio between the first resolution and the second resolution. An image processing apparatus comprising: determining means for determining a pixel at an edge portion of the image processing apparatus.   The image processing apparatus according to claim 1, wherein the determining unit determines the pixel of the edge portion by dividing the value of the target pixel by the ratio.   The image processing apparatus further includes a correcting unit that performs a process of reducing a pixel value on a pixel located within a predetermined pixel from a pixel of the edge determined by the determining unit in the first resolution image. The image processing apparatus according to claim 1, wherein the image processing apparatus is an image processing apparatus.   The image processing apparatus according to claim 3, wherein the correction unit performs a process of reducing the pixel value by using an adjustment parameter determined according to a distance from the pixel of the edge portion.   The resolution conversion unit is configured to select the second resolution corresponding to the area according to the number of black pixels in the area composed of S × S pixels (S is a natural number of 2 or more) in the first resolution image. 4. The image processing apparatus according to claim 1, wherein an image having the first resolution is converted into an image having the second resolution by determining a pixel value of the image. 5.   The image processing apparatus according to claim 1, wherein the first resolution image is a binary image. A resolution conversion step of converting an image of the first resolution into an image of a second resolution lower than the first resolution;
A determination step of referring to a target pixel in the image of the second resolution and a peripheral pixel of the target pixel and determining whether the target pixel is an edge;
The image of the first resolution corresponding to the target pixel based on the value of the target pixel determined to be an edge portion by the determination step and the ratio between the first resolution and the second resolution. An image processing method comprising: a determining step for determining a pixel of an edge portion in   The program for functioning a computer as each means of the image processing apparatus as described in any one of Claims 1 thru | or 7.

Images