JP5458946B2 - Image processing apparatus, program, and recording medium - Google Patents

Description

  The present invention relates to an image processing apparatus such as an electrophotographic or ink-jet printer or a multifunction peripheral that expresses gradation using dots, and particularly to an edge portion of image data after halftone processing in such an image processing apparatus. The present invention relates to a technique for improving generated jaggy.

  In an electrophotographic printer or an inkjet printer, dither processing is often performed as halftone processing for converting multi-value image data into a predetermined number of output gradations. In the dither processing, for example, a screen pattern having a high line number of 300 lines or more with an emphasis on resolution is used for text, and a screen pattern of about 200 lines with an emphasis on gradation is used for graphics and images.

  Now, there is a problem that jaggy occurs due to the screen pattern of dither processing at the edge portion of halftone characters and line drawings (for example, paragraphs [0002] to [0004] of Patent Document 1; Patent Document 1). 2 paragraphs [0017]-[0019]). The size of the jaggy depends on the screen angle and the edge angle. When this angle is close, the jaggy appears remarkably.

  For example, Patent Documents 1 and 2 are known as prior art documents relating to such jaggy. In the technique disclosed in Patent Document 1, when there is no dot around the contour pixel, a dot is added to correct jaggies. Further, by determining the size (output level) of the output dot in accordance with the pixel value level of the contour pixel before halftone processing, density fluctuation due to the difference in output characteristics between isolated dots and continuous dots is reduced. Further, in the technique disclosed in Patent Literature 2, edge correction data is generated by performing smoothing processing on image data before halftone processing at an edge portion, and the edge correction data and image data after halftone processing are combined. By comparing and outputting the one having the higher density, dots are added between the dots in the edge portion to correct jaggies.

  However, in the technique of Patent Document 1, since dots of a plurality of sizes are mixed in the edge portion at medium density or higher (see FIG. 10 (c) of Patent Document 1), dot reproducibility (dot density characteristics) changes. In such a case, local density fluctuations may occur at the edge portion. Further, since dots are added between the dots and correction is made so that the dots having a low output level are darkened, there arises a problem that the lines are thickened with high-density characters and line drawings. Also in the technique of Patent Document 2, since the correction is performed so that dots are added between the dots, the same problem that the lines are thickened in high-density characters and line drawings occurs.

  An object of the present invention is to effectively prevent jaggies that occur at the edge of an image while suppressing the thickness of characters and line drawings in an image processing apparatus such as an electrophotographic or ink jet printer or a multi-function printer that expresses gradation using dots. There is to improve.

The invention described in claim 1
Halftone processing means for performing halftone processing on the first image data;
Edge detecting means for detecting an edge portion of the image from the first image data;
Among the pixels (denoted as edge pixels) of the image data (denoted as second image data) after the halftone processing in the edge portion detected by the edge detection means, the edge pixels with respect to the edge pixel of dot OFF Is corrected so as to output a dot having an output level corresponding to the density before the halftone process, and the dot output level is lowered for the dot ON edge pixel having a density higher than a predetermined density before the halftone process. Edge correction processing means for correcting
An image processing apparatus comprising:

According to a second aspect of the present invention, in the image processing apparatus according to the first aspect of the present invention,
The edge correction processing means performs correction so as to lower the dot output level for all dot ON edge pixels having a density of a predetermined level or higher before halftone processing.

According to a third aspect of the present invention, in the image processing apparatus of the first aspect of the present invention,
The edge correction processing means performs correction so as to lower the dot output level only for a part of the dot ON edge pixels having a predetermined density or more before halftone processing.

According to a fourth aspect of the present invention, in the image processing apparatus of the first aspect of the present invention,
The edge correction processing means includes
Means for determining a coefficient based on a pixel value of each pixel of the first image data;
Means for determining a first correction candidate value based on the pixel value for each pixel of the first image data;
Means for multiplying the pixel value of each pixel of the second image data by the coefficient determined corresponding to the pixel, and determining a second correction candidate value by quantizing the multiplication result;
For each pixel of the second image data, a darker value of the first correction candidate value and the second correction candidate value determined corresponding to the pixel is a correction value corresponding to the pixel. And means for replacing the pixel value of the edge pixel of the second image data with the correction value corresponding to the edge pixel,
The means for determining the coefficient decreases the value of the coefficient as the pixel value increases, and the means for determining the first correction candidate value increases the first correction candidate value as the pixel value increases. Is.

The invention according to claim 5 is the image processing apparatus according to claim 1,
The edge correction processing means includes
Means for determining a coefficient based on a pixel value of each pixel of the first image data;
Means for determining a first correction candidate value based on the pixel value for each pixel of the first image data;
For each pixel of the first image data, the pixel value is multiplied by the coefficient determined corresponding to the pixel, and the multiplication result is subjected to the same halftone processing as the halftone processing means. Means for determining two correction candidate values;
For each pixel of the second image data, a darker value of the first correction candidate value and the second correction candidate value determined corresponding to the pixel is a correction value corresponding to the pixel. And means for replacing the pixel value of the edge pixel of the second image data with the correction value corresponding to the edge pixel,
The means for determining the coefficient decreases the value of the coefficient as the pixel value increases, and the means for determining the first correction candidate value increases the first correction candidate value as the pixel value increases. Is.

The invention according to claim 6 is the image processing apparatus according to claim 4 or 5, wherein
Output to the edge pixel to be corrected by lowering the output level of the dot by controlling the value of the coefficient determined for the edge pixel in accordance with the distance from the edge boundary of the edge pixel by means for determining the coefficient The degree of lowering the level is weakened as the distance from the edge boundary of the edge pixel increases.

The invention according to claim 7 is the image processing apparatus according to claim 4 or 5, wherein
The means for determining the coefficient includes a pixel belonging to a line having a predetermined line width or less, a pixel belonging to a character having a predetermined size, a pixel adjacent to a white line having a predetermined line width or less, or a white having a predetermined size or less. According to whether or not it is a pixel corresponding to one of the pixels adjacent to the character (denoted as a special pixel), the coefficient value determined for the pixel is controlled,
The means for determining the first correction candidate value controls the value of the first correction candidate value determined for the pixel according to whether or not the pixel corresponds to the special pixel.
When the edge pixel corresponds to the special pixel, the degree to which the output level for the edge pixel to be corrected to lower the dot output level is increased compared to the case where the edge pixel does not correspond to the special pixel. The output level of the dots to be output is more likely to be lower when the edge pixel corresponds to the special pixel than when the edge pixel does not correspond to the special pixel.

  The invention described in claim 8 is a program that causes a computer to function as each unit of the image processing apparatus according to any one of claims 4 to 7.

  A ninth aspect of the present invention is a computer-readable recording medium in which a program for causing a computer to function as each means of the image processing apparatus according to any one of the fourth to seventh aspects is recorded.

  In the present invention, among the image data edge pixels after halftone processing, correction is performed so that a dot having an output level corresponding to the density of the edge pixel before halftone processing is output to the edge pixel of dot OFF. As a result, it is possible to improve the jaggy of the edge portion, and at the same time, the dot ON edge pixel having a density higher than a predetermined density is corrected so as to lower the dot output level before halftone processing. It is possible to effectively suppress the thickening of characters and line drawings due to the addition of dots for. Further, as the distance from the edge boundary of the edge pixel of the dot ON is larger, the excessive effect due to the correction of the dot output level of the edge pixel of the dot ON is suppressed by weakening the degree of lowering the dot output level of the edge pixel. be able to. In addition, the degree to which the output level for the edge pixel to be corrected for lowering the dot output level is reduced is determined by the edge pixel being a special pixel (a pixel belonging to a line having a predetermined line width or less, a pixel belonging to a character having a predetermined size). In the case of a pixel adjacent to a white line having a predetermined line width or less, or a pixel adjacent to a white character having a predetermined size or less), the pixel is strengthened and the dot OFF. By effectively making the output level of the dots to be output to the edge pixel of the edge pixel lower than when the edge pixel corresponds to the special pixel and not to the special pixel, it is possible to effectively reduce the fatness of small characters and fine lines. It is possible to achieve such effects as suppressing jaggies while effectively preventing blurring of small white characters and fine white lines.

1 is a block diagram of an image processing apparatus according to an embodiment of the present invention. It is a block diagram for demonstrating the internal structure of an edge correction process part. It is a block diagram for demonstrating the internal structure of an edge correction process part. It is a figure which shows the example of the dither threshold value matrix for a halftone process. It is explanatory drawing of the mask for edge detection. It is explanatory drawing of the mask for edge detection. It is a graph which shows the relationship between the correction coefficient (alpha) and the pixel value (density) of input image data. It is a graph which shows the relationship between the correction candidate value β and the pixel value (density) of input image data. It is a graph which shows the relationship between the correction coefficient (alpha) and the pixel value (density) of input image data. It is a graph which shows the relationship between the correction candidate value β and the pixel value (density) of input image data. It is a schematic diagram which shows the example of the image data before edge correction. It is a schematic diagram which shows the example of the image data after an edge correction process. It is a schematic diagram which shows the example of the image data after an edge correction process. It is a schematic diagram which shows the example of the image data after an edge correction process.

  Embodiments of the present invention will be described below with reference to the drawings.

  FIG. 1 is a block diagram for explaining the overall configuration of an image processing apparatus according to an embodiment of the present invention. Specifically, this image processing apparatus is, for example, a printer (image forming apparatus) of an electrophotographic system or an ink jet system in which gradation is expressed using a plurality of types of dots having different sizes. This image processing apparatus may be realized as an independent printer, or may be realized as an apparatus embedded in a device such as a multifunction peripheral (MFP). Note that FIG. 1 does not show a print engine mechanism that actually forms an image on a recording medium using a plurality of types of dots having different sizes in accordance with the image data OUTDATA after the edge correction processing.

  In FIG. 1, INDATA is image data to be printed, and is described here as raster data of 8 bits / pixel (256 gradations). The image data storage unit 102 is a means for temporarily storing the image data INDATA. TAG is attribute information for each pixel of the image data INDATA, and is described here as three types of object information of text / graphic / image, but other attribute information can be included (Example 4 described later). reference). The attribute information storage unit 103 is means for temporarily storing the attribute information TAG in association with each pixel of the image data INDATA. The PDL data processing unit 101 analyzes PDL data input from the outside, generates image data INDATA and attribute information TAG to be printed, and writes the data into the image data storage unit 102 and the attribute information storage unit 103 It is. Note that text / graphic / image object information and the like can be obtained directly by analysis of PDL data, but attribute information or image feature information (see Example 4 described later) that cannot be obtained by the analysis is PDL. In the data processing 101, extraction processing is performed based on the image data INDATA and the attribute information TAG.

  The halftone processing unit 104 performs dither processing as halftone processing on the image data INDATA, and outputs image data sDATA. The halftone processing unit 104 selects and uses a screen pattern optimal for the type of object to which the processing target pixel belongs, according to the attribute information TAG.

  The edge detection unit 105 detects an edge portion of the image (the pixel on the dark side of the portion having a large density difference) from the image data INDATA, and outputs detection information as an edge detection signal DE.

  The edge correction processing unit 106 performs processing for correcting the edge portion of the image data sDATA after the halftone processing so that jaggies are not noticeable, and outputs the image data OUTDATA after the edge correction.

  Image data sDATA and OUTDATA are signals representing the type (output level) of dots output by the printer engine mechanism. Here, it is assumed that the number of output gradations of the printer is 4, and each pixel of the image data sDATA and OUTDATA can take four output levels: dot OFF, small dot, medium dot, and large dot. Here, the large dot is the dot with the highest output level. Such quaternary data can be expressed with a minimum of 2 bits. For convenience, it is assumed here that each image data is 8 bits (0 to 255) / pixel data, and 0 (decimal expression) is set to dot OFF. , 85 is assigned to the small dot, 170 is assigned to the medium dot, and 255 is assigned to the large dot.

Then, the halftone processing unit 104 uses the dither threshold matrix shown in FIG.
If INDATA <th1, sDATA = 0 (dot off)
If th1 ≦ INDATA <th2, sDATA = 85 (small dot)
If th2 ≦ INDATA <th3, sDATA = 170 (medium dot)
If th3 ≦ INDATA, sDATA = 255 (large dot)
It is assumed that this conversion is performed.

  FIG. 11 is a diagram schematically showing an example of image data sDATA after halftone processing using the dither threshold matrix. A portion indicated by an arrow is an edge boundary, and represents a result of processing image data INDATA of a white background on the left side of the arrow and a gray image on the right side. The density on the right side of the arrow in the image data INDATA is 65/255 in (a), 105/255 in (b), and 200/255 in (c). Here, for example, “65/255” means 65 (decimal) in the 8-bit representation (0 to 255). Others are the same.

  In (a) to (c), since there is a dot-off pixel (white pixel) in the edge portion, this is recognized as a jaggy in the edge portion. In (a), only large dots are used. In (b), large dots and small dots are mixed. In (c), large dots and medium dots are mixed.

  In order to make the jaggies inconspicuous while suppressing the thickening of characters and line drawings, the edge correction processing unit 106 applies the pixel OFF of the pixel of the edge portion to the image data sDATA after the halftone processing. Correction is made so that dots of an output level corresponding to the density before halftone processing are output, and the dot output level is reduced if the density before halftone processing is equal to or greater than a predetermined value for the dot-on pixels in the edge portion. Such edge correction processing is performed, details of which will be described later.

Hereinafter, mainly the edge detection unit 105 and the edge correction processing unit 106 will be described more specifically with some examples.
[Example 1]
In the first embodiment, the edge detection unit 105 detects an edge portion of an image with a 2-pixel width from the image data INDATA using the mask shown in FIG. That is, with reference to 5 × 5 pixels (B0 to B7, C0 to C15) around the target pixel A shown in FIG. 5, Δ is calculated by the following calculation formula.
Δ = A-MIN (B, C)
However,
B = MIN (B0, B1, ..., B7)
C = MIN (C0, C1, ..., C15)
Here, MIN () means the minimum value in ().
Then, when Δ> Th, the target pixel A is detected as an edge pixel. Thereby, the edge portion can be detected with a width of 2 pixels.

  Next, the edge correction processing unit 106 will be described. FIG. 2 is a block diagram illustrating an internal configuration of the edge correction processing unit 106 according to the first embodiment.

  As shown in FIG. 2, the edge correction processing unit 106 includes an edge correction data generation unit 201 and a selector 202. The edge correction data generation unit 201 is a part that generates edge correction data eDATA to be replaced with the image data sDATA after halftone processing for the edge pixels detected by the edge detection unit 105. The edge correction data eDATA is 8-bit data having the same four values (0, 85, 170, 255) as the number of output gradations as in the case of the image data sDATA, OUTDATA, but can also be expressed by 2 bits. is there. The selector 202 selects the image data sDATA output from the halftone processing unit 104 and outputs it as image data OUTDATA for the pixel whose attribute information TAG indicates an image object, regardless of the edge detection signal DE. On the other hand, when the attribute information TAG indicates a text object or graphic object, the selector 202 selects the edge correction data eDATA for the edge pixel and outputs it as image data OUTDATA, and selects the image data sDATA for the non-edge pixel. And output as image data OUTDATA. That is, except for the image object, the image data sDATA for the edge portion is replaced with the edge correction data eDATA to obtain the image data OUTDATA in which the jaggy of the edge portion is corrected.

  The edge correction data generation unit 201 includes an α determination unit 211, a β determination unit 212, an α multiplication / quantization unit 213, and a maximum value selection unit 214.

  The α determination unit 211 and the β determination unit 212 are means for determining a correction coefficient α and a correction candidate value β according to the pixel value of the image data INDATA, that is, the density. Specifically, as shown in FIG. 7, the value of the correction coefficient α obtained by the α determining unit 211 is 1.0 at a low density less than a predetermined density (INDATA <100 in this example), and the density becomes high ( Decreases as INDATA increases, and decreases to 0.6 at a high concentration above a predetermined concentration (INDATA ≧ 200 in this example). As shown in FIG. 8, the correction candidate value β is a value that increases as the density of the image data INDATA increases, and takes the same four values (0, 85, 170, and 255) as the image data sDATA and OUTDATA.

The α multiplication / quantization unit 213 multiplies the halftone processed image data sDATA by the correction coefficient α determined by the α determination unit 211, and simply quantizes the multiplication result to four values to obtain a correction candidate value. This is output as cDATA. That is,
If sDATA × α <42, cDATA = 0
If 42 ≦ sDATA × α <128, cDATA = 85
If 128 ≦ sDATA × α <212, cDATA = 170
If 212 ≦ sDATA × α, cDATA = 255
Thus, the correction candidate value cDATA is determined.

  The maximum value selection unit 214 selects a correction candidate value cDATA output from the α multiplication / quantization unit 213 and the larger (darker) value of the other correction candidate value β as a correction value. Is output as edge correction data eDATA.

  The effect of edge correction according to the first embodiment will be described. FIG. 12 schematically shows edge-corrected image data OUTDATA when the same image data INDATA as in FIG. 11 is processed. Arrows indicate edge boundaries. The right pixel adjacent to the edge boundary and the right adjacent pixel are detected as pixels (edge pixels) of the edge portion.

  In the case of FIG. 12A, α = 1.0 and β = 85 from FIGS. cDATA = sDATA, and the larger value of cDATA and β is output as the correction value eDATA. Therefore, the value of the dot OFF pixels (two columns in the vertical direction) at the edge portion of each line in the data sDATA is replaced with the correction value 85/255. That is, the edge correction is performed so that the small dot (85/255) is output to the edge-off dot-off pixel, and as a result, jaggy is improved.

  In the case of FIG. 12B, α = 0.98 and β = 85 from FIGS. In this case as well, as in the case of (a), the dot OFF pixel at the edge of the image data sDATA is corrected to output a small dot (85/255), and jaggy is improved.

  In the case of FIG. 12C, α = 0.6 and β = 170 from FIGS. In the pixel where the value of the data sDATA in the edge portion is 255, when a large dot is output, the correction candidate value cDATA = β = 170, so one (whichever is sufficient) is selected as the correction value eDATA. It is corrected to output medium dots. Further, in a pixel whose value of the data sDATA in the edge portion is less than 255, β (= 170) is larger than cDATA, and correction is made so that eDATA = 170, that is, a medium dot is output. Therefore, the output level of the edge portion becomes uniform with medium dots, and jaggy is improved. In this way, pixels with dot OFF in the edge portion are corrected to output medium dots, while pixels that output large dots in the edge portion are corrected to lower the output level to medium dots, so edge correction The character and line drawing due to will not occur.

  The characteristic shape of α shown in FIG. 7 is determined according to the output characteristics of the printer (dot size, toner scattering, etc.). On the low-density side, even if correction is made so that dots are output to the pixels with dot OFF, no thickening of characters or line drawings is seen. On the contrary, if the correction is performed so that the output level of the image data after halftone processing is lowered on the low density side, the edges may appear blurred. Therefore, the point at which the value of α starts to be lowered (INDATA = 100 in FIG. 6) should not be set to a lower density side than necessary, but should be set around the density at which characters and line drawings start to appear thick.

[Example 2]
In the second embodiment, the edge detection unit 105 uses the mask shown in FIG. 5 to detect the edge portion of the image with a two-pixel width from the image data INDATA. At this time, the edge pixel (outside edge) adjacent to the edge boundary is detected. Edge pixels) and edge pixels (inner edge pixels) adjacent to the edge pixels are identified. That is, the edge detection unit 105 refers to 5 × 5 pixels (B0 to B7, C0 to C15) around the pixel of interest A, and calculates Δ1 and Δ2 with the following calculation formulas.
Δ1 = A−B
Δ2 = A−C
However,
B = MIN (B0, B1, ..., B7)
C = MIN (C0, C1,..., C15) * MIN is the meaning of the minimum value, and if Δ1> Th1, the target pixel A is detected as an outer edge pixel, Δ1 ≦ = Th1, and Δ2> If it is Th, the target pixel A is detected as an inner edge pixel. In this manner, the edge detection unit 105 identifies and detects the edge portion as a two-pixel width outer edge pixel having a small distance from the edge boundary and an inner edge pixel having a large distance from the edge boundary. The detection information is output as an edge detection signal DE.

  The edge correction processing unit 106 according to the second embodiment has an internal configuration as shown in FIG. However, although not shown in FIG. 2, the edge detection signal DE is input to the α determination unit 211. Then, the α determining unit 211 switches the value of the correction coefficient α depending on whether the pixel is an inner edge pixel or an outer edge pixel based on the edge detection signal DE. Specifically, the correction coefficient α is set to a value as shown in FIG. 7 in the outer edge pixel as in the first embodiment, but α = 1.0 is always set in the inner edge pixel. As in the first embodiment, the β determination unit 212 determines the correction candidate value β according to FIG.

  The effect of edge correction according to the second embodiment will be described. FIG. 13 schematically shows edge-corrected image data OUTDATA when the same image data INDATA as in FIG. 11 is processed. Arrows indicate edge boundaries. The right pixel adjacent to the edge boundary is detected as an outer edge pixel, and the right adjacent pixel is detected as an inner edge pixel.

  As apparent from a comparison between FIG. 13 and FIG. 12, the image data OUTDATA in FIGS. 13A and 13B is the same as that in the first embodiment, and the jaggy of the edge portion is improved.

  In the case of FIG. 13C, α = 1.0 at the inner edge pixel (the edge pixel far from the edge boundary), so that the large dot is not corrected to the middle dot, but the outer edge pixel (edge boundary) As in the case of the first embodiment, the output level is corrected from a large dot to a medium dot.

  The degree of weight of characters and line drawings varies depending on the output characteristics of the printer. In the case of a printer having output characteristics in which the effect is excessive and the characters and line drawings become too thin in the correction as in the first embodiment, the edge boundary of the edge pixel of the dot ON as in the second embodiment. If the distance from the edge is large, the edge correction considering the distance from the edge boundary that weakens the degree to which the output level of the dot for the edge pixel is reduced will make the character and line drawing thicker more appropriately. It becomes possible to improve jaggy while suppressing.

[Example 3]
In the third embodiment, the edge detection unit 105 detects an edge portion (a darker pixel of a portion having a large density difference) with a three-pixel width. That is, by using the mask shown in FIG. 6, 7 × 7 pixels (B0 to B7, C0 to C15, D0 to D23) around the target pixel A of the image data INDATA are referred to, and Δ is calculated by the following calculation formula.
Δ = A−MIN (B, C, D)
However,
B = MIN (B0, B1, ..., B7)
C = MIN (C0, C1, ..., C15)
D = MIN (D0, D1,..., D23) * MIN is the meaning of the minimum value, and when Δ> Th, the target pixel A is detected as an edge pixel. Thereby, the edge portion can be detected with a width of 3 pixels.

  Although the edge portion is detected with a width of 3 pixels here, the detection width of the edge portion depends on the output resolution of the printer. When the output resolution is high, it is necessary to increase the detection width of the edge portion to some extent. However, if the detection width is increased, the processing time and circuit scale required to detect the edge portion increase, and therefore an appropriate detection width is set in consideration of the trade-off.

  FIG. 3 is a block diagram for explaining the internal configuration of the edge correction processing unit 106 according to the third embodiment. 3, the same elements as those in FIG. 2 or corresponding elements are denoted by the same reference numerals.

  The edge correction processing unit 106 according to the third embodiment includes an edge correction data generation unit 201 and a selector 202, but the internal configuration of the edge correction data generation unit 201 is partially different from that shown in FIG. That is, the α multiplication / quantization unit 213 in FIG. 2 is replaced with an α multiplication unit 215 and a halftone processing unit 216. The α multiplier 215 is means for multiplying the image data INDATA by the correction coefficient α determined by the α determiner 211. The halftone processing unit 216 performs dither processing as the halftone processing same as the halftone processing unit 104 in FIG. 1 on the image data after the α multiplication unit 215 multiplies the correction coefficient α to give four values (0, 0, 85, 170, 255) is output as a quantized candidate value cDATA, and an optimal screen pattern is used for each object in accordance with the attribute information TAG. As in the case of the first embodiment, each of the α determination unit 211 and the β determination unit 212 determines the correction coefficient α and the correction candidate value β as shown in FIGS. 7 and 8 according to the density of the image data INDATA. . The larger (darker) value of the correction candidate value cDATA and the correction candidate value β is selected as a correction value by the maximum value selection unit 214 and is output as edge correction data eDATA.

  The effect of edge correction according to the third embodiment will be described. FIG. 14 schematically shows edge-corrected image data OUTDATA when the same image data INDATA as in FIG. 11 is processed. Arrows indicate edge boundaries. Three pixels on the right side of the edge boundary are detected as edge portions. Here, it is assumed that the dither threshold matrix shown in FIG. 4 is used by the halftone processing unit 216 and the halftone processing unit 104.

  14 (a) and 14 (b) are the same as those in the first embodiment (FIG. 12) except that the edge portion has a width of 3 pixels. By correcting so that a small dot (85/255) is output to a pixel with dot OFF, jaggy is improved.

  In the case of FIG. 14C, α = 0.6 and β = 170 from FIGS. The value obtained by multiplying the 8-bit image data INDATA by the correction coefficient α is 200 × 0.6 = 120. The correction candidate value cDATA obtained by dithering the halftone processing unit 216 and the correction candidate value β (= The larger value of 170) is output. As a result, in the edge portion (three vertical columns) with a width of three pixels, the dot OFF pixels of the image data sDATA are corrected so that medium dots are output, and some pixels that output large dots are medium. By correcting so that dots are output, jaggy is improved.

  In the third embodiment, unlike the first embodiment, the correction candidate value cDATA is dithered, so that the screen pattern of the data sDATA is maintained to some extent even in the edge portion after the edge correction. Compared to Examples 1 and 2, the uncomfortable feeling associated with edge correction is less likely to occur.

  In the third embodiment, as in the second embodiment, the distance from the edge boundary of the edge pixel is identified by edge detection, and the correction coefficient α for the edge pixel is switched according to the distance from the edge boundary. It is also possible. Specifically, for example, the value of the correction coefficient α is determined according to FIG. 7 for the edge pixel closest to the edge boundary among the three edge pixels, and the correction coefficient α is determined for the edge pixel farthest from the edge boundary. A value of 1.0 is used, and an intermediate value (for example, an average value) of correction coefficient α values for edge pixels on both sides is used as the value of correction coefficient α for an edge pixel between the two edge pixels. Can be.

[Example 4]
In the fourth embodiment, the edge detection method may be the same as in the first, second, or third embodiment, and the internal configuration of the edge correction processing unit 106 is the same as in the first, second, or third embodiment. Good. However, the attribute information TAG is referred to by the α determination unit 211 and the β determination unit 212 shown in FIG.

  Now, for thick lines and large characters, even if the lines are thick due to jaggy correction, the thickness is difficult to recognize, whereas thin lines and small characters are likely to be recognized. Also, for small white characters and white thin lines on the color background, when jaggies are corrected at the edge of the background, the edges of the background become darker (thickened), and the white characters and white thin lines become thinner. , Easy to fade.

In view of the above, in the fourth embodiment, the PDL data processing unit 101
(1) Pixels belonging to lines with a predetermined line width or less
(2) Pixels belonging to characters of a predetermined size or less
(3) Pixel adjacent to a white line with a predetermined line width or less
(4) Acquire attribute information indicating whether or not the pixel is adjacent to a white character having a predetermined size or less, and add it to the attribute information TAG.

  Whether or not the pixel is (1) or (2) can generally be determined directly by analysis of PDL data. Further, for each pixel of the image data developed on the image data storage unit 102, by referring to the attribute information on the attribute information storage unit 103 and checking whether the pixel is adjacent to the white character or white line pixel, It can be determined whether the pixel is (3) or (4).

  The added attribute information is referred to by the α determination unit 211 and the β determination unit 212. Then, by controlling the correction coefficient α and the correction candidate value β according to whether the pixel corresponds to any one of the above pixels (1) to (4), a small character associated with edge correction In addition, it effectively suppresses the thickening of thin lines and thin white characters and faint lines of white lines.

  This will be described more specifically. The α determination unit 211 determines the value of the correction coefficient α according to FIG. 9 for a pixel corresponding to any one of the pixels (1) to (4) described above (denoted as a special pixel), and does not correspond to a special pixel. For the pixel, the value of the correction coefficient α is determined according to FIG. 7 as in the first, second, and third embodiments. However, when the edge detection is performed by identifying the inner edge pixel and the outer edge pixel as in the second embodiment, the correction coefficient α is always set to 1.0 for the inner edge pixel. The β determination unit 212 determines the correction candidate value β according to FIG. 10 for the pixels corresponding to the special pixels, and for the pixels not corresponding to the special pixels, as in the first, second, and third embodiments, FIG. Then, the correction candidate value β is determined.

  In FIG. 9 applied to a pixel corresponding to a special pixel, the value of the correction coefficient α on the high density side is made smaller than that in FIG. 7 applied to other pixels. Therefore, when a dot ON edge pixel corresponds to a special pixel and FIG. 9 is applied to determine the correction coefficient α, the large dot (255/255) of the edge pixel is a small dot (85/255) or a medium dot. (170/255) is easily corrected (the degree to which the dot output level is lowered increases).

  Further, by determining the correction candidate value β for the pixel corresponding to the special pixel as shown in FIG. 10, when the edge pixel of dot OFF corresponds to the special pixel, the dot output to the edge pixel is a small dot ( 85/255) or medium dots (170/255) (prone to low output levels).

  Therefore, it is possible to improve jaggies while effectively suppressing the thickening of small characters and fine lines and effectively preventing blurring of small white characters and fine white lines.

  As mentioned above, although embodiment of this invention was described, this invention is not limited only to such embodiment, It can take various deformation | transformation forms.

  In addition, the means such as the halftone processing unit 104, the edge detection unit 105, and the edge correction processing unit 106 in the image processing apparatus according to the above-described embodiment can be realized by dedicated hardware, and can be generally executed by a program. It is also possible to cause a computer with a simple configuration to function as these means. Such a program is also included in the present invention. Various recording (storage) media that can be read by a computer, such as a magnetic disk, an optical disk, a magneto-optical disk, and a semiconductor storage element, on which such a program is recorded are also included in the present invention.

101 PDL data processing unit 102 Image data storage unit 103 Attribute information storage unit 104 Halftone processing unit 105 Edge detection unit 106 Edge correction processing unit 201 Edge correction data generation unit 202 Selector 211 α determination unit 212 β determination unit 213 α multiplication / quantum Conversion unit 214 maximum value selection unit 215 α multiplication unit 216 halftone processing unit

Japanese Patent No. 4111190 Japanese Patent No. 4217706

Claims (9)

Halftone processing means for performing halftone processing on the first image data;
Edge detecting means for detecting an edge portion of the image from the first image data;
Among the pixels (denoted as edge pixels) of the image data (denoted as second image data) after the halftone processing in the edge portion detected by the edge detection means, the edge pixels with respect to the edge pixel of dot OFF Is corrected so as to output a dot having an output level corresponding to the density before the halftone process, and the dot output level is lowered for the dot ON edge pixel having a density higher than a predetermined density before the halftone process. Edge correction processing means for correcting
An image processing apparatus comprising: The image processing apparatus according to claim 1.
The image processing apparatus according to claim 1, wherein the edge correction processing unit performs correction so as to lower a dot output level for all of the dot ON edge pixels having a density of a predetermined density or more before halftone processing. The image processing apparatus according to claim 1.
The image processing apparatus according to claim 1, wherein the edge correction processing unit performs correction so as to lower a dot output level for only a part of dot ON edge pixels having a predetermined density or more before halftone processing. The image processing apparatus according to claim 1.
The edge correction processing means includes
Means for determining a coefficient based on a pixel value of each pixel of the first image data;
Means for determining a first correction candidate value based on the pixel value for each pixel of the first image data;
Means for multiplying the pixel value of each pixel of the second image data by the coefficient determined corresponding to the pixel, and determining a second correction candidate value by quantizing the multiplication result;
For each pixel of the second image data, a darker value of the first correction candidate value and the second correction candidate value determined corresponding to the pixel is a correction value corresponding to the pixel. And means for replacing the pixel value of the edge pixel of the second image data with the correction value corresponding to the edge pixel,
The means for determining the coefficient decreases the value of the coefficient as the pixel value increases, and the means for determining the first correction candidate value increases the first correction candidate value as the pixel value increases. Image processing device. The image processing apparatus according to claim 1.
The edge correction processing means includes
Means for determining a coefficient based on a pixel value of each pixel of the first image data;
Means for determining a first correction candidate value based on the pixel value for each pixel of the first image data;
For each pixel of the first image data, the pixel value is multiplied by the coefficient determined corresponding to the pixel, and the multiplication result is subjected to the same halftone processing as the halftone processing means. Means for determining two correction candidate values;
For each pixel of the second image data, a darker value of the first correction candidate value and the second correction candidate value determined corresponding to the pixel is a correction value corresponding to the pixel. And means for replacing the pixel value of the edge pixel of the second image data with the correction value corresponding to the edge pixel,
The means for determining the coefficient decreases the value of the coefficient as the pixel value increases, and the means for determining the first correction candidate value increases the first correction candidate value as the pixel value increases. Image processing device. The image processing apparatus according to claim 4 or 5,
Output to the edge pixel to be corrected by lowering the output level of the dot by controlling the value of the coefficient determined for the edge pixel in accordance with the distance from the edge boundary of the edge pixel by means for determining the coefficient An image processing apparatus, characterized in that the degree of lowering the level decreases as the distance from the edge boundary of the edge pixel increases. The image processing apparatus according to claim 4 or 5,
The means for determining the coefficient includes a pixel belonging to a line having a predetermined line width or less, a pixel belonging to a character having a predetermined size, a pixel adjacent to a white line having a predetermined line width or less, or a white having a predetermined size or less. According to whether or not it is a pixel corresponding to one of the pixels adjacent to the character (denoted as a special pixel), the coefficient value determined for the pixel is controlled,
The means for determining the first correction candidate value controls the value of the first correction candidate value determined for the pixel according to whether or not the pixel corresponds to the special pixel.
When the edge pixel corresponds to the special pixel, the degree to which the output level for the edge pixel to be corrected to lower the dot output level is increased compared to the case where the edge pixel does not correspond to the special pixel. An image processing apparatus characterized in that an output level of a dot to be output is likely to be a lower output level when the edge pixel corresponds to the special pixel than when the edge pixel does not correspond to the special pixel.   A program that causes a computer to function as each unit of the image processing apparatus according to any one of claims 4 to 7.   A computer-readable recording medium on which a program for causing a computer to function as each unit of the image processing apparatus according to claim 4 is recorded.

Images