JP2009100230A - Image processing apparatus and method thereof - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To create a halftone screen which is applicable to a low-resolution electrophotographic printer. <P>SOLUTION: Control points of the halftone screen are input(S20), the halftone screen is Voronoi-segmented into polygons surrounding the control points (S21, S22) and dot-lattice points included in respective polygons in the halftone screen are displaced non-periodically and at random (S24). Then, Voronoi polygons each of which includes the displaced dot-lattice point are generated (S25) and a threshold matrix is created for making each halftone dot grow around while taking the centroid of each Voronoi polygon as the center of growth in a shape approximately similar to the Voronoi polygon. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to creation of a halftone screen and image processing using the halftone screen.

  An electrophotographic printing apparatus having an exposure process in which a light beam is scanned to remove charges from a uniformly charged surface layer of a photosensitive drum such as an organic photoconductor (OPC) or amorphous silicon has nonlinear characteristics. In addition, the complexity of electrophotographic processes such as development, transfer, and fixing are factors of nonlinear characteristics.

  This non-linear characteristic causes interference between printed dots. For example, even if an independent dot is printed, it is difficult to record the dot, and the dot is surely recorded when a cluster state of several dots is reached. In addition, if the distance between dots is small, the toner may move and the dots may be connected. In addition, as in the inkjet method, the process of recording dots by attaching ink droplets to media has micro phenomena between the ink and the media, but interference between printed dots is unlikely to occur, and dots are recorded reliably. can do.

  The printed image by the electrophotographic printing apparatus is greatly influenced by the fluctuation of the spatial frequency due to the non-linear characteristic of the electrophotographic process. When forming a halftone image by an electrophotographic process, halftone of a halftone dot system is used in consideration of nonlinearity. If halftone dots are used, the basic spatial frequency is fixed, and dots can be recorded stably without being affected by variations in the spatial frequency. For example, if the number of halftone dots is N lines / inch, then the halftone dot pitch P is 25.4 / N [mm]. That is, the fundamental spatial frequency is 1 / (2P) = N / (2 × 25.4), which means that the fundamental spatial frequency is fixed. Therefore, if the printing apparatus is designed so that the electrophotographic process is always stabilized at the fundamental spatial frequency, a printed image can be stably formed. For example, in a 1200 dpi printer, the basic spatial frequency of a 200 line / inch halftone screen is 4 cycles / mm. In other words, if the electrophotographic process is stabilized at a spatial frequency of 4 cycles / mm, the image reproduction characteristics of the printing apparatus are improved.

  The AM modulation method based on the halftone dot method provides stable image reproduction characteristics. On the other hand, in color printing, moire tends to occur when toners of CMYK colors are overlapped. In order to suppress moire, the screen angle is changed for each color component, and the moire beat generated between the color components is driven to the high frequency side so that the moire is not visually noticeable. For example, the screen angle of Y is set to 30 degrees, and the screen angles of C, M, and K are set to 0 or 60 degrees to suppress moire due to overlapping of color toners.

  In digital halftone processing, the screen angle cannot be arbitrarily set because the resolution of the digital image is discrete. However, moire can be suppressed by selecting an optimal and discrete screen angle for each color component.

  However, even if the screen angle is introduced and optimized, a unique pattern caused by overlapping color components remains only by driving the moire beat to the high frequency side. This is a so-called rosette pattern, which becomes an obstacle when outputting a high-quality image. In particular, when outputting a high-quality photographic image, smooth image quality reproduction is required like a silver salt process photograph, and the rosette pattern is a major obstacle.

  As another approach, there is a method of reproducing gradations by an FM modulation method using error diffusion or a blue noise mask. The FM modulation method is widely used in the ink jet method and the thermal transfer method because the arrangement of printed dots is random, the gradation is good, and there is no moiré due to overlapping of color components. However, in the FM modulation method, the dot interval changes, and the dot interval cannot be freely controlled. For example, the dot interval gradually decreases as the density value increases. For this reason, the spatial frequency characteristic reaches a high frequency and is influenced by the frequency characteristic of the printing apparatus. Therefore, the FM modulation method is not suitable for an electrophotographic printing apparatus that is easily affected by spatial frequency fluctuations.

  As a method for improving the above problems, a hybrid halftone method has attracted attention. This method is an intermediate method between the AM modulation method and the FM modulation method, and has both features. In the hybrid halftone method, dots are clustered like halftone dots, and the interval between dots is changed. Since the variation in the dot interval is irregular and anisotropic, the spatial correlation between the dots decreases, and the generation of moire is suppressed.

  The hybrid halftone method is disclosed in, for example, Non-Patent Document 1, Patent Document 1, and the like. According to the techniques described in these documents, dots are clustered like halftone dots, the periodicity is lost by moving the lattice points (the center position of the halftone dots) by random numbers and rotation operations, and the generation of moiré is reduced. Suppress. Various methods have been proposed to create a hybrid halftone screen, both of which feature the advantages of the AM modulation method (printing stability) and the advantages of the FM modulation method (moire elimination, high resolution), and a high number of lines. High-quality image output can be expected.

  However, the hybrid halftone method targets a high-resolution printer such as 2400 dpi and assumes that one halftone dot is formed with a high resolution. On the other hand, the resolution of an electrophotographic printer for office use is about 600 dpi, and it is difficult to achieve both a high number of halftone lines and sufficient gradation characteristics.

  For example, in a 600 dpi electrophotographic printer, if the target is a hybrid halftone screen (hereinafter referred to as HHS) of 200 lines or more, in the AM system, the configuration of one halftone dot is 3 × 3 pixels. The number of gradations obtained is only 10. Also, the gradation is about 37 in a 1200 dpi printer. Furthermore, if black dots and white dots are arranged in a checkered pattern in order to avoid crushing of the white area in the shadow portion, the apparent number of lines further decreases.

  In addition, the hybrid halftone method rotates the dots randomly to make the angle direction isotropic and suppress the generation of moire. However, unlike the FM modulation method, the dot non-periodicity is completely eliminated. is not. That is, since there are locally regular lattice points, if the screens in the same direction overlap by chance in each color, local moire occurs.

JP 2005-136612 JP “Aperiodic cluster-type color halftone screen” Journal of the Institute of Image Electronics Engineers of Japan, Vol. 35, No. 5, pp. 566-575 (2006)

  An object of the present invention is to create a halftone screen applicable to an electrophotographic low-resolution printer.

  Another object of the present invention is to create a halftone screen that suppresses the generation of moiré and provides good gradation reproduction.

  The present invention has the following configuration as one means for achieving the above object.

  The image processing according to the present invention inputs halftone screen control points, divides the halftone screen into polygons surrounding the control points, and includes a halftone screen halftone grid included in the polygons. A point is aperiodically and randomly displaced to generate a Voronoi polygon including a halftone dot after the displacement, and is substantially similar to the Voronoi polygon with the centroid of the Voronoi polygon as the growth center A threshold matrix for growing halftone dots in a shape is created.

  According to the present invention, a halftone screen applicable to an electrophotographic low-resolution printer can be created.

  In addition, it is possible to create a halftone screen that suppresses the generation of moire and obtains good gradation reproduction.

  Hereinafter, image processing according to an embodiment of the present invention will be described in detail with reference to the drawings.

[Device configuration]
FIG. 1 is a block diagram illustrating a configuration example of an image processing apparatus according to an embodiment.

  The functions of a multifunction peripheral (MFP) 10 having a scanner 11 and an electrophotographic printer 12 are controlled by a built-in controller 13.

  The CPU 17 of the controller 13 executes an operating system (OS) and various programs stored in the ROM 14 and the hard disk drive (HDD) 16 using the RAM 15 as a work memory. The HDD 16 stores programs such as a control program and an image processing program, and image data.

  The CPU 17 displays a user interface on the display unit 18 and inputs a user instruction from a software key of the display unit 18 or a keyboard of the operation panel 19. For example, when the user instruction indicates copying, a document image read by the scanner 11 is printed by the printer 12 (copy function).

  Although not shown in the drawing, the communication unit 20 is a communication interface that connects to a public line or a network. If the user instruction indicates facsimile transmission, the CPU 17 controls the communication unit 20 to transmit the original image read by the scanner 11 to the destination designated by the user (facsimile function). When the user instruction indicates push scan, the document image read by the scanner 11 is transmitted to the designated server device by controlling the communication unit 20 (push scan function). When the communication unit 20 receives a facsimile image, the CPU 17 prints the received image with the printer 12 (facsimile function). When the communication unit 20 receives a print job, the CPU 17 prints an image with the printer 12 according to the print job (printer function). When the communication unit 20 receives a pull scan job, the CPU 17 transmits a document image read by the scanner 11 to the designated server device or client device according to the scan job (pull scan function).

● Dot Generator FIG. 2 is a block diagram showing a configuration example of a dot generator for generating HHS. The dot generator is configured as a part of the controller 13.

  The synchronization signal input unit 35 receives from the printer 12 a horizontal synchronization signal Hsync indicating the scanning timing of one line, a vertical synchronization signal Vsync indicating the scanning timing of one page, and a pixel clock Vclock as the synchronization signals. These synchronization signals are sequentially input to the image memory 31 assigned to the RAM 15, and image data corresponding to the scanning position of the photosensitive drum 30 is output. The synchronization signal is sequentially input to a dither threshold memory 32 assigned to the ROM 14 or RAM 15, and a threshold matrix described later corresponding to the scanning position of the photosensitive drum 30 is output.

The comparator 33 inputs the image data corresponding to the scanning position of the photosensitive drum 30 and the threshold matrix, compares the threshold of each cell of the image data and the threshold matrix, and outputs a binary signal according to the following formula for each cell. Output.
When D ≥ Th0, output signal = '1'
When D <Th0, output signal = '0' (1)
Where Th0 is a threshold,
D is image data

  The laser driver 34 controls the light emission of the beam light source 27 by driving the beam light source 27 according to the binary signal output from the comparator 33. That is, when the output signal of the comparator 33 is ‘1’, the light beam 26 is output to the beam light source 27 (laser on), and when it is ‘0’, the light beam 26 is not output to the beam light source 27 (laser off).

[Create HHS]
FIG. 3 is a flowchart for explaining processing for creating an HHS, which is executed by the CPU 17.

  The CPU 17 inputs a control point to generate an HSS (S20). The control point is not a point indicating the position of a halftone dot, but a point sequence for determining the arrangement state of the halftone grid. The control points are randomly arranged with an interval larger than the average interval of the finally obtained halftone lattice.

  FIG. 4 is a diagram showing an example of arrangement of control points. Control points 137 to 145 are randomly arranged in the M × N pixel region 130. In addition, the method of randomizing the arrangement of control points is, for example, a method in which a random number is given to each control point of a square lattice that is regularly arranged in the initial stage, and a two-dimensional movement according to the random number is given to the control point, There is a method of randomly arranging the control points by rotating the control points.

  The control points of the M × N pixel region 130 are set as a set, and the set of control points is two-dimensionally arranged periodically. The reason for this is to make a halftone dot grid, which will be described later, periodic in units of M × N pixels, so that the threshold matrix fits in M × N pixels.

  Next, the CPU 17 creates a Delaunay triangle connecting adjacent control points (S21), creates a Voronoi polygon from the Delaunay triangle (S22), and determines the center of gravity of the Voronoi polygon ( S23).

  FIG. 5 is a diagram showing a set of Delaunay triangles connecting control points 137 to 145. The Delaunay triangle is configured such that control points other than the control points corresponding to the vertices do not enter the circumscribed circle. In other words, the Delaunay triangle is uniquely determined when a control point is given.

FIG. 6 is a diagram showing a set of Voronoi polygons connecting the outer center points of the Delaunay triangle. The Voronoi polygon for the control point Pi is a set of points V (Pi) expressed by the following equation, and the Voronoi polygon represents the sphere of influence of the control point Pi.
V (Pi) = (P | d (P, Pi) <d (P, Pj), ∀j ≠ i}… (2)
Where Pj is the control point adjacent to Pi,
d (P, Pi) is the distance between point P and control point Pi,
d (P, Pj) is the distance between point P and control point Pj

The sides of the Voronoi polygon (Voronoi boundary) connect the outer points of the Delaunay triangle, and the Delaunay triangle is uniquely determined with respect to the control point. Therefore, the Voronoi polygon is also uniquely determined. The outer center point (Gx, Gy) of the Delaunay triangle is expressed by the following equation.
Gx = {X0 ² (Y1-Y2) + X1 ² (Y2-Y0) + X2 ² (Y0-Y1)-(Y0-Y1) (Y1-Y2) (Y2-Y0)} / L
Gy = -Gx (X2-X1) / (Y2-Y1) + (X2-X1) (X1 + X2) / 2 / (Y2-Y1) + (Y1 + Y2) / 2… (3)
Where L = 2 {X0 (Y1-Y2) + X1 (Y2-Y0) + X2 (Y0-Y1)},
(X0, Y0), (X1, Y1), (X2, Y2) are the coordinates of the vertex of the Delaunay triangle

  The CPU 17 forms a Voronoi polygon having one control point inside, and determines the center of gravity of each Voronoi polygon. FIG. 7 is a diagram showing the relationship between the center of gravity of the Voronoi polygon and the control points. The center of gravity indicated by the x mark depends on the shape of the Voronoi polygon, and is usually at a position different from the control point (indicated by the mark).

  Next, the CPU 17 gives fluctuation and rotation about the center of gravity of the Voronoi polygon to the initial halftone dot sequence (S24).

  FIG. 8 is a diagram showing a state in which the initial halftone dot sequence is superimposed on the Voronoi polygon. The interval p between halftone dot lattice points is set according to the number of lines in the HSS, and is set more finely as the number of lines increases. Note that FIG. 8 shows a square halftone dot sequence, but if it is desired to be anisotropic, a rectangular halftone dot sequence may be used. The initial halftone dot sequence represents the characteristics of the final halftone dot, although it is an initial condition, and should be selected carefully.

  The CPU 17 gives fluctuation to each halftone dot and displaces the position of each halftone dot in the two-dimensional x and y directions. This operation is for giving fluctuation to the interval p, eliminating the periodicity of the halftone dot sequence, and arranging the halftone dot randomly.

Next, the CPU 17 gives the rotation of the angle set for each Voronoi polygon to the halftone dot lattice point to which the fluctuation is given. FIG. 9 is a diagram showing the halftone dot after rotation. As shown in Table 1 as an example of the rotation angle of each control point, a different rotation is given to each control point (Voronoi polygon). When the initial halftone dot sequence is a square lattice, the rotation angle is 0 to 90 degrees.
table 1
──────┬┬─┬┬─┬┬─┬┬──┬──┬──┬──┬──┬──
Control point │ 137 │ 138 │ 139 │ 140 │ 141 │ 142 │ 143 │ 144 │ 145
──────┼┼─┼┼─┼┼─┼┼──┼──┼──┼──┼──┼──
Rotation angle θ ₀ ° │ 60 │ 45 │ 30 │ 70 │ 60 │ 30 │ 45 │ 50 │ 45
──────┴┴─┴┴─┴┴─┴┴──┴──┴──┴──┴──┴──

The halftone dot in the Voronoi polygon is rotated around the center of gravity of the Voronoi polygon at the rotation angle set for the corresponding control point. The coordinates of the halftone dot after rotation are expressed by the following equation.
Px '= (Px-Jx) cosθ-(Py-Jy) sinθ + Jx
Py '= (Px-Jx) sinθ + (Py-Jy) cosθ + Jy… (4)
Where (Px, Py) is the coordinates of the halftone dot before rotation,
(Px ', Py') is the coordinates of the halftone dot after rotation,
(Jx, Jy) is the coordinate of the center of rotation (the center of gravity of the Voronoi polygon)

In equation (4), θ is the rotation angle, and decreases as the distance from the rotation center increases, as shown in the following equation.
θ = θ ₀ (1-r / r ₀ )… (5)
Where r is the distance from the center of rotation to the halftone dot (Px, Py),
r ₀ is the radius of the circumscribed circle of the Voronoi polygon,
θ ₀ is the rotation angle at the center of rotation

FIG. 10 is a diagram showing variations of the rotation operation. FIG. 10 (a) shows the arrangement of the halftone dot lattice points after rotating at the rotation angle θ according to the equation (5). There is a discontinuity of the primary differential value at the boundary r ₀ and the boundary is conspicuous.

FIG. 10 (b) shows an arrangement of halftone dot lattice points after rotating at a rotation angle θ according to the following equation. The closer r is to the boundary r ₀ , the greater the displacement of the halftone dot and the more noticeable the boundary.
θ = θ ₀ {1-(r / r ₀ ) ² }… (6)

FIG. 10 (c) shows the arrangement of halftone dot lattice points after rotation at a rotation angle θ according to the following equation. The primary differential value at the boundary r ₀ is continuous, and a smooth displacement is obtained at the boundary.
θ = θ ₀ [1-sin ² {(1/2) × πr / r ₀ }]… (7)

  These rotation operations are preferably used according to the purpose. Use Equation (6) (Fig. 10 (b)) to greatly disturb the halftone grid, and use Equation (7) (Fig. 10 (c)) to give a smooth displacement to the halftone grid. And preferred.

  Due to the rotation, the halftone dot in the Voronoi polygon may protrude outside the Voronoi polygon. When the halftone dot is outside the Voronoi polygon, a blank portion of the halftone dot is generated in the boundary region of the Voronoi polygon. As described above, since the M × N pixel Voronoi polygons are two-dimensionally arranged so that the threshold value matrix fits in the M × N pixels, the generation of blank portions is inconvenient. Therefore, after rotation, we extract halftone grid points (hereinafter referred to as cross-border grid points) that protrude from the Voronoi polygon and enter the adjacent Voronoi polygon, and cross-border grid points are the halftone dots of the Voronoi polygon into which the cross-border grid points have intruded. Use grid points. Alternatively, the halftone dot grid points that protrude from the Voronoi polygon are folded in line symmetry at Voronoi boundaries (sides of the Voronoi polygon) and placed inside the Voronoi polygon.

  FIG. 11 is a view showing the halftone dot after rotation. However, for the sake of explanation, the density of the halftone dot is lower than the halftone dot shown in FIG. As shown in FIG. 11, halftone dot lattice points located around the Voronoi boundary are not uniform because they approach or separate from the halftone dot points of the adjacent Voronoi polygon. This is because random rotation of Voronoi polygon units cannot control the density of halftone lattice points. In this state, the dispersion of halftone dots is not uniform and the image quality is deteriorated.

  Therefore, the CPU 17 corrects the position of the halftone dot that is too close. That is, the Voronoi polygon surrounding the halftone dot is obtained to determine the center of gravity of the Voronoi polygon (S25), and the process of moving the halftone dot to the center of gravity (S26) is repeated as many times as necessary (S27).

  FIG. 12 is a diagram for explaining the correction of the positions of halftone lattice points. In the Voronoi polygon having halftone lattice points 101 to 106 shown in FIG. 12A, there are halftone lattice points 101 and 102, 103 and 104, and 105 and 106 that are too close. FIG. 12B shows the centroids 101a, 102a, 103a, 104a, 105a, and 106a of each Voronoi polygon. The CPU 17 replaces the halftone grid points 101 to 106 with the centroids 101a to 106a, corrects the positions of the halftone grid points, and improves halftone grid points that are too close. FIG. 13 is a diagram showing halftone dot lattice points as a result of repeating the correction of the halftone dot positions (S25 and S26) three times with respect to the halftone dot dots in FIG. The number of repetitions varies depending on the degree of density of the halftone lattice points, but usually it may be repeated about 3 to 5 times.

  Next, as a result of correcting the halftone dot that has been too close, the CPU 17 folds the halftone dot that protrudes from the M × N pixel region 130 in a line-symmetric manner to the side of the M × N pixel region 130 (S28). By this operation, the number of pixels in the M × N pixel region 130 is maintained.

[Create threshold matrix]
FIG. 14 is a flowchart for explaining processing for creating a threshold matrix.

  The CPU 17 inputs halftone dot information (S30), and creates a Voronoi polygon including the halftone dot (S31).

  FIG. 15 is a diagram showing a set of Voronoi polygons including halftone lattice points. Each Voronoi polygon corresponds to a halftone dot. FIG. 16 shows a halftone dot pattern at a halftone dot area ratio of 70%. The halftone dot grows substantially similar to the Voronoi polygon with the center of gravity of the Voronoi polygon as the growth center. That is, the CPU 17 creates a threshold value matrix corresponding to such halftone dot growth (blackening).

  17 and 18 are enlarged views of halftone dots 120 to 124. FIG. 17 shows a halftone dot area ratio of 30%, and FIG. 18 shows a halftone dot area ratio of 70%. The center of blackening is the center of gravity of the Voronoi polygon, and the halftone dot is substantially similar to the Voronoi polygon regardless of the degree of blackening. The halftone grid points 120 to 124 are at positions different from the center of gravity of the growth center.

  Next, the CPU 17 classifies the Voronoi polygon according to the resolution (dot size) of the electrophotographic printer (S32), and sets the threshold matrix cells. FIG. 19 is an enlarged view of the halftone dots 120 to 124 after the division, and the Voronoi boundary (side) becomes jagged depending on the division. FIG. 20 is an enlarged view of the halftone dots 120 to 124 when the halftone dot area ratio is 40%. The halftone dots are made to be similar to the Voronoi polygon, but they are not completely similar due to the effect of the division according to the resolution.

  When the resolution of the printer is low, the Voronoi polygon does not contain a sufficient number of cells, and a sufficient number of gradations cannot be obtained. Considering a 600 dpi printer, when 200 lines are secured, the average halftone dot spacing is 3 cells, and one Voronoi polygon has about 9 to 10 cells. Therefore, the number of gradations that can be expressed by halftone dots is about 10. Even with a 1200 dpi printer, 37 gradations are not enough. On the other hand, image data input from a digital camera or scanner has 256 gradations of 8 bits for each color. In order to reproduce this gradation, we want at least 256 gradations, and we want 300 to 512 gradations to guarantee a wide dynamic range by applying gamma conversion and various color conversions.

  Therefore, the CPU 17 groups the Voronoi polygons (S33), and creates a threshold matrix for each Voronoi polygon according to the group to which it belongs (S34).

  FIG. 21 is a diagram illustrating grouping of Voronoi polygons. For example, classify Voronoi polygons into a first group and a second group, create a threshold matrix of thresholds a1, a2, a3,..., AS for the first group and threshold b1 for the second group. , B2, b3,..., BT threshold matrix. Both threshold values are alternately arranged as shown in FIG. When the Voronoi polygons are classified into two groups, for example, the Voronoi polygons with odd numbers of halftone lattice points shown in FIG. 20 are the first group, and the even Voronoi polygons are the second group. It can be divided into

  If the halftone dots of the two groups are integrated, the number of gradations S + T + 1 is obtained. Further, if the number of groups is increased, the number of gradations increases, and a high gradation number can be obtained even in a low-resolution printer. However, since the area of each Voronoi polygon is different and the number of cells included in one Voronoi polygon is different, it is necessary to adjust the number of thresholds. For example, if the number of cells of a Voronoi polygon with the first group is R (<S), the threshold values a1, a2, a3,..., AS are thinned out to, for example, a1, a3, a5,. Thus, even if the threshold value is thinned out, it does not overlap with the threshold value of the second group.

  In this way, the CPU 17 creates Voronoi polygon threshold matrices for the entire HHS, and stores these threshold matrices in, for example, the HDD 16 in association with the positions of the halftone lattice points before displacement.

[Increase the number of gradations]
FIG. 22 is a diagram for explaining setting of threshold matrices for two Voronoi polygons belonging to different groups.

  The Voronoi polygon shown in FIG. 22 (a) has 51 cells, the Voronoi polygon shown in FIG.22 (b) has 36 cells, and the numbers attached to each cell are the Voronoi polygons. The order of blackening so as to be substantially similar is shown.

  If the threshold matrices of both Voronoi polygons are set as shown in FIGS. 22 (c) and 22 (d), 51 + 36 + 1 = 88 gradations can be expressed in the dynamic range of 0 to 255. However, the target 300-512 gradation is insufficient. In order to solve this, a method of increasing the number of groups is conceivable. However, when the number of groups is increased, the resolution is lowered. Therefore, as shown in FIGS. 22 (e) and 22 (f), two types of threshold matrices are set in the Voronoi polygon of FIG. 22 (b). The threshold value matrix of FIG. 22 (e) is the same as that of FIG. 22 (d), but the threshold value matrix of FIG. 22 (f) is different from FIG. 22 (d). The threshold values in the threshold matrix in FIG. 22 (f) are different from the threshold values in the threshold matrix set in the Voronoi polygon in FIG. 22 (a).

  FIG. 23 is a block diagram illustrating a configuration example of a dot generator when the number of gradations of a print image is increased. A configuration different from FIG. 2 is that there are two dither threshold memories 32a and 32b. That is, for a certain Voronoi polygon, the threshold matrix shown in FIG. 22 (d) is stored in the dither threshold memory 32a, and the threshold matrix shown in FIG. 22 (f) is stored in the dither threshold memory 32b.

The comparator 33 inputs the image data corresponding to the scanning position of the photosensitive drum 30 and two threshold matrixes, compares the image data with the threshold value of each cell, and outputs a binary signal according to the following formula for each cell. To do.
Output signal = '2' when Th0 ≤ D
When Th1 ≤ D <Th0, output signal = '1' (8)
When D <Th1, output signal = '0'
Where Th0 and Th1 are thresholds,
D is image data

  The laser driver 34 controls the light emission of the beam light source 27 in accordance with the ternary signal output from the comparator 33. That is, as shown in FIG. 24A, when the output signal of the comparator 33 is ‘2’, the beam light source 27 outputs the light beam 26 having the pulse width T. Further, in the case of ‘1’, the light beam 26 having the pulse width T / 2 is output to the beam light source 27, and in the case of ‘0’, the light beam 26 is not output to the beam light source 27. In this way, the number of gradations is increased by pulse width modulation, and (51 + 36) × 2 + 1 = 175 gradations can be obtained in the example shown in FIG. Note that the target 300 gradations are achieved by combining halftone grid points, increasing the number of groups, and increasing multilevel (in the above example, ternarization, further quaternarization, etc.).

  However, as shown in FIG. 24 (b), the light intensity peak value with the pulse width T / 2 is lower than the light intensity peak value with the pulse width T, and the exposure with the pulse width T / 2 is very unstable. It is. Therefore, the laser driver 34 emits light from the beam light source 27 so that the dots formed with the pulse width T / 2 (hereinafter, half dots) are adjacent to the dots formed with the pulse width T (hereinafter referred to as all dots). Control timing.

  For example, FIG. 24 (c) is an example in which the light emission timing of the beam light source 27 is controlled so that a half dot is formed on the left side of the dot width and then all the dots are formed. Since an isolated half dot is formed, the half dot is very unstable. On the other hand, FIG. 24 (d) is an example of controlling the light emission timing of the beam light source 27 so that a half dot is formed on the right side of the dot width and then all dots are formed. Therefore, half dots can be stably formed.

[Gradation image]
Fig. 25 is an enlarged view of the gradation formed by the halftone dots, and blackening starts from the center of gravity of each Voronoi polygon corresponding to the halftone dot, and each halftone dot gradually becomes similar to the corresponding Voronoi polygon. Shows how it blackens.

  The periodicity of halftone dots disappears when a fluctuation or rotation operation is given to the halftone dot sequence, and each halftone dot has a random arrangement. However, the average interval of halftone dots is the interval of the halftone dot sequence given as the initial condition, and the average interval does not change significantly unlike the FM modulation method. When the HHS of this example is evaluated in the frequency space, the peak of the primary frequency spectrum is prominent and is distributed concentrically by the rotation operation. Therefore, it appears that the positions of primary peaks found in the AM modulation system are averaged concentrically.

[Rotation operation for each color component]
The rotation operation described in FIGS. 8 and 9 is set as follows for each color plate. The rotation operation in Table 1 described above is a rotation operation of the yellow component (Y plate).
Table 2 (C version)
──────┬┬─┬┬─┬┬─┬┬──┬──┬──┬──┬──┬──
Control point │ 137 │ 138 │ 139 │ 140 │ 141 │ 142 │ 143 │ 144 │ 145
──────┼┼─┼┼─┼┼─┼┼──┼──┼──┼──┼──┼──
Rotation angle θ ₀ ° │ 60 │ 45 │ 30 │ 70 │ 60 │ 30 │ 45 │ 50 │ 45
──────┴┴─┴┴─┴┴─┴┴──┴──┴──┴──┴──┴──
Table 3 (M version)
──────┬┬─┬┬─┬┬─┬┬──┬──┬──┬──┬──┬──
Control point │ 137 │ 138 │ 139 │ 140 │ 141 │ 142 │ 143 │ 144 │ 145
──────┼┼─┼┼─┼┼─┼┼──┼──┼──┼──┼──┼──
Rotation angle θ ₀ ° │ -60│ -45│ -30│ -70│ -60│ -30│ -45│ -50│ -45
──────┴┴─┴┴─┴┴─┴┴──┴──┴──┴──┴──┴──
Table 4 (K version)
──────┬┬─┬┬─┬┬─┬┬──┬──┬──┬──┬──┬──
Control point │ 137 │ 138 │ 139 │ 140 │ 141 │ 142 │ 143 │ 144 │ 145
──────┼┼─┼┼─┼┼─┼┼──┼──┼──┼──┼──┼──
Rotation angle θ ₀ ° │ 30 │ 15 │ 45 │ 20 │ 45 │ 70 │ 30 │ 65 │ 60
──────┴┴─┴┴─┴┴─┴┴──┴──┴──┴──┴──┴──

  Since the halftone dot is fluctuated, the correlation for each color plate is lowered, but the correlation is further lowered by giving different rotation angles. For the Y version, since the MTF value of the yellow visibility is low, there is no problem with using any threshold matrix. However, when priority is given to gradation, a halftone dot grid is utilized by taking advantage of the characteristics of electrophotography. It is better to aim for stable output by widening the average interval between dots (rough halftone dots). For this reason, even if the rotation angle is the same as that of the C plate, the interval of the halftone dot is different, the dots of the C plate do not completely overlap with the dots of the Y plate, and no moire occurs in the secondary color. .

  FIG. 26 is a diagram showing halftone dot points immediately after rotation of the C plate (FIG. 26 (a)) and the M plate (FIG. 26 (b)). As described above, in order to perform “correction of dot lattice points that are too close”, the dot lattice points that are away from the center of rotation are corrected to have a random positional relationship. However, since the halftone dot near the rotation center is not corrected much, it is necessary to remove the moire between the color components by the rotation operation. In this embodiment, the C and M plates rotate in opposite directions, so that the patterns of halftone dot after rotation are greatly different, and the occurrence of moire between color components is low.

  FIG. 27 is a diagram showing a rotation operation based on another concept, in which the initial halftone dot dot iron of the M plate has a screen angle. Even if the same rotation angle as that of the C plate is used, the initial state is different, so the pattern of the halftone dot after rotation is different, and as a result, no moire occurs between the color components. The initial halftone dot sequence can take many variables / parameters such as lattice spacing and screen angle. In general, it is desirable to perform the rotation operation on the basis of screen arrangements having angles different by 30 degrees so that moire between color components does not occur in the initial halftone dot array.

  In the above description, the controller 13 of the printer 10 has been described as generating HHS. However, an HHS having an average lattice interval corresponding to the number of lines of the image to be formed may be created in advance and stored in the HDD 16 or the like, and the HHS necessary for image formation may be loaded into the dither threshold memory 32 and used.

  As described above, by using the HHS of this embodiment, it is possible to print a high-quality image having a high number of lines, high gradation, and no moire in a printer having a relatively low resolution such as 400 dpi or 600 dpi.

  The image processing according to the second embodiment of the present invention will be described below. Note that the same reference numerals in the second embodiment denote the same parts as in the first embodiment, and a detailed description thereof will be omitted.

  FIG. 28 is a diagram showing halftone dot lattice points in the M × N pixel region 130 in the second embodiment.

  Halftone dot 131 belongs to the first group (for black dots), halftone dot 132 belongs to the second group (for black dots), and halftone dot 133 belongs to the third group (for white dots). . The halftone dot points of these groups are arranged in a random state after being subjected to fluctuation and rotation operations. At that time, it is an important factor for determining the final image quality that the halftone dot is uniformly distributed within each group and that the halftone dot is uniformly distributed as a whole. The grouping is determined before the initial halftone dot sequence is introduced, and then the fluctuations and rotations are performed, or the fluctuations and rotations are performed without grouping at first, and then divided into groups. May be.

  FIG. 29 is a diagram showing a state where the M × N pixel region 130 is divided by Voronoi. For each Voronoi polygon, the center of gravity (marked with x) is obtained and becomes the center of blackening of the halftone dot.

  FIG. 30 is a diagram showing a halftone dot pattern at medium density (tone level 150). A threshold value matrix is set so that the threshold values of the halftone dots of the first and second groups do not overlap each other. Blackening of the third group of halftone dots (white dots) starts from the periphery of the Voronoi polygon after 100% of the blackening of the halftone dots of the first and second groups has been completed. The gradation level at which white dots start to be blackened is determined by the area ratio of white dots to black dots. If the number of white dots and black dots is approximately equal, and the total area of white dots and black dots are approximately equal, for example, 8 bits, black dot blacking is finished at gradation level 128. Then, the blackening of the white dots starts from the gradation level 129.

  Usually, when there are many white dots, the distribution of black dots is affected, and it is difficult to obtain a visually uniform and uniform halftone dot pattern. For this reason, in general, the number of white dots is limited to a smaller number than that of black dots. If the number of white dots is about 1/4 of the total halftone (area ratio is about 25%), blackening of the black dots ends at 3/4 of the dynamic range (gradation level 192), and the black of the white dots Starts at around the gradation level 193. FIG. 31 is a diagram showing a threshold matrix for white dots. Blackening is started from gradation level 195, and while maintaining a halftone dot shape (white dot) substantially similar to the Voronoi polygon, a halftone dot (white dot) from the periphery of the Voronoi polygon toward the center of gravity (reduction center) ).

  FIG. 32 is an enlarged view showing a gradation formed by halftone dots. In the gradation formed by the halftone dots not including white dots shown in FIG. 25, white thin lines with a width of one or two dots remain around the black dots in the shadow portion (high density region). This is an inevitable phenomenon in the blackening process at the Voronoi boundary. When white dots are introduced, the area ratio of the Voronoi boundary decreases, and the gradation of the shadow part is reproduced by the area of the white dots, so that white fine lines are not left around the black dots as shown in FIG. be able to.

  It is preferable to assign a Voronoi polygon with a small area to the third group (white dots). The reason is that if there is a non-blackened portion (including white dots) having a large area when black dots are blackened, the uniformity of the entire image is lost and a visually non-uniform pattern is exhibited. Therefore, it is preferable to assign a Voronoi polygon having a small area to the white dot in order to improve the image quality. Therefore, after the positions of halftone lattice points are finally determined, the area of each Voronoi polygon may be obtained, and a predetermined number of Voronoi polygons may be selected in ascending order of area. However, it is determined whether or not the selected Voronoi polygon is uniformly distributed without being locally biased. If there is a skew, it is necessary to take another measure such as selecting another Voronoi polygon.

  Furthermore, in order to increase the number of gray levels of white dots, it is also effective to divide white dots into a plurality of groups and set a threshold matrix in which thresholds do not overlap between groups, as with black dots.

  Hereinafter, image processing according to the third embodiment of the present invention will be described. Note that the same reference numerals in the third embodiment denote the same parts as in the first and second embodiments, and a detailed description thereof will be omitted.

  In the following, an example will be described in which additional information is embedded in image data using the rotation angle of a halftone dot. Additional information embedded in image data includes copyright information (name of photographer, creator, owner, etc., creation date / time), confidential information (tracking information such as printer device number, print date, print instructor, etc.), Meta information etc. When embedding these additional information, it is necessary to maintain the image quality while minimizing the influence on the image quality. There is also a security requirement to prevent a third party from easily reading the embedded additional information.

  FIG. 33 is a block diagram illustrating a configuration example of the dot generator according to the third embodiment. Additional information 36 is embedded in the threshold matrix read from the HDD 16 and set in the dither threshold memory 32.

  The additional information 36 represents information in byte units, and gives the rotation angle of the control point corresponding to the bit. For example, if the first byte of the additional information 36 is a character “Y”, bit information of 0x59 = ‘01011001’ needs to be embedded in the image.

FIG. 34 is a diagram showing an example of a region in which the additional information 36 is embedded. The additional information 36 is embedded by changing the rotation angle of the halftone dot in the nine Voronoi polygons. Of the 9 bits of bits 8-0, bit 8 is used as a parity bit, and embedded information (for example, '01011001') is assigned to the Voronoi polygon of bits 7-0. Since fluctuations are also given to the halftone dot, the following rule is imposed on the rotation angle θ in order to improve the separation performance and reduce the determination error.
When bit value = '0', θ <45 °
When bit value = '1' θ ≥ 45 ° (9)

The first byte (for example, character Y) of the additional information 36 is embedded in the C version as follows.
Table 5
──────┬┬─┬┬─┬┬─┬┬──┬──┬──┬──┬──┬──
Control point │ 137 │ 138 │ 139 │ 140 │ 141 │ 142 │ 143 │ 144 │ 145
──────┼┼─┼┼─┼┼─┼┼──┼──┼──┼──┼──┼──
Bit │ 8 │ 7 │ 6 │ 5 │ 4 │ 3 │ 2 │ 1 │ 0
──────┼┼─┼┼─┼┼─┼┼──┼──┼──┼──┼──┼──
Additional information Y │ 0 │ 0 │ 1 │ 0 │ 1 │ 1 │ 0 │ 0 │ 1
──────┼┼─┼┼─┼┼─┼┼──┼──┼──┼──┼──┼──
Rotation angle θ ₀ ° │ 20 │ 30 │ 60 │ 40 │ 70 │ 50 │ 30 │ 15 │ 60
──────┴┴─┴┴─┴┴─┴┴──┴──┴──┴──┴──┴──

FIG. 35 is a diagram showing a halftone dot with embedded additional information 36, and a rotation operation according to Table 5 is performed for each Voronoi polygon. In Example 1, the range r ₀ to be rotated is a circumscribed circle of a Voronoi polygon. However, in order to reduce the influence of fluctuations of halftone dot lattice points around the Voronoi polygon, it is preferable to limit r ₀ within a circle included in the Voronoi polygon. Then, the halftone dot outside the r ₀ circle is set as the halftone dot in the initial state without performing the rotation operation. In this case, only the data in the r ₀ circle of the threshold value matrix that is not embedded with the additional information stored in the HDD 16 may be rewritten according to the additional information 36.

FIG. 36 is a diagram showing a halftone dot in which, for example, the second byte (for example, the letter “K”) of the additional information is embedded. is there.
Table 6
──────┬┬─┬┬─┬┬─┬┬──┬──┬──┬──┬──┬──
Control point │ 137 │ 138 │ 139 │ 140 │ 141 │ 142 │ 143 │ 144 │ 145
──────┼┼─┼┼─┼┼─┼┼──┼──┼──┼──┼──┼──
Bit │ 8 │ 7 │ 6 │ 5 │ 4 │ 3 │ 2 │ 1 │ 0
──────┼┼─┼┼─┼┼─┼┼──┼──┼──┼──┼──┼──
Additional information K │ 0 │ 0 │ 1 │ 0 │ 0 │ 1 │ 0 │ 1 │ 1
──────┼┼─┼┼─┼┼─┼┼──┼──┼──┼──┼──┼──
Rotation angle θ ₀ ° │ 20 │ 30 │ 60 │ 40 │ 30 │ 50 │ 30 │ 75 │ 60
──────┴┴─┴┴─┴┴─┴┴──┴──┴──┴──┴──┴──

  In Table 6, the rotation angles of bits 7, 4, 2 and 6 and 0 are the same, but a more random value is set in consideration of the moire suppression according to the rule of equation (9). May be.

  FIG. 37 is a flowchart for explaining a method for extracting additional information embedded in a print image. It is assumed that information on the control point and information on the rotation angle of the color plate are known.

  The CPU 17 inputs information indicating control points corresponding to the print image, and performs Voronoi division (S151). Then, one of the Voronoi polygons created by Voronoi division is extracted and the center of gravity is obtained (S152).

  Next, the CPU 17 acquires the image data of the print image from the scanner 11, for example, associates the image data with the control points in units of M × N pixels, and corresponds to the vicinity of the center of gravity of the Voronoi polygon obtained in step S152. The coordinates of the halftone dot of the color plate are detected (S153). Then, the screen angle (rotation angle) is obtained from the detected coordinates, and if the rotation angle ≧ 45 °, it is determined as “1”, and if the rotation angle <45 °, it is determined as “0” (S154).

  The CPU 17 repeats steps S152 to S154 based on the determination in step S155, and extracts bit values from all Voronoi polygons in the M × N pixel region. Then, when the processing of a certain M × N pixel area is completed, the process proceeds to the next M × N pixel area (S156), and steps S152 to S156 are repeated according to the determination in step S157, so that the bit value from all M × N pixel areas To extract.

  The rotation angle can be detected by calculation from the dots in the vicinity of the center of gravity. However, the method of performing the Hough transform and calculating the rotation angle θ from the result is accurate. However, the determination of whether the rotation angle is larger or smaller than 45 ° does not have to be very accurate, and can be performed with a simple calculation.

  In this way, additional information can be embedded in the HSS halftone dot.

[Design guidelines]
Applying the HHS of this embodiment to an electrophotographic printer and outputting a high-quality image includes the following parameters as design items.
・ The average number of Voronoi polygons corresponding to one halftone dot is 64 (average 8 × 8 pixels) ・ The number of halftone dots of a Voronoi polygon corresponding to one control point is 64 (average) 8 × 8 pixels) or more ・ The number of control points per cycle (the number of Voronoi polygons) is 9 (average 3 × 3) or more

  Therefore, since 8 × 8 × 3 = 192, a high-quality HSS can be obtained by using a dither matrix of at least 192 × 192 pixels.

  However, the average number of lines of the output image needs to be 200 lines, and when the printer is 600 dpi, the average number of pixels of the Polonoi polygon corresponding to one halftone dot is 9 to satisfy the above condition. Therefore, in order to obtain high gradation, the Polonoi polygons corresponding to the halftone lattice points are divided into four groups, and the types of threshold matrixes in each group are increased, and the output of the comparator 33 is converted into nine values. There is a need. As a result, the number of gradations that can be expressed is 9 × 4 × 8 = 288 gradations, which can approach the target gradation number of 300. If nine-value conversion is severe, the number of groups is increased and the level of multi-value conversion is lowered.

  In the case of a 1200 dpi printer, the average number of pixels of the Polonoi polygon corresponding to one halftone dot is 36, so it is divided into 4 groups and ternarized to 36 x 4 x 2 = 288 floors. Tones can be obtained. Furthermore, if the quaternarization is performed, 36 × 4 × 3 = 432 gradations can be obtained, and a wide dynamic range can be guaranteed even when gamma conversion or various color conversions are performed.

[Other embodiments]
Note that the present invention can be applied to a system constituted by a plurality of devices (for example, a computer, an interface device, a reader, a printer, etc.), but an apparatus (for example, a copier, a facsimile machine, a control device) composed of a single device. Etc.).

  Another object of the present invention is to supply a storage medium storing a computer program for realizing the functions of the above-described embodiments to a system or apparatus, and the computer (CPU or MPU) of the system or apparatus executes the computer program. But it is achieved. In this case, the software read from the storage medium itself realizes the functions of the above embodiments, and the computer program and the computer-readable storage medium storing the computer program constitute the present invention. .

  Further, the above functions are not only realized by the execution of the computer program. That is, according to the instruction of the computer program, the operating system (OS) and / or the first, second, third,... This includes the case where the above function is realized.

  The computer program may be written in a memory of a device such as a function expansion card or unit connected to the computer. That is, it includes the case where the CPU of the first, second, third,... Device performs part or all of the actual processing according to the instructions of the computer program, thereby realizing the above functions.

  When the present invention is applied to the storage medium, the storage medium stores a computer program corresponding to or related to the flowchart described above.

A block diagram showing a configuration example of an image processing apparatus of an embodiment, Block diagram showing a configuration example of a dot generator that generates HHS, Flowchart explaining the process of creating HHS, The figure which shows the example of arrangement of a control point, A diagram showing a set of Delaunay triangles connecting control points, A diagram showing a set of Voronoi polygons connecting the Delaunay triangle centroids, Figure showing the relationship between the center of gravity of the Voronoi polygon and the control points, The figure which shows the state which superimposed the initial halftone lattice point sequence on the Voronoi polygon, The figure which shows the halftone dot after rotation, Figure showing variations of rotation operation, The figure which shows the halftone dot after rotation, The figure explaining correction of the position of a halftone dot, FIG. 11 is a diagram showing halftone dot lattice points as a result of repeating the correction of the halftone dot positions with respect to the halftone dot dots in FIG. A flowchart for explaining processing for creating a threshold matrix; A diagram showing a set of Voronoi polygons including halftone grid points, A diagram showing a halftone dot pattern at a halftone dot area ratio of 70%, An enlarged view of halftone dot of halftone dot, An enlarged view of halftone dot of halftone dot, An enlarged view of the halftone dot of the halftone dot after classification, An enlarged view of a halftone dot of a halftone dot when the halftone dot area ratio is 40%, A diagram illustrating the grouping of Voronoi polygons, The figure explaining the setting of the threshold matrix of two Voronoi polygons with different groups Block diagram showing a configuration example of a dot generator when increasing the number of gradations of a print image, The figure explaining light emission control of a beam light source, Figure showing an enlarged gradation formed by halftone dots, The figure which shows the halftone dot after the rotation of C version and M version, The figure which shows the rotation operation of another way of thinking, FIG. 5 is a diagram showing dot lattice points in the M × N pixel region in Example 2. A diagram showing how the M × N pixel region is divided into Voronois, The figure which shows the halftone dot pattern in medium density, The figure which shows the threshold value matrix for white dots, Figure showing an enlarged gradation formed by halftone dots, Block diagram showing a configuration example of a dot generator of Example 3, The figure which shows an example of the area | region which embeds additional information, A figure showing a halftone dot with embedded additional information, A figure showing a halftone dot embedded with, for example, a second byte of additional information, 6 is a flowchart illustrating a method for extracting additional information embedded in a print image.

Claims (12)

Input means for inputting the control points of the halftone screen;
Splitting means for Voronoi splitting the halftone screen into polygons surrounding the control points;
Displacement means for aperiodically and randomly displacing the halftone screen halftone dot included in the polygon,
Generating means for generating a Voronoi polygon including a halftone dot after the displacement;
An image processing apparatus comprising: a creation unit that creates a threshold matrix for growing halftone dots in a shape substantially similar to the Voronoi polygon, with the center of gravity of the Voronoi polygon as a growth center.   2. The image according to claim 1, wherein the displacing unit displaces the halftone dot by giving fluctuation to the halftone dot and rotating around the center of gravity of the polygon. Processing equipment.   The creation means creates the threshold value matrix obtained by dividing the Voronoi polygon according to the dot size of an image forming apparatus, and sets a threshold value to be compared with image data in each cell of the threshold value matrix. Item 3. The image processing device according to Item 1 or Item 2.   The creating means creates the threshold matrix for the Voronoi polygon generated by the creating means, and stores the threshold matrix in a memory in association with the position of the halftone dot before the displacement. 4. The image processing device according to claim 1, wherein the image processing device is an image processing device.   5. The image processing device according to claim 1, wherein the halftone dot before the displacement has an interval corresponding to the number of lines of the image to be formed. And further comprising classification means for dividing the Voronoi polygon into a plurality of groups,
6. The image processing according to claim 1, wherein the creating unit sets the threshold values of threshold matrixes corresponding to Voronoi polygons having different groups so as not to have the same value. apparatus. Further, the Voronoi polygon is classified into a group for black dots and a group for white dots,
The creating means creates, for the Voronoi polygon of the group for white dots, a threshold value matrix that reduces white dots in a shape substantially similar to the Voronoi polygon, with the center of gravity of the Voronoi polygon as a reduction center. 6. The image processing apparatus according to claim 1, wherein the image processing apparatus is characterized in that:   8. The image processing apparatus according to claim 7, wherein the classification unit divides a Voronoi polygon having a small area into the white dot group. A memory for storing a threshold value matrix created by the image processing device according to any one of claims 1 to 8,
Comparing means for inputting image data in accordance with a synchronization signal of the image forming apparatus, inputting a threshold value matrix corresponding to the image data from the memory, and comparing the threshold value of each cell of the image data and the threshold value matrix;
An image processing apparatus comprising: a driving unit that drives a beam light source of the image forming apparatus according to the comparison result. Enter the control point of the halftone screen,
Dividing the halftone screen into polygons surrounding the control points,
The halftone screen halftone dot included in the polygon is aperiodically and randomly displaced,
Generating a Voronoi polygon including the halftone dot after the displacement;
An image processing method characterized by creating a threshold matrix for growing halftone dots in a shape substantially similar to the Voronoi polygon with the center of gravity of the Voronoi polygon as a growth center.   10. A computer program for controlling a computer device to function as each unit of the image processing device according to claim 1.   12. A computer-readable storage medium in which the computer program according to claim 11 is recorded.

Images