JP2010219577A - Image processing apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image processing apparatus capable of sensing an original image, a background portion and a latent image portion and converting them into a four-value image data in which density is different among the original image, the background portion and the latent image portion and outputting the converted image data. <P>SOLUTION: The image processing apparatus includes: a binary image storage section for storing binary image data for at least two lines of a horizontal scanning direction; a coupling dark pixel detecting section for extracting dark pixels from the binary image data, and detecting the number of pixels of the coupled dark pixel consisting of dark pixels adjacent in the horizontal direction and the center position; a candidate pixel extracting section for extracting candidate pixels of an isolated point and a dot from the coupled dark pixel up to the present line by comparison between a result of detection of the present line and a result of detection of one line before; and a four-value image generating section for estimating the isolated point and the dot from the candidate pixels on the basis of the maximum value of the number of pixels about the coupled dark pixels adjacent to the vertical scanning direction and the number of coupled lines, and allocating gradation density corresponding to the original image, the isolated point and the dot and generating a four-value image data. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to an image processing apparatus, and more specifically, includes a background portion in which density adjustment is performed by changing the density of isolated points and a latent image portion in which density adjustment is performed by changing the size of halftone dots. The present invention relates to an image processing apparatus that converts binary image data in which a tint block image is embedded into quaternary image data and outputs it.

  2. Description of the Related Art Conventionally, a technique for embedding a tint block image and printing is known as a technique for suppressing information leakage due to printed matter and duplication of the printed matter (for example, Patent Documents 1 to 3). For example, when printing document data on paper, a character string such as copy prohibition is embedded as a copy-forgery-inhibited pattern image in a document image made up of document data. The copy-forgery-inhibited pattern image is composed of a background portion made up of high-resolution isolated dots and a latent image portion made up of low-resolution halftone dots. By using the phenomenon in which the latent image portion is lifted by removing the image, the duplication is suppressed.

  In order for the copy-forgery-inhibited pattern image to exhibit the effect of suppressing duplication, it is necessary that the isolated dots constituting the background portion disappear during duplication, and the halftone dots must be printed so clearly that the latent image portion can be visually recognized. Further, in order to make the copy-forgery-inhibited pattern image inconspicuous in the printed matter, the background portion and the latent image portion must have substantially the same density when viewed macroscopically. In general, a halftone dot is an aggregate (lumps) of a large number of printing dots from a microscopic viewpoint, and macro density adjustment is performed by changing the number of printing dots constituting the halftone dot to increase the size of the halftone dot. This is done by increasing or decreasing the size. On the other hand, an isolated point is composed of about one print dot, and macro density adjustment is performed by changing the density of isolated points.

  In the case of an image processing apparatus that generates binary image data composed of white pixels and black pixels by embedding a copy-forgery-inhibited pattern image in a document image, and outputs the data to a printer, the absolute density of the background portion and the latent image portion, and between the background portion and the latent image portion The density of the copy-forgery-inhibited pattern image can be adjusted by software so that the density difference between the background portion and the latent image portion and the original image are all optimal. However, even if the density of the original image and the copy-forgery-inhibited pattern image is optimized in advance, the dot reproduction performance of the printer varies greatly depending on mechanical factors, individual differences, aging, etc., and therefore it depends on the printer used for printing binary image data. However, there is a problem that an effective tint block image cannot be formed.

  By using a smoothing technique that smooths edges by adjusting the size and position of printing dots, printing is performed at a resolution higher than the resolution of input data, for example, 600 dpi (dot per inch), for example, equivalent to 1200 dpi or 2400 dpi. Printers that can do this are known. There is also known an electrophotographic printer capable of macroscopic density adjustment by changing the size of printing dots by adjusting the exposure time. Therefore, when printing binary image data in which a copy-forgery-inhibited pattern image is embedded, pixels constituting the original image, the background portion, and the latent image portion are detected, and the print resolution and printing are performed between the original image, the background portion, and the latent image portion. It is conceivable to print with different densities.

  Detection of halftone dots constituting the latent image portion is performed by, for example, extracting adjacent black pixels in the horizontal scanning direction and the vertical scanning direction, and such connected black pixels appear periodically in the horizontal scanning direction and the vertical scanning direction. It was done by determining whether or not. That is, if at least three halftone dots do not exist continuously in both the horizontal and vertical scanning directions, they are not detected as halftone dots. For this reason, depending on the shape and size of characters, symbols, figures, etc. embedded as a latent image portion, an area where only one or two halftone dots exist is generated, and the halftone dots in the area cannot be detected correctly. There was a problem. Further, in a tint block image created in advance, even if three or more halftone dots constituting the latent image portion exist continuously in both the horizontal and vertical scanning directions, the halftone dots are synthesized by combining with the original image. It is also conceivable that there will be a region where only one or two exists.

  Further, in the above halftone dot detection method, since it is necessary to determine the periodicity in both the horizontal and vertical scanning directions, a vertical line buffer for holding binary image data for one line in the horizontal scanning direction is used. There is also a problem that a capacity of at least three times the upper limit size of the halftone dot in the scanning direction must be secured.

JP 2007-47383 A JP 2008-271563 A JP 2005-94327 A

  The present invention has been made in view of the above circumstances, and detects a document image, a background portion, and a latent image portion based on binary image data in which a copy-forgery-inhibited pattern image is embedded, and the document image, the background portion, and the latent image portion. An object of the present invention is to provide an image processing apparatus capable of converting and outputting quaternary image data having different densities. In particular, an object of the present invention is to provide an image processing apparatus in which the detection accuracy of halftone dots constituting a latent image portion of a tint block image is improved. It is another object of the present invention to provide an image processing apparatus in which the capacity of a line buffer required for detecting halftone dots is reduced.

  In the image processing apparatus according to the first aspect of the present invention, a copy-forgery-inhibited pattern image including a background portion in which density adjustment is performed by changing the density of isolated points and a latent image portion in which density adjustment is performed by changing the size of halftone dots. An image processing apparatus for converting embedded binary image data into quaternary image data and outputting the quaternary image data, the binary image storage unit holding the binary image data for at least two lines in the horizontal scanning direction; A black pixel is extracted from the binary image data, and the number of connected black pixels and the center position of the black pixels adjacent to each other in the horizontal scanning direction are detected. A candidate pixel extraction unit that extracts candidate pixels as candidates for the isolated points and halftone dots from the connected black pixels up to the current line by comparing the detection result with the detection result of the previous line; and a vertical scanning direction Adjacent to above The isolated point and the halftone dot are estimated from the candidate pixel based on the maximum value of the number of pixels and the number of connected lines regarding the black pixel, and the gradation density corresponding to the original image, the isolated point and the halftone dot is assigned. And a quaternary image generation unit that generates quaternary image data.

  In this image processing apparatus, binary image data is converted into quaternary image data by assigning gradation densities corresponding to the original image, isolated points, and halftone dots to black pixels, and output. Therefore, when such an image processing apparatus is used, when the binary image data in which the copy-forgery-inhibited pattern image is embedded is printed, pixels constituting the original image, the background portion, and the latent image portion are detected, and the original image, the isolated point is detected. In addition, printing is performed using quaternary image data to which gradation densities corresponding to halftone dots are assigned, so that printing can be performed with different print resolutions or print densities between the original image, the background portion, and the latent image portion. At that time, candidate pixels are extracted by comparing the current line and the previous line with the number and center position of connected black pixels made up of black pixels adjacent in the horizontal scanning direction, and connected adjacent in the vertical scanning direction. Isolated points and halftone dots are estimated based on the maximum number of pixels and the number of connected lines regarding black pixels.

  According to such a configuration, candidate pixels are extracted by comparing the number of pixels and the center position of the connected black pixels between the current line and the previous line, and based on the maximum number of pixels and the number of connected lines. Since the isolated points and halftone dots are estimated, the halftone dots can be correctly detected even in a region where only one or two halftone dots exist. Therefore, based on the binary image data in which the copy-forgery-inhibited pattern image is embedded, the original image, the background portion, and the latent image portion are detected, and converted into four-value image data in which the density is different between the original image, the background portion, and the latent image portion. Thus, it is possible to improve the accuracy of halftone dot detection. In addition, since binary image data for at least two lines is held and the number of connected black pixels and the center position are compared between the current line and the previous line, the isolated points and halftone dots are estimated. The capacity of the line buffer required to detect halftone dots can be reduced as compared with the conventional detection method in which halftone dots are detected by judging the periodicity in both the vertical scanning directions.

  In the image processing apparatus according to the second aspect of the present invention, in addition to the above configuration, the candidate pixel extraction unit excludes connected black pixels whose number of pixels exceeds the upper limit size of the halftone dot in the horizontal scanning direction from the candidate pixels. Composed.

  In the image processing apparatus according to the third aspect of the present invention, in addition to the above-described configuration, the candidate pixel extraction unit selects a connected black pixel in which the number of connected lines up to the current line exceeds the upper limit size of the halftone dot in the vertical scanning direction. Configured to be excluded from.

  In the image processing apparatus according to the fourth aspect of the present invention, in addition to the above-described configuration, the candidate pixel extraction unit may perform the maximum value and the connection for the connected black pixels of the current line in which there is no black pixel adjacent to the previous line. Configured to reset the number of lines.

  In the image processing apparatus according to the fifth aspect of the present invention, in addition to the above configuration, the quaternary image generation unit determines a candidate pixel having the maximum value and the number of connected lines of 1 for the current line as an isolated point, A candidate pixel associated with a connected black pixel in which the maximum value and the number of connected lines are both N (N is an integer not less than 3 and not more than the upper limit size of the halftone dot in the vertical scanning direction) is determined as a halftone dot. Composed.

  According to the image processing apparatus of the present invention, candidate pixels are extracted by comparing the number of pixels of the connected black pixels and the center position between the current line and the previous line, and the maximum number of pixels and the number of connected lines are obtained. Since isolated points and halftone dots are estimated based on this, halftone dots can be detected correctly even in a region where only one or two halftone dots exist. Therefore, based on the binary image data in which the copy-forgery-inhibited pattern image is embedded, the original image, the background portion, and the latent image portion are detected, and converted into four-value image data in which the density is different between the original image, the background portion, and the latent image portion. Thus, it is possible to improve the accuracy of halftone dot detection. In addition, since binary image data for at least two lines is held and the number of connected black pixels and the center position are compared between the current line and the previous line, the isolated points and halftone dots are estimated. The capacity of the line buffer required to detect halftone dots can be reduced as compared with the conventional detection method in which halftone dots are detected by judging the periodicity in both the vertical scanning directions.

1 is a system diagram showing an example of a schematic configuration of a tint block adding system 100 including an image processing apparatus 20 according to an embodiment of the present invention. It is the figure which showed an example of the copy-forgery-inhibited pattern image B1 produced in PC10 of FIG. FIG. 2 is a block diagram illustrating a configuration example of a main part of the tint block addition system 100 of FIG. 1, in which an example of a functional configuration in the image processing apparatus 20 is illustrated. It is explanatory drawing which showed typically an example of operation | movement of the image processing apparatus 20 of FIG. 3, and the process at the time of converting a binary image into a quaternary image is shown. FIG. 4 is an explanatory diagram schematically illustrating an operation example of the image processing apparatus 20 in FIG. 3, in which the number of connected lines does not match the maximum value of the number of pixels and is not a halftone dot. It is explanatory drawing which showed typically the operation example of the image processing apparatus 20 of FIG. 3, and the case where the center position of a connection black pixel does not correspond by the present line and the front line is shown. FIG. 4 is an explanatory diagram schematically showing an operation example of the image processing apparatus 20 in FIG. 3, and shows a case where the difference in the number of pixels between the current line and the previous line is neither 0 nor ± 2. FIG. 4 is an explanatory diagram schematically showing an operation example of the image processing apparatus 20 in FIG. 3, and shows a case where all of the conditions (1) to (5) are satisfied and extracted as halftone dot candidate pixels. It is explanatory drawing which showed typically the operation example of the image processing apparatus 20 of FIG. 3, and the case where it should not be set as the candidate pixel of a halftone dot is shown. FIG. 4 is a diagram illustrating an operation example of the image processing apparatus 20 in FIG. 3, and illustrates an example of a pixel value conversion table used when assigning gradation densities. FIG. 4 is a diagram illustrating an operation example of a printer 30 that adjusts print resolution and print density based on quaternary image data generated by the image processing apparatus 20 of FIG. 3. 4 is a flowchart showing an example of the operation of the image processing apparatus 20 in FIG. 3. 4 is a flowchart showing an example of the operation of the image processing apparatus 20 in FIG. 3. 4 is a flowchart showing an example of candidate pixel extraction processing in the image processing apparatus 20 of FIG. 3. 4 is a flowchart showing an example of candidate pixel extraction processing in the image processing apparatus 20 of FIG. 3.

<System diagram>
FIG. 1 is a system diagram showing an example of a schematic configuration of a tint block adding system 100 including an image processing apparatus 20 according to an embodiment of the present invention. This copy-forgery-inhibited pattern addition system 100 includes a PC (personal computer) 10, an image processing apparatus 20, and a printer 30, and converts binary image data in which a copy-forgery-inhibited pattern image is embedded into four-value image data and prints it on paper A1. .

  The PC 10 is an information processing terminal that includes an application program for creating a tint block, creates a desired tint block image by executing the application program, and adds the tint block image to the document image to generate binary image data. is there. A copy-forgery-inhibited pattern image is a binary image composed of a background portion composed of high-resolution isolated points and a latent image portion composed of low-resolution halftone dots, and a predetermined character string, symbol, figure or the like is a latent image. Embedded as part.

  The document image is a binary image for printing that is read from a document using a scanner (not shown) or created by the PC 10. By combining the copy-forgery-inhibited pattern image and the original image, binary image data including white pixels and black pixels is created.

  The image processing device 20 detects a document image, an isolated point, and a halftone dot based on binary image data input from the PC 10 and assigns a gradation density corresponding to the document image, the isolated point, and the halftone dot to obtain a binary value. The image data is converted into quaternary image data and output to the printer 30 is performed.

  The printer 30 is a printing device that adjusts the printing resolution and printing density based on the four-value image data from the image processing device 20 and prints the composite image in which the copy-forgery-inhibited pattern image is embedded on the paper A1 to generate the printed matter A2. is there.

  As such a printer 30, for example, an electrophotographic printer that irradiates a photosensitive drum with light to be exposed, attaches toner, and transfers the toner onto a sheet can be used. The print resolution is adjusted by controlling the timing of turning on or off the exposure and changing the size and position of the print dots to smooth the edges. The print density is adjusted by changing the size of the print dots by controlling the exposure time.

  Here, the printing resolution is the number of printing dots per unit length, and printing is performed at a resolution equivalent to 1200 dpi or 2400 dpi, for example, with respect to the resolution of 600 dpi of input data by smoothing by the above-described exposure timing control. Can do.

  When duplicating the printed matter A2 using the copy (copier) machine 40, if the input resolution when reading the print dots is less than or equal to the printing resolution of the printer 30, a copy-forgery-inhibited pattern image that was not noticeable in the printed matter A2 is copied. Can stand out. In other words, in the printed matter A2, the background portion and the latent image portion have substantially the same density when viewed macroscopically, but in the duplicate A3, the isolated points constituting the background portion of the tint block image are removed, and the halftone dots The latent image portion formed by

  The printer 30 adjusts the print resolution and the print density based on the four-value image data to which the gradation density corresponding to the original image, the isolated point, and the halftone dot is assigned, so that the space between the original image, the background portion, and the latent image portion. Thus, it is possible to print with different print resolution and print density. Therefore, it is possible to form an effective tint block image corresponding to the dot reproduction performance of the printer 30.

<Background image>
2A to 2C are diagrams showing an example of the tint block image B1 created in the PC 10 of FIG. FIG. 2A shows the entire tint block image B1 including the background portion B2 constituted by isolated points and the latent image portion B3 constituted by halftone dots. FIG. 2B shows an enlarged view of the background portion B2, and FIG. 2C shows an enlarged view of the latent image portion B3.

  An isolated point is composed of one black pixel or several adjacent black pixels, and the density adjustment of the background portion B2 is performed by changing the density of isolated points, that is, the number of isolated points per unit area. Done. On the other hand, a halftone dot is an aggregate of a large number of black pixels adjacent to each other, and is larger than an isolated point. The density adjustment of the latent image portion B3 is performed by changing the size of the halftone dots.

  A halftone dot is represented by a figure having a symmetrical shape with respect to the horizontal scanning direction, that is, the horizontal direction in this drawing, and the vertical scanning direction, that is, the vertical direction, for example, a square, a circle, or the like. By embedding such a copy-forgery-inhibited pattern image B1 in the document image, binary image data is created.

<Image processing device>
FIG. 3 is a block diagram illustrating a configuration example of a main part of the tint block addition system 100 in FIG. 1, and illustrates an example of a functional configuration in the image processing apparatus 20. The image processing apparatus 20 includes a line buffer 1, a buffer control unit 2, a connected black pixel detection unit 3, a pixel number and center position storage unit 4, a candidate pixel extraction unit 5, a connection parameter storage unit 6, and a quaternary image generation unit 7. Consists of.

  The line buffer 1 is a binary image storage unit that can hold binary image data to which a copy-forgery-inhibited pattern image is added for at least two lines in the horizontal scanning direction. The horizontal scanning direction here is the horizontal direction of the binary image.

  Here, when binary image data is input from an external device such as the PC 10 in units of the horizontal width of the binary image, a line capacity that is twice or more the horizontal width of the binary image is secured as the line buffer 1. Shall.

  The buffer control unit 2 controls the reading and writing of binary image data with respect to the line buffer 1, buffers the binary image data sequentially input from the PC 10, and sends it to the connected black pixel detection unit 3 and the quaternary image generation unit 7. The operation to output is performed.

  Here, binary image data for three lines is held by the line buffer 1, and each image data is updated at a period corresponding to the horizontal width of the image. That is, if the line to be processed as a connected black pixel detection target is called the current line, the previous line is called the previous line, and the next line to be processed is called the next line, the line buffer 1 has the previous line, The binary image data of the current line and the next line is held.

  The connected black pixel detection unit 3 scans the binary image data on the line buffer 1 to extract black pixels, detects the number of connected black pixels and the center position in the horizontal scanning direction, and outputs pixel number information. The center position information is stored in the pixel number and center position storage unit 4.

  The black pixel is, for example, a pixel having a pixel value = 1, and the connected black pixels are black pixels adjacent in the horizontal scanning direction. Here, one isolated black pixel is also referred to as a connected black pixel with the number of pixels = 1. Further, the center position of the connected black pixels is a relative position in the horizontal line. Here, when the number of pixels is an even number, the center position of the connected black pixels is one of two pixels sandwiching the central boundary line. For example, it is assumed that the pixel located in front is determined as the center position.

  The pixel number and center position storage unit 4 holds the pixel number information and center position information detected for the current line, and the pixel number information and center position information detected for the previous line. The number of pixels and center position storage unit 4 is configured by, for example, a plurality of registers respectively associated with pixels for two lines, and the black corresponding to the connected black pixels in the register corresponding to the center position of the connected black pixels. By storing the number of pixels, the center position information is retained.

  The candidate pixel extraction unit 5 compares the detection result of the current line by the connection black pixel detection unit 3 with the detection result of the previous line, thereby selecting candidates for isolated points and halftone dots from the connection black pixels up to the current line. An operation of extracting a pixel and outputting the extraction result to the quaternary image generation unit 7 is performed.

  In the connection parameter storage unit 6, the maximum value of the number of pixels and the number of connection lines regarding the connection black pixels adjacent in the vertical scanning direction in the binary image are held as connection parameters. The number of connected lines is the number of lines included in an aggregate composed of connected black pixels adjacent in the vertical scanning direction.

  First, the upper limit size in the horizontal scanning direction of halftone dots constituting the latent image portion of the copy-forgery-inhibited pattern image is M, and connected black pixels having a pixel number exceeding M are excluded from isolated points and halftone dots. exclude.

  For example, by rewriting the number of pixels held in the register corresponding to the center position of such a connected black pixel to 0, the black pixels constituting the connected black pixel are excluded from the candidate pixels. Alternatively, as a register for holding the number of connected black pixels, a register capable of storing an integer of 0 or more and M or less is used, and for a connected black pixel whose number of pixels exceeds M, the number of pixels = 0 may be stored. good.

  As the connection parameter storage unit 6, a register capable of storing an integer of 0 or more and M or less can be used.

  Next, attention is paid to a register in which a numerical value other than 0 is stored as the number of connected black pixels for the current line, and the number of pixels of this register corresponds to the same position in the horizontal scanning direction as that register in the previous line. Compare the number of pixels in the register.

Then, it is sequentially determined whether or not the following conditions (1) and (2) are satisfied. Subsequently, referring to the disable signal of the isolated points and halftone dots, the number of connected lines, and the binary image data of the next line, it is sequentially determined whether or not the conditions (3) to (5) are satisfied.
(1) The number of pixels in the previous line ≧ 1.
(2) Difference in the number of pixels = 0 or ± 2.
(3) The disable signal for isolated points and halftone dots is not output in the processing of the previous line.
(4) The number of connected lines has not reached the upper limit size M of halftone dots by the previous line.
(5) There is no black pixel in any of the lower left, lower and lower right of the connected black pixels in the current line.

  When the condition (1) is not satisfied, the binary image data of the previous line and the current line in the line buffer 1 is referred to. At this time, if a black pixel exists in any of the upper left, upper, or upper right of the connected black pixels in the current line, it is not left-right symmetric. Therefore, the connected black pixels in the current line are not considered as isolated pixels and halftone dot candidate pixels. . Then, in order to exclude the connected black pixel of the next line having the same center position in the horizontal scanning direction as the connected black pixel of the current line from the candidate pixels, the isolated point and the halftone dot with respect to the connected black pixel of such next line The disable signal is output.

  On the other hand, if there are no black pixels at the upper left, upper, and upper right of the connected black pixels in the current line, there is a possibility that the line is the start line of an isolated point or a halftone dot. The value and the number of connected lines are reset, and it is determined whether or not the condition (5) is satisfied. That is, the maximum value of the number of pixels and the number of connected lines are reset for the connected black pixels of the current line in which there are no black pixels adjacent to the previous line.

  If the condition (2) is not satisfied, it is not left-right symmetric, or even if it is left-right symmetric, the change in the number of pixels is large. Therefore, the connected black pixel of the current line is not used as a candidate pixel for an isolated point or a halftone dot. Then, in order to exclude the connected black pixel of the next line having the same center position in the horizontal scanning direction as the connected black pixel of the current line from the candidate pixels, the isolated point and the halftone dot with respect to the connected black pixel of such next line The disable signal is output.

  If the condition (3) is not satisfied, the connected black pixels in the previous line are connected in the vertical direction to the connected black pixels that are disabled in the isolated points and halftone dots. And not. The disable signals for the isolated points and halftone dots are continuously output as they are.

  When the condition (4) is not satisfied, the number of connected lines up to the current line exceeds the upper limit size M of the halftone dot in the vertical scanning direction. Exclude from Then, in order to exclude the connected black pixel of the next line having the same center position in the horizontal scanning direction as the connected black pixel of the current line from the candidate pixels, the isolated point and the halftone dot with respect to the connected black pixel of such next line The disable signal is output.

  The condition (5) is determined by referring to the binary image data of the next line and the current line in the line buffer 1. If this condition (5) is not satisfied, that is, if a black pixel exists in any of the lower left, lower or lower right of the connected black pixels in the current line, they are connected or adjacent to the connected black pixels in the next line in the vertical direction. Therefore, the connected black pixels on the current line are not set as isolated point and halftone dot candidate pixels.

  Here, the connected black pixel of the current line may be a connected black pixel constituting a halftone dot, but since it does not reach the end line of the black pixel connected in the vertical direction, an isolated point and a halftone dot are present at this stage. If the condition (5) is satisfied with the end line of the black pixels connected in the vertical direction instead of the candidate pixels, the isolated pixels and the halftone dots are selected. Therefore, the isolated point and halftone dot disable signals are not output to the connected black pixels of the next line that have the same center position in the horizontal scanning direction as the connected black pixels of the current line.

  When all of the conditions (1) to (4) are satisfied and when the condition (1) is not satisfied, and there are no black pixels on the upper left, upper and upper right of the connected black pixels on the current line, If the number of connected black pixels on the current line exceeds the maximum value obtained by the determination up to the previous line, the maximum value of the number of pixels in the connected parameter storage unit 6 is updated. Further, the number of connection lines in the connection parameter storage unit 6 is incremented.

  On the other hand, when all of the conditions (1) to (5) are satisfied and when the condition (1) is not satisfied, there are no black pixels in the upper left, upper and upper right of the connected black pixels in the current line. In addition, when the condition (5) is satisfied, the connected black pixels of the current line are set as isolated point and halftone dot candidate pixels.

  The quaternary image generation unit 7 includes an isolated point / halftone dot estimation unit 21, a pixel value conversion unit 22, and a quaternary quaternary memory 23. The quaternary image generation unit 7 estimates an isolated point and a halftone dot with reference to the connection parameters, and the original image. Then, an operation is performed in which gradation density corresponding to isolated points and halftone dots is assigned to generate quaternary image data.

  The isolated point and halftone dot estimation unit 21 performs an operation of estimating isolated points and halftone dots from candidate pixels based on the maximum number of pixels and the number of connected lines held in the connection parameter storage unit 6. Here, regarding an aggregate of pixels composed of candidate pixels of connected black pixels, if the maximum size in the horizontal scanning direction matches the maximum size in the vertical scanning direction, it is determined as an isolated point or a halftone dot. The isolated point is assumed to be composed of one connected black pixel.

Specifically, referring to the connection parameter, it is determined which of the following conditions (a) to (d) is satisfied, and isolated points and halftone dots are estimated based on the determination result.
(A) Maximum number of pixels = number of connected lines = 1
(B) Maximum number of pixels = number of connected lines = N (N is an arbitrary integer not less than 3 and not more than the upper limit size M in the vertical scanning direction of halftone dots)
(C) Maximum number of pixels = number of connected lines = 2
(D) Maximum number of pixels ≠ number of connected lines

  When the condition (a) is satisfied, the connected black pixel of the current line is determined as an isolated point, and an isolated point enable signal is output to the connected black pixel of the current line.

  If the condition (b) is satisfied, the candidate pixel associated with the connected black pixel of the current line is determined as a halftone dot, and a halftone dot enable signal is output to the candidate pixel associated with the connected black pixel of the current line. To do.

  Specifically, for the connected black pixels on the current line, a halftone dot signal is output to pixels (white pixels and black pixels) in a range corresponding to the maximum value of the number of pixels with reference to the center position. At that time, for a total of M lines from the current line to the (M−1) th previous line, a halftone dot enable signal is transmitted to the pixel including the connected black pixel having the same center position in the horizontal scanning direction as the connected black pixel of the current line. Is output. That is, the halftone dot enable signal is output to all pixels in an N × N square area surrounding the halftone dot with the current line as the lower end.

  When the halftone dot size is fixed to a constant value F of 3 or more, the condition (b) may be replaced with the maximum number of pixels = number of connected lines = F.

  When the condition (c) is satisfied, two black pixels adjacent in the horizontal scanning direction are connected in the vertical scanning direction, and in order to clearly distinguish the isolated point and the halftone dot, normally, the current line The candidate pixels associated with the connected black pixels are not included in either the isolated point or the halftone dot, and no enable signal is output. However, the condition (c) may be included in the condition (a) or (b) if the candidate pixel satisfying the condition (c) is used as an isolated point or halftone dot constituting the copy-forgery-inhibited pattern image.

  When the condition (d) is satisfied, even if it is bilaterally symmetric, it is long in the horizontal scanning direction or the vertical scanning direction. Therefore, the candidate pixels associated with the connected black pixels of the current line are not isolated points and halftone dots, and the enable signal is Neither is output.

  The pixel value conversion unit 22 converts the binary image data from the buffer control unit 2 into quaternary image data based on the isolated point and halftone dot estimation results by the isolated point and halftone dot estimation unit 21, and outputs it to the printer 30. The operation to be performed. In this case, conversion to four-value image data is performed using a four-value memory 23 having a queue structure having a line capacity corresponding to the upper limit size M in the vertical scanning direction of halftone dots.

  Specifically, first, the binary image data of the current line is enqueued in the quaternary memory 23. Next, for each pixel from the current line stored in the quaternary memory 23 to the (M−1) th previous line, an isolated point enable signal and a halftone dot enable signal from the isolated point and halftone dot estimation unit 21 are obtained. By referring to and assigning gradation densities according to the original image, isolated points, and halftone dots, quaternary image data is generated from the binary image data.

  At this time, the isolated point enable signal is output only to the candidate pixel (black pixel) of the current line, whereas the halftone dot enable signal includes an N × N square region including the current line and surrounding the halftone dot. Are output to all pixels. Therefore, depending on the shape of the halftone dot, the halftone dot enable signal may be output to the white pixel. However, in the case of the white pixel, the binary image data is output as it is without being converted.

  The assignment of the gradation density according to the original image, the isolated point, and the halftone dot is performed based on, for example, a pixel value conversion table that holds the gradation density in association with the isolated point or halftone dot enable.

  Next, when the conversion process from the binary image data to the quaternary image data is completed, the quaternary image data of the (M−1) th line before the current line is dequeued. Then, in conjunction with the input of the line data to the line buffer 1, the binary image data is enqueued in the quaternary memory 23, the line data on the quaternary memory 23 is updated, and memory access and conversion for data conversion are performed. The dequeuing of the later four-value image data is repeated.

  If the current line is a line preceding the Mth line from the first line, the binary image data of the (M−1) th previous line does not exist in the quaternary memory 23, so the binary Only conversion processing from image data to quaternary image data is performed, and queuing of quaternary image data is not performed. If the current line is the last line, all the image data existing in the quaternary memory 23 is fixed, and therefore all the quaternary image data from the current line to the (M−1) th previous line. Is dequeued.

  Here, when extracting candidate pixels of isolated points and halftone dots by comparing the detection result of the number of pixels and the center position for the current line with the detection result for the previous line, the conditions (1) to (5) It is conceivable that halftone dot candidate pixels cannot be extracted appropriately only by determination. That is, even when the aggregate of connected black pixels is left-right symmetric or not vertically symmetric, or when the outline is not a convex polygon even if left-right symmetric and vertically symmetric, the candidate pixel for halftone dots Will be extracted.

  The term “convex polygon” as used herein refers to a polygon in which any two points on the contour line are located outside the line segment connecting the two points between the two points. It is.

Therefore, in order to exclude such a case, after determining the conditions (1) to (5), it is determined whether or not the following conditions (6) and (7) are satisfied as necessary. It may be added.
(6) The same number of lines where the difference in the number of pixels from the previous line is +2 and the line which is −2 appear by the current line.
(7) A line in which the difference in the number of pixels from the previous line appears to −2, and the difference in the number of pixels from the previous line up to the current line becomes +2. Has not appeared.

<Processing process>
FIG. 4 is an explanatory diagram schematically showing an example of the operation of the image processing apparatus 20 in FIG. 3, and shows a processing process when converting a binary image into a quaternary image. The line buffer 1 always holds binary image data for three lines, that is, the next line, the current line, and the previous line, and is updated at a period corresponding to the horizontal width 31 of the binary image.

  For the previous line data, the first to fifth pixels are white pixels, the sixth to eighth pixels are black pixels, the ninth to twelfth pixels are white pixels, and the thirteenth to sixteenth pixels are The black pixel and the 17th and subsequent pixels are white pixels.

  In the current line data, the first pixel is a white pixel, the second pixel is a black pixel, the third pixel to the fifth pixel are white pixels, the sixth pixel to the eighth pixel are black pixels, the ninth pixel to the thirteenth pixel Up to the white pixels, the 14th to 15th pixels are black pixels, and the 16th and subsequent pixels are white pixels.

  In the next line data, the first pixel to the fifth pixel are white pixels, the sixth pixel to the eighth pixel are black pixels, and the ninth and subsequent pixels are white pixels.

  The connected black pixel detection unit 3 scans the binary image data on the line buffer 1 to extract a black pixel corresponding to the print dot, for example, a pixel having a pixel value “1” and adjacent black in the horizontal scanning direction. The number of pixels and the center position of the connected black pixels 32 made up of pixels are determined. The determination result is held by storing the number of connected black pixels 32 in a register corresponding to the center position of the connected black pixels 32 among the registers holding the number of pixels for the connected black pixels 32 of the current line. .

  Specifically, the current line pixel corresponding to the center position of the connected black pixel 32 from the sixth pixel to the eighth pixel is set in the current line pixel number register corresponding to the second connected black pixel 32. The number register stores the number of pixels “3”, and the current line pixel number register corresponding to the center position of the connected black pixels 32 from the 14th pixel to the 15th pixel stores “2”.

  In the previous line pixel number register for holding the number of pixels of the previous line, the number of pixels “3” and 13th pixel are stored in the previous line pixel number register corresponding to the center position of the connected black pixels 32 from the sixth pixel to the eighth pixel. The pixel number “4” is stored in the previous line pixel number register corresponding to the center position of the connected black pixels 32 from the first to the 16th pixel.

  The candidate pixel extraction unit 5 compares the detection result of the number of pixels and the center position for the current line with the detection result for the previous line, performs the determination of the conditions (1) to (5), and extracts the candidate pixels. If necessary, an isolated point and halftone dot disable signal is output.

  In this example, the condition (1) is not satisfied for the second connected black pixel 32 of the current line, but there is no black pixel in any of the upper left, upper, and upper right of the connected black pixel 32. Since the condition (5) is also satisfied, it is a candidate pixel for an isolated point. Further, the connected black pixel 32 from the sixth pixel to the eighth pixel is excluded from the candidate pixels because the condition (3) is not satisfied. Further, the connected black pixels 32 from the 14th pixel to the 15th pixel satisfy all the conditions (1) to (5), and thus are candidate pixels.

  The register that holds the maximum value of the number of pixels related to the connected black pixels 32 adjacent in the vertical scanning direction stores the maximum value after being updated along with the extraction of the candidate pixels. Specifically, the maximum value “1” is stored in the maximum value register corresponding to the connected black pixel 32 of the second pixel, and the maximum value is stored in the maximum value register corresponding to the center position of the connected black pixel 32 from the sixth pixel to the eighth pixel. The value “3” and the maximum value “4” are stored in the maximum value registers corresponding to the center positions of the connected black pixels 32 from the 14th pixel to the 15th pixel, respectively.

  Further, the register that holds the number of connected lines in the vertical scanning direction stores the number of connected lines that is updated as candidate pixels are extracted. Specifically, the number of connected lines is “1” in the vertical connection number register corresponding to the connected black pixel 32 of the second pixel, and the number of vertical connections corresponding to the center positions of the connected black pixels 32 from the sixth pixel to the eighth pixel. The number of connected lines “9” is stored in the register, and the number of connected lines “4” is stored in the vertical connected number register corresponding to the center position of the connected black pixels 32 from the 14th pixel to the 15th pixel.

  The quaternary image generation unit 7 estimates isolated points and halftone dots by referring to a register that holds the maximum value of the number of pixels and a register that holds the number of connected lines, and according to the original image, the isolated points, and the halftone dots. The gradation density is assigned to generate quaternary image data. The isolated points and halftone dots are estimated based on the determination result as to which of the conditions (a) to (d) is satisfied.

  In this example, the condition (a) is satisfied for the second connected black pixel 32 of the current line, so the connected black pixel 32 of the current line is determined to be an isolated point, and an isolated point enable signal is output. ing.

  Since the 14th to 15th connected black pixels 32 satisfy the condition (b), the candidate pixel associated with the connected black pixel 32 of the current line is determined as a halftone dot. An enable signal is output. Specifically, for the connected black pixels 32 on the current line, a halftone dot signal is output to pixels in a range corresponding to the maximum value “4” of the number of pixels with the center position as a reference. That is, the halftone dot enable signal is output to all pixels in a 4 × 4 square area surrounding the halftone dot with the current line as the lower end.

  In the quaternary memory 23, binary image data of the current line is enqueued, and an isolated point enable signal and a halftone dot enable signal are stored for each line data from the stored current line to the (M-1) th previous line. Referring to FIG. 4, quaternary image data is generated from binary image data by assigning gradation densities corresponding to the original image, isolated points, and halftone dots.

  Specifically, gradation density “1” is assigned to the original image, gradation density “2” is assigned to the isolated point, gradation density “3” is assigned to the halftone dot, and gradation density “0” is assigned to the white pixel. Thus, quaternary image data is generated. In this figure, the description of the pixel value of the white pixel is omitted.

  When the conversion process from the binary image data to the quaternary image data is completed, the (M−1) previous quaternary image data is dequeued.

<Extraction of halftone dot candidates>
5 to 9 are explanatory diagrams schematically showing an operation example of the image processing apparatus 20 of FIG. 3, in which an example of the connected black pixels 32 extracted for five lines including the current line is shown. Yes. FIGS. 5A and 5B show a case where the number of connected lines in the vertical scanning direction and the maximum value of the number of pixels do not match and do not form halftone dots. The black pixel 33 with a circle is a black pixel at the center position of the connected black pixel 32.

  In FIG. 5A, five connected black pixels 32 having a pixel number “2” centered on the black pixel 33 of the third pixel are connected in the vertical scanning direction, the number of connected lines = 5, and the maximum value. The case of = 2 is shown. In FIG. 5B, a connected black pixel 32 having the pixel number “1” and a connected black pixel 32 having the pixel number “3” are connected in the vertical scanning direction with the black pixel 33 of the third pixel as the center position. In this example, the number of connected lines = 5 and the maximum value = 3. In these cases, each candidate pixel is not a halftone dot because it is long in the vertical scanning direction even if it is laterally and vertically symmetric.

  FIGS. 6A and 6B show a case where the center position of the connected black pixel 32 does not match between the current line and the previous line, and the halftone dot candidate pixel is not used. FIG. 6A shows a connected black pixel 32 having a pixel number “4” centered on the second black pixel 33 and a pixel number “5” centered on the third black pixel 33. A case where the connected black pixels 32 are connected in the vertical scanning direction and the center positions of the third line and the second line do not coincide is shown.

  FIG. 6B shows a connected black pixel 32 having a pixel number “3” centered on the second black pixel 33 and a pixel number “5” centered on the third pixel black pixel 33. A case where the connected black pixels 32 are connected in the vertical scanning direction and the center positions of the third line and the second line do not coincide is shown.

  In these cases, since the center positions of the connected black pixels do not match between the current line and the previous line, the condition (1) is not satisfied and the left and right are not symmetrical. For this reason, these connected black pixels 32 are not considered as dot candidate pixels.

  FIGS. 7A and 7B show a case where the difference in the number of pixels between the current line and the previous line is neither 0 nor ± 2, and is not a halftone dot candidate pixel. In FIG. 7A, a connected black pixel 32 with a pixel number “4” and a connected black pixel 32 with a pixel number “5” are connected in the vertical scanning direction with the black pixel 33 of the third pixel as the center position. A case where the difference in the number of pixels between the third line and the second line is +1 is shown.

  In FIG. 7B, a connected black pixel 32 with a pixel number “3”, a connected black pixel 32 with a pixel number “5”, and a pixel number “1” with the black pixel 33 of the third pixel as the center position. A case where the connected black pixels 32 are connected in the vertical scanning direction and the difference in the number of pixels between the fourth line and the third line is −4 is shown.

  In these cases, the condition (2) is not satisfied and is not symmetric, or even if it is symmetric, the change in the number of pixels is large.

  FIGS. 8A to 8C show a case where all of the conditions (1) to (5) are satisfied and should be extracted as halftone dot candidate pixels. FIG. 8A shows a case where the connected black pixels 32 having the number of pixels “5” centered on the black pixel 33 of the third pixel are connected in the vertical scanning direction. In FIG. 8B, a connected black pixel 32 having a pixel number “3” and a connected black pixel 32 having a pixel number “5” are connected in the vertical scanning direction with the black pixel 33 of the third pixel as the center position. The case is shown.

  In FIG. 8C, a connected black pixel 32 with the pixel number “1”, a connected black pixel 32 with the pixel number “3”, and a pixel number “5” with the black pixel 33 of the third pixel as the center position. A case where the connected black pixels 32 are connected in the vertical scanning direction is shown.

  In FIGS. 9A and 9B, all of the conditions (1) to (5) are satisfied, but the condition (6) or (7) is not satisfied and the halftone dot candidate pixel should be used. The case where there is no is shown. In FIG. 9A, a connected black pixel 32 having a pixel number “3” and a connected black pixel 32 having a pixel number “5” are connected in the vertical scanning direction with the black pixel 33 of the third pixel as the center position. In addition, a case is shown in which the same number of lines where the difference in the number of pixels from the previous line is +2 and the line which is −2 do not appear up to the current line.

  In this case, the aggregate composed of the connected black pixels 32 is not symmetrical even if it is bilaterally symmetric, so it should not be a candidate pixel for a halftone dot.

  In FIG. 9B, a connected black pixel 32 having a pixel number “5” and a connected black pixel 32 having a pixel number “3” are connected in the vertical scanning direction with the black pixel 33 of the third pixel as the center position. A line having a pixel number difference of −2 from the previous line appears, and a line having a pixel number difference of +2 from the previous line until the current line The case where it appears is shown.

  In this case, even if the aggregate of the connected black pixels 32 is bilaterally symmetric and vertically symmetric, the outline is not a convex polygon and should not be a halftone dot candidate pixel.

<Pixel value conversion table>
FIG. 10 is a diagram illustrating an operation example of the image processing apparatus 20 in FIG. 3, and illustrates an example of a pixel value conversion table used when assigning gradation densities according to a document image, isolated points, and halftone dots. ing. This pixel value conversion table is a table that holds the gradation density in the case of no enable, the isolated point enabled, and the halftone enabled in association with the binary image data.

  Specifically, the white pixel (pixel value = 0) of the binary image is assigned the same pixel value “0” as that of the binary image data as the gradation density when the enable is not performed and when the halftone is enabled. ing. For the black pixel (pixel value = 1) of the binary image, the same pixel value “1” as that of the binary image data is assigned as the gradation density when the enable is not performed.

  On the other hand, for a black pixel (pixel value = 1) of the binary image, the pixel value “2” is assigned as the gradation density when the isolated point is enabled, and the pixel value “ 3 "is assigned.

  By assigning gradation density using such a pixel value conversion table, it is possible to generate quaternary image data of pixel values “0” to “3” from the binary image data.

<Printer settings>
11A to 11C are diagrams illustrating an operation example of the printer 30 that adjusts the print resolution and the print density based on the quaternary image data generated by the image processing apparatus 20 of FIG. An example of setting the print resolution and print density when the dot reproduction performance of the printer 30 is different is shown.

  FIG. 11A shows a setting example of a printer with standard dot reproduction performance, FIG. 11B shows a setting example of a printer with high dot reproduction performance, and FIG. Shows a setting example of a printer with low dot reproduction performance.

  For the original image (pixel value = 1), in either case, either printing resolution 600 dpi or 1200 dpi is selected depending on the smoothing selection, and a predetermined standard value is selected as the printing density. Is done.

  For the background portion (pixel value = 2) and latent image portion (pixel value = 3) of the copy-forgery-inhibited pattern image, the print resolution and print density are selected according to the dot reproduction performance of the printer.

  Specifically, in the case of a printer with high dot reproduction performance, there is a problem that the copy-forgery-inhibited pattern image is too dark on the printed matter A2 and the original image is difficult to identify. Further, when such a printed matter A2 is duplicated by the copying machine 40, there is a problem that the background portion does not disappear and the latent image portion does not stand out.

  Therefore, it is considered that the print density of the background portion is lowered to a level at which the background portion disappears at the time of copying, and the print density of the latent image portion is also lowered in order to match the macro density between the background portion and the latent image portion. It is done.

  At this time, if the print density is lowered with the same reduction width for the background portion and the latent image portion, the latent image portion becomes too thin and unnoticeable at the time of duplication. The range of lowering is different. Specifically, the background portion has a reduction amount of 20% with respect to the standard value, while the latent image portion has a reduction amount of 10%.

  Thus, by making the lowering width of the latent image portion smaller than that of the background portion, it is possible to prevent the latent image portion from becoming too thin during duplication.

  On the other hand, in the case of a printer with low dot reproduction performance, there is a problem in the printed matter A2 that the background portion is too thin and the latent image portion becomes conspicuous. Further, when such a printed matter A2 is duplicated by the copying machine 40, there is a problem that the latent image portion disappears.

  Therefore, it is conceivable to print by increasing the print density of the background portion in order to increase the print density of the latent image portion to a level that does not disappear during duplication and to match the macro density between the background portion and the latent image portion.

  At this time, if the print density is increased with the same increase width for the background portion and the latent image portion, the latent image portion becomes conspicuous in the printed matter, so the increase width of the print density is different between the background portion and the latent image portion. . Specifically, the background portion has a raised width of 20% with respect to the standard value, while the latent image portion has a raised width of 10%.

  Thus, by making the raising width of the latent image portion smaller than that of the background portion, it is possible to prevent the latent image portion from conspicuous in the printed matter.

  Steps S101 to S117 in FIGS. 12 and 13 are flowcharts showing an example of the operation of the image processing apparatus 20 in FIG. First, the connected black pixel detection unit 3 scans the binary image data on the line buffer 1 to extract black pixels from the current line, and determines the number of connected black pixels composed of adjacent black pixels and the center position thereof. Then, the pixel number information is stored in a register corresponding to the center position of each connected black pixel (steps S101 to S103).

  Next, the candidate pixel extraction unit 5 excludes connected black pixels whose number of pixels exceeds the upper limit size M of halftone dots from isolated point and halftone dot candidate pixels. The pixel number information is rewritten to the pixel number = 0 (steps S104 and S108).

  Further, the candidate pixel extraction unit 5 pays attention to a register in which a numerical value other than 0 is stored as the number of connected black pixels, and the number of pixels in this register and the same position in the horizontal scanning direction as the register in the previous line. Compare the number of pixels in the corresponding register. Then, isolated point and halftone dot candidate pixels are extracted based on the conditions (1) to (5) (steps S104 and S105).

  Next, the isolated point and halftone dot estimation unit 21 of the quaternary image generation unit 7 refers to the connection parameter, and sets the maximum value of the number of pixels and the number of connection lines in the vertical scanning direction to the maximum value = number of connection lines = 1. If it is satisfied, the connected black pixel on the current line is determined as an isolated point, and an isolated point enable signal is output to the black pixel at the center position (steps S106 and S109).

  When the maximum value of the number of pixels and the number of connected lines in the vertical scanning direction satisfy the maximum value = number of connected lines = N, the candidate pixel associated with the connected black pixel of the current line is determined as a halftone dot, A halftone dot enable signal is output to pixels in a range corresponding to the maximum number of pixels with the black pixel at the center as a reference (steps S106, S107, S110).

  Next, the pixel value conversion unit 22 enqueues the binary image data of the current line on the line buffer 1 into the quaternary memory 23 (step S111). For each pixel from the current line held in the quaternary memory 23 to the (M−1) th previous line, based on the isolated point and halftone dot enable signal from the isolated point and halftone dot estimation unit 21. The pixel values are converted to generate quaternary image data from the binary image data (step S112).

  Next, the pixel value conversion unit 22 proceeds to the processing of the next line without dequeuing the quaternary image data if the current line is not the Mth (upper limit size of halftone dot) line after the first line. The processing procedure after step S101 is repeated (steps S113 and S117).

  On the other hand, even if the current line is the Mth and subsequent lines from the first line, if the current line is the last line, the quaternary image data of all lines held in the quaternary memory 23 is dequeued and the next line Then, the processing procedure after step S101 is repeated (steps S113 to S115).

  Further, if the current line is the M-th line from the first line and the current line is not the last line, the (M−1) -th previous line of 4-level image data stored in the 4-level memory 23 is stored. The process is dequeued and the process proceeds to the next line, and the processing procedure from step S101 is repeated (steps S113, S114, S116, and S117).

  Steps S201 to S213 in FIGS. 14 and 15 are flowcharts showing an example of candidate pixel extraction processing in the image processing apparatus 20 in FIG. The candidate pixel extraction unit 5 extracts the isolated point and halftone dot candidate pixels from the black pixels up to the current line by comparing the detection result of the current line and the detection result of the previous line by the connected black pixel detection unit 3. .

  At that time, it is first determined whether or not the current line is the first line. If the current line is not the first line, attention is paid to a register in which a numerical value other than 0 is stored as the number of connected black pixels for the current line. Then, the number of pixels in this register is compared with the number of pixels in the register corresponding to the same position in the horizontal scanning direction as that register in the previous line to determine whether the number of pixels in the previous line is 1 or more. (Steps S201 and S202).

  At this time, if the number of pixels of the previous line is not 1 or more, the binary image data of the previous line and the current line on the line buffer 1 is referred to and either the upper left, upper or upper right of the connected black pixels of the current line. It is determined whether or not a black pixel exists (steps S202 and S211).

  If there is a black pixel in the upper left, upper or upper right of the connected black pixel in the current line, the connected black pixel in the next line having the same center position in the horizontal scanning direction as the connected black pixel in the current line is excluded from the candidate pixels. Therefore, an isolated point and halftone dot disable signal is output to such a connected black pixel in the next line, and this process ends (step S212).

  On the other hand, if there is no black pixel in any of the upper left, upper and upper right of the connected black pixels of the current line, the maximum value and the number of connected lines in the connected parameter storage unit 6 are reset to “0” (step S213). .

  If the number of pixels in the previous line is 1 or more, the difference in the number of pixels between the current line and the previous line is calculated, and it is determined whether the difference in the number of pixels is 0 or ± 2 ( Steps S202 and S203).

  If the difference in the number of pixels is not 0 or ± 2, such a next line is connected in order to exclude a connected black pixel in the next line having the same center position in the horizontal scanning direction as that of the current line. Disabling signals for isolated points and halftone dots are output to the black pixels, and this process is terminated (steps S203 and S212).

  On the other hand, if the difference in the number of pixels is either 0 or ± 2, whether or not the black pixel of the previous line adjacent in the vertical scanning direction is a candidate pixel by referring to the disable signal in the process of the previous line Is determined (steps S203 and S204).

  If the black pixel on the previous line is not a candidate pixel, the disable signal of the isolated point and halftone dot is continuously output as it is, and this process is terminated (steps S204 and S212). If the black pixel on the previous line is a candidate pixel, it is determined whether or not the number of connected lines up to the previous line has reached the upper limit size M of halftone dots (steps S204 and S205).

  If the number of connected lines up to the previous line has reached the upper limit size M, in order to exclude the connected black pixel of the next line that has the same center position in the horizontal scanning direction as the connected black pixel of the current line, An isolated point and halftone dot disable signal is output to the connected black pixel of the next line, and the process is terminated (steps S205 and S212).

  If the current line is the first line in step S201, if the connection parameter is reset in step S213, and if the number of connected lines up to the previous line has not reached the upper limit size M in step S205, the current line It is determined whether or not the number of connected black pixels exceeds the maximum value obtained in the determination up to the previous line. If the number of pixels exceeds the maximum value, the number of pixels in the connection parameter storage unit 6 is determined. The maximum value is updated (steps S206 and S207). Then, the number of connection lines in the connection parameter storage unit 6 is incremented (step S208).

  Next, the binary image data of the current line and the next line on the line buffer 1 are referred to. Then, it is determined whether or not a black pixel exists in any of the lower left, lower or lower right of the connected black pixels in the current line (step S209).

  At this time, if there is a black pixel in any of the lower left, lower, or lower right of the connected black pixels in the current line, this process ends. On the other hand, if there are no black pixels in the lower left, lower and lower right of the connected black pixels in the current line, it is determined that the connected black pixel in the current line is a candidate pixel for an isolated point and a halftone dot. The process ends (step S210).

  According to the present embodiment, candidate pixels are extracted by comparing the number of pixels and the center position of the connected black pixels 32 between the current line and the previous line, and the isolated points are based on the maximum number of pixels and the number of connected lines. Since the halftone dots are estimated, the halftone dots can be detected correctly even in a region where only one or two halftone dots exist. Accordingly, the original image, the background portion B2, and the latent image portion B3 are detected based on the binary image data in which the copy-forgery-inhibited pattern image B1 is embedded, and the density is varied between the original image, the background portion B2, and the latent image portion B3. It is possible to improve halftone dot detection accuracy when converted into value image data and output.

  In addition, since binary image data for at least two lines is stored and the number of connected black pixels 32 and the center position are compared between the current line and the previous line, isolated points and halftone dots are estimated. Compared to the conventional detection method for detecting halftone dots by judging the periodicity in both scanning directions, the capacity of the line buffer 1 required for detection of halftone dots can be reduced.

DESCRIPTION OF SYMBOLS 1 Line buffer 2 Buffer control part 3 Connection black pixel detection part 4 Pixel number and center position memory | storage part 5 Candidate pixel extraction part 6 Connection parameter memory | storage part 7 Four value image generation part 10 PC
20 Image processing device 21 Isolated point and halftone dot estimation unit 22 Pixel value conversion unit 23 Quaternary quaternary memory 30 Printer 31 Horizontal width of binary image 32 Concatenated black pixel 40 Copy machine 100 Background pattern addition system A1 Paper A2 Printed matter A3 Duplicate Object B1 Copy-forgery-inhibited pattern image B2 Background portion B3 Latent image portion

Claims (5)

Binary image data in which a copy-forgery-inhibited pattern image composed of a background portion in which density adjustment is performed by changing the density of isolated points and a latent image portion in which density adjustment is performed by changing the size of halftone dots is embedded as a quaternary image. An image processing apparatus that converts data to output and outputs the data,
A binary image storage unit for holding the binary image data for at least two lines in the horizontal scanning direction;
A connected black pixel detection unit that extracts black pixels from the binary image data and detects the number and center position of the connected black pixels including black pixels adjacent in the horizontal scanning direction;
Extract candidate pixels as candidates for the isolated points and halftone dots from the connected black pixels up to the current line by comparing the detection result of the current line with the detection result of the previous line by the connected black pixel detection unit. A candidate pixel extraction unit to perform,
The isolated points and halftone dots are estimated from the candidate pixels based on the maximum number of pixels and the number of connected lines related to the connected black pixels adjacent in the vertical scanning direction, and according to the original image, the isolated points, and the halftone dots. An image processing apparatus comprising: a quaternary image generation unit that assigns gradation density and generates the quaternary image data.   The image processing apparatus according to claim 1, wherein the candidate pixel extraction unit excludes connected black pixels whose number of pixels exceeds an upper limit size in a horizontal scanning direction of the halftone dots from candidate pixels.   2. The image processing according to claim 1, wherein the candidate pixel extraction unit excludes, from the candidate pixels, connected black pixels whose number of connected lines up to the current line exceeds an upper limit size of the halftone dot in the vertical scanning direction. apparatus.   The said candidate pixel extraction part resets the said maximum value and the said number of connected lines about the connection black pixel of the present line in which the black pixel adjacent to a previous line does not exist, The said number of connection lines is characterized by the above-mentioned. The image processing apparatus according to any one of the above.   The quaternary image generation unit determines a candidate pixel having the maximum value and the number of connected lines of 1 for the current line as an isolated point, and the maximum value and the number of connected lines are both N (N is 3). 2. The image processing apparatus according to claim 1, wherein the candidate pixel associated with the connected black pixel which is equal to or larger than the upper limit size in the vertical scanning direction of the halftone dot is determined as the halftone dot.

Images