JP2012239012A - Image processing device, image processing method, and computer program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To perform high-quality edge correction processing for an image data edge.SOLUTION: An image processing device comprises: density determination means for determining a density of target pixel data; error diffusion processing means for performing error diffusion processing for the target pixel data; dither processing means for performing dither processing for the target pixel data; edging processing means for performing edging processing on the basis of the target pixel data and data generated by the dither processing means; and selection means for selecting the data generated by the edging processing means as data for the edge correction processing with respect to the target pixel when the density determination means determines that the target pixel is a high density having a higher value than a threshold value and for selecting the data generated by the error diffusion processing means as the data for the edge correction processing with respect to the target pixel when the density determination means determines that the target pixel is a low density having a lower value than the threshold value.

Description

  The present invention relates to an image processing apparatus, an image processing method, and a computer program capable of executing correction processing on an edge of an image.

  In general, a digital image processing apparatus such as a digital copying apparatus that reproduces an image by outputting digitized image data from a digital printer such as a laser beam printer is widely used instead of a conventional analog image processing apparatus due to the development of digital equipment. ing. This digital image processing apparatus generally employs a method of reproducing gradation by halftone processing such as a dither processing method in order to reproduce a halftone.

  The gradation reproduction by the dither processing method has a problem that a break called jaggy occurs in a portion with a lot of high-frequency components such as a halftone character and an edge portion of a fine line portion. This is because the frequency component included in the input image is higher than the period of the dither, so that aliasing noise occurs in the dither period, the apparent resolution is lowered, and the contour of the edge portion is obscured. .

  In order to solve this problem, a technique is widely used in which an edge of an input image is determined, edge processing is performed on the edge portion, and a contour is emphasized (see, for example, Patent Document 1).

  Moreover, in order to reproduce the high frequency component of an edge part, what uses edge part error diffusion (for example, refer patent document 2) is also proposed.

JP 2005-341249 A JP 63-288565 A

  However, the edge correction processing according to Patent Document 1 is a very effective method for a high density edge portion where the edging process can be performed stably, but the dot reproducibility is not good for the low density edge portion. The edge may become stable and the edge may not be corrected appropriately.

  Further, the edge correction processing according to Patent Document 2 improves edge reproduction in a low density portion by increasing the resolution by error diffusion processing and setting an optimum quantization value. However, in the high density edge portion, since dots are randomly generated by error diffusion processing, there is a problem that jaggy is not reduced when dots are not added to the jaggy portion generated by the screen.

  An image processing apparatus according to the present invention is an image processing apparatus that performs edge correction processing on edges in an image, and includes density determination means for determining a density of data of a target pixel, and error diffusion for the data of the target pixel. An error diffusion processing unit that performs processing, a dither processing unit that performs dither processing on the data of the target pixel, and an edge that performs edge fringing processing based on the data of the target pixel and data generated by the dither processing unit If the edge determination processing unit and the target pixel are determined to have a high density, which is a density higher than a threshold value, by the density determination unit, the edge edge processing unit is used as data of the edge correction processing for the target pixel. Is selected, and the pixel of interest is a low density whose density is smaller than a threshold value by the density determination means. And when it is determined as data of the edge correction process for the pixel of interest, characterized by having a selection means for selecting the data generated by the error diffusion processing means.

  Alternatively, an image processing apparatus according to the present invention is an image processing apparatus that performs edge correction processing on an edge in an image, the density determination unit that determines the density of data of the pixel of interest, and the data of the pixel of interest. Error diffusion processing means for performing error diffusion processing, dither processing means for performing dither processing on the data of the target pixel, and an edge for performing edge border processing from the data of the target pixel and data generated by the dither processing means Edge processing means, and data mixing means for generating data after edge correction processing based on data generated by the edge border processing means and data generated by the error diffusion processing means as the edge correction processing And the mixing means increases the density of the target pixel so that the edge fringing process is performed with respect to the error diffusion process. The case is increased, characterized by increasing the proportion of the error diffusion processing on the edge bordering process the lower the concentration of the target pixel.

  Alternatively, an image processing apparatus according to the present invention is an image processing apparatus that performs edge correction processing on an edge in an image, the edge intensity of data of a target pixel, and a density contrast with peripheral pixels of the target pixel. Edge strength calculating means for calculating a certain edge strength, error diffusion processing means for performing error diffusion processing on the data of the target pixel, dither processing means for performing dither processing on the data of the target pixel, When the edge strength of the target pixel calculated by the edge strength calculating means is large, the edge border processing means for performing edge border processing based on the data of the target pixel and the data generated by the dither processing means, If the data generated by the edge border processing means is selected and the result of the edge strength calculating means is small, the error is It characterized by having a selection means for selecting the data generated by the diffusion processing unit.

  According to the present invention, high-quality edge correction processing can be performed on the edge of image data.

It is a block diagram which shows the functional structure of the image processing apparatus which concerns on 1st Embodiment. 2 is a block diagram showing a functional configuration of a buffer 102. FIG. 3 is a block diagram showing a functional configuration of an edge strength detection unit 106. FIG. 4 is an example of a filter used in the filter calculation unit 300. 3 is a block diagram illustrating a functional configuration of an edge determination unit 110. FIG. FIG. 6 is a processing flowchart of the edge fringing correction unit 105. It is a processing flowchart of the correction process determination part 107 which concerns on 1st Embodiment. It is a figure which shows the edge correction process result using the output EN from the edge fringing correction | amendment part 105. FIG. It is a figure which shows the edge correction process result using output ED from the error diffusion process part 104. FIG. It is a figure at the time of performing the edge correction process by an edge edging, and the edge correction process by an error diffusion process with respect to the image data with a high density | concentration of an edge part. It is a figure at the time of performing the edge correction process by an edge edging, and the edge correction process by an error diffusion process with respect to the image data with a low density | concentration of an edge part. According to the second embodiment. It is a block diagram which shows the functional structure of an image processing apparatus. It is a processing flow figure of the mixture ratio selection part 1201 which concerns on 2nd Embodiment. It is a block diagram which shows the functional structure of the mixing part 1202 which concerns on 2nd Embodiment. It is an example of the table stored in the mixture ratio table 1401 which concerns on 2nd Embodiment. FIG. 10 is a diagram illustrating a result of edge correction processing using an output EN from the edge fringing correction unit 105 and an output ED from the error diffusion processing unit 104. It is a block diagram which shows the functional structure of the image processing apparatus which concerns on 3rd Embodiment.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings. The embodiment described below is an example as means for realizing the present invention, and the present invention can be applied to a modified or modified embodiment described below without departing from the spirit of the present invention.

[First Embodiment]
FIG. 1 is a block diagram showing a functional configuration of an image processing apparatus according to the first embodiment of the present invention. In the figure, an image processing apparatus 101 and a printer 11 are connected. Hereinafter, the image processing apparatus 101 will be described sequentially.

  The buffer 102 temporarily accumulates input image data, and stores data for a plurality of lines. In the present embodiment, it is assumed that the image data representing each pixel in 8 bits has a capacity for storing one page of image data. Here, the input image data IMAGE is an image input from an external device (not shown) (for example, a computer device or a controller document reading device), and the input image data IMAGE is, for example, 8-bit gradation data. It is.

  The image data stored in the buffer 102 is input to the dither processing unit 103, the error diffusion processing unit 104, the edge fringing correction unit 105, the edge strength calculation unit 106, and the correction processing determination unit 107 in pixel order in raster order. Here, image data input to each image processing unit in raster order is expressed as a pixel of interest.

  The dither processing unit 103 generates gradation pixel data DITHER (hereinafter referred to as DITHER) from the target pixel data included in the input image data according to a dither processing method. The dither processing unit 103 is a process of selecting a threshold value to be applied to the target pixel data from a threshold matrix stored in advance, comparing the selected threshold value with the target pixel data, and outputting a quantized value. In the threshold matrix, thresholds are held as a matrix based on the positional relationship of pixels. By the dither processing, a threshold value based on the positional relationship of the target pixel is selected from the threshold matrix and compared with the target pixel value. At this time, if the target pixel exceeds the threshold value, a quantized value corresponding to the threshold value is output, and if the target pixel is lower than the threshold value, a quantized value obtained by lowering the quantization level by one is output. The quantization width differs depending on the resolution of the printer 111. For example, when applied to a printer capable of pulse width control capable of expressing one pixel with a plurality of gradations, the quantization width of the number of gradations that can be expressed. Can have up to.

  On the other hand, the error diffusion processing unit 104 generates gradation pixel data ED from the target pixel data included in the input image data according to the error diffusion method.

  The edge fringing correction unit 105 outputs data EN used for edge fringing correction based on the target pixel data and DITHER output from the dither processing unit 103.

  The edge strength calculation unit 106 detects the edge strength from the target pixel data included in the input image data and its peripheral pixel data, and outputs an output signal 1061. Then, the edge determination unit 110 outputs an edge determination signal EDGE_DETECT according to the input edge strength (output signal 1061).

  The correction processing determination unit 107 uses the data EN output from the edge fringing correction unit 105 based on the density of the target pixel data included in the input image data, or the gradation pixel data ED output from the error diffusion processing unit 104. Specify whether to use. Specifically, the selection signal SELECT for selecting which of the data EN and the gradation pixel data ED is output to the edge correction selection unit 108.

  The edge correction processing selection unit 108 selects the edge correction processing ED by error diffusion or the edge correction processing EN by edge processing based on the instructed selection signal SELECT, and outputs either of them as EDGE. The EDGE output here is used by the flat edge selection unit 109 only when the EDGE_DETECT signal indicates an edge. Here, the data for one pixel is sequentially output.

  The flat edge processing selection unit 109 selects either DITER output from the dither processing unit 103 or edge correction processing image data EDGE output from the edge correction processing selection unit 108 in accordance with the selection signal SELECT output from the edge determination unit 110. Select. Then, it is output as an output image data signal 1091.

  The printer 111 is, for example, a laser beam printer, and performs image formation by determining the laser irradiation time of each pixel in accordance with the output image data 1091. In the case where the printer 111 is an LED printer using an LED array, the LED irradiation time is determined to form an image. In the present embodiment, a printer capable of pulse width control capable of expressing one pixel with a plurality of gradations will be described. The details of each block in FIG. 1 will be described below.

○ Buffer 102
FIG. 2 is a block diagram illustrating functions of the buffer 102. In the buffer 102, first, the input image data IMAGE is stored in the page memory 201. The image data IMAGE of the page memory is vertical (n pixels) × horizontal (m pixels). The data stored in the page memory 201 is selected by the selector 203 in a 3 × 3 pixel (202) matrix centered on the pixel of interest C. The three pixels in the first row of the matrix are sent to the edge strength calculation unit 106 through the signal line 204a, the three pixels in the second row including the pixel of interest through the signal line 204b, and the three pixels in the third row through the signal line 204c. At this time, assuming that (row number, column number), the selector 203 sequentially selects the target pixel C as (2, 2) (2, 3)... (2, n−1). Similarly, the selector 203 selects the target pixel C from the second row to the (n−1) th row. At this time, the pixels sent to the edge intensity calculation unit 106 as 204a to 204c are also changed sequentially.

  At this time, the target pixel data C is also sent to the dither processing unit 103, the error diffusion processing unit 104, the edge fringing correction unit 105, and the correction processing determination unit 107 at the same time.

○ Edge strength calculation unit 106
FIG. 3 is a block diagram illustrating a functional configuration of the edge strength calculation unit 106. Image data input from the buffer 102 through the signal lines 204 a to 204 c is stored in the buffer 301. The buffer 301 includes three shift registers 301a to 301c that can store image data for three pixels. Accordingly, the buffer 301 can store image data for nine pixels. Here, the pixel located at the center is set as the target pixel 308.

  The image data for nine pixels stored in the buffer 301 is input to the filter operation unit 300 through the signal line 302 in order from the head address of the buffer 301a. When transmission of the data stored in the buffer 301a is completed, transmission is performed sequentially from the top address of the buffer 301b. Thereafter, data from 301a to 301c are transmitted in the same manner.

  The filter calculation unit 300 performs filter calculation processing on the input image data for nine pixels. The filter calculation unit 300 determines whether or not the target pixel C is an edge based on the density difference (density contrast) between the target pixel C and the surrounding pixels. In the following, a Laplacian filter is used in FIG. 4 as a filter used for the filter calculation processing described in detail. The filter operation unit 300 performs a multiplication process of the image data sequentially input from the buffer 301 and the filter, and performs an addition process on the result of the multiplication process of all the pixels. For example, when the Laplacian filter of FIG. 4 is used as the filter, the image data at the head address of the buffer 301a is input, and at the same time, a value is read from the head address 400a of the filter, and multiplication processing with the input data is performed. When the address of the buffer 301a is changed and the next data is input, the read address of the filter is also shifted, and the value read from the address 400b is multiplied by the input data. In the same manner, input data and a filter are sequentially multiplied, and a value obtained by adding the results is output as an edge strength value 1061. This addition process substantially corresponds to a process for evaluating the density difference between the pixel of interest and its surrounding pixels.

○ Edge determination unit 110
FIG. 5 is a block diagram illustrating a functional configuration of the edge determination unit 110. The edge determination determination unit 502 refers to the edge determination threshold value from the edge intensity value 1061 input from the edge intensity detection unit 106, determines whether or not to perform edge correction processing on the target pixel 308, and outputs it as an edge determination result EDGE_DETECT To do.

  A method of selecting output image data by the edge fringing correction unit 105 will be described with reference to FIG. The edge fringing correction unit 105 selects either DITER or C each time image data C of the pixel of interest is sequentially input, and outputs it as an output signal EN. First, the DITER output from the dither processing unit 103 is referred to (S601). When DITHER = 0, the image data C of the target pixel is selected and output as output data EN (S602). If DITTER ≠ 0, DITHER is output as output data EN (S603).

○ Correction processing determination unit 107
FIG. 11 is a processing flow of the correction processing determination unit 107. First, the correction process determination unit 107 reads a threshold value TH in order to select an edge correction process (S1101). In the present embodiment, the threshold TH is read from a value stored in advance, but a value determined by density information measured by a sensor may be read, or TH may be changed for each color. That is, the threshold value TH is not uniquely determined, but an optimum edge correction process may be selected.

  The threshold value TH is compared with the image data C (target pixel data C) input to the correction processing determination unit 107 (S1102). This corresponds to density determination. If the input image data C is greater than or equal to the threshold value TH, a signal for selecting edge correction processing for high density is output as SELECT = 0 (S1103). If the input image data C is smaller than the threshold value TH, a signal for selecting edge correction processing for low density is output as SELECT = 1 (S1104).

○ Edge correction processing selection unit 108
Next, the edge correction process selection unit 108 will be described. The edge correction process selection unit 108 selects the edge fringing correction EN or the error diffusion process ED based on the value of the selection signal SELECT input from the correction process determination unit 107.

  When the selection signal SELECT indicates edge correction processing for high density, as described above, the correction processing by the edge fringing correction unit 105 is optimal. Therefore, in this case, the edge correction processing selection unit 108 selects the EN data generated by the edge edging processing unit 105 as the edge correction processing data, and sets it as the output value EDGE (EDGE = EN).

  On the other hand, when the selection signal SELECT selects edge correction processing for low density, the edge correction processing selection unit 108 generates ED data generated by the error diffusion processing unit 104 as edge correction processing data. The output value EDGE (EDGE = ED).

○ Flat edge processing selection unit 109
The flat edge processing selection unit 109 outputs EDGE as output image data 1091 as a flat part when it is regarded as an edge by the EDGE_DETECT signal from the edge determination unit 110, and as a flat part when it is not regarded as an edge.

  As a result, the flat edge processing selection unit 109 determines that the output EN of the edge fringing correction unit 105 is determined that the target pixel C is an edge by the edge determination unit 110 and the edge correction processing selection unit 108 determines a high density. Is selected as edge correction data. Further, as a result, the flat edge processing selection unit 109 determines that the output ED of the error diffusion processing unit 104 has the target pixel C determined as an edge by the edge determination unit 110 and a low density by the edge correction processing selection unit 108. In this case, it is selected as edge correction data.

  The above is the detailed description of each block in FIG. In the following, an image when the edge fringing correction process by the edge fringing correction unit 105 is reflected and when the error diffusion processing by the error diffusion processing unit 104 is reflected as the edge correction processing will be described with reference to FIGS. .

  First, the case where the output EN of the edge fringing correction unit 105 is selected as the edge correction processing by the flat edge processing selection unit 109 will be described in detail with reference to FIG.

  FIG. 8A shows an example of input image data. The image data of FIG. 8B is obtained by dithering the input image data of FIG. 8A by the dither processing unit 103. In the edge portion having an angle close to the phase component of the dither, the quantization error becomes large as in 702, and the edge cannot be reproduced properly, and jaggy occurs. FIG. 8C is a diagram in which an edge is determined by the edge determination unit 110 and the determination result is shown as image data. The edge intensity detection unit 106 calculates an edge intensity value from the image data of 9 pixels stored in the line buffer 301 as described above. Here, it is assumed that the image data stored in the line buffer 301 is 703, for example.

  Next, the filter operation unit 300 performs a filter operation between the data stored in the line buffer 301 and the edge determination filter. Here, it is assumed that the edge determination filter is a Laplacian filter shown in FIG. When the filter operation is performed with the data 703 stored in the line buffer and the Laplacian filter 400, 384 is output as an edge intensity value through the signal line 1061.

  The edge determination unit 110 compares the edge intensity value input via the signal line 1061 with the threshold value read from the edge determination threshold value 501. Then, the edge determination unit 110 outputs an edge detection signal as an EDGE_DETECT when the edge intensity value is greater than or equal to the threshold, and an edge non-detection signal when the edge strength value is smaller than the threshold. Here, assuming that the edge determination threshold value stored in 501 is 192, the edge intensity value is larger, and therefore it can be considered that the center pixel of the input image data 703 has detected an edge.

  FIG. 8C shows a result obtained by performing the above edge determination processing on the input image data in FIG. For image data, a pixel indicating an edge detection unit is indicated by an image 703a, and a pixel indicating an edge non-detection portion is indicated by an image 703b. As shown in FIG. 8C, it can be confirmed that only the edge portion in the image is detected.

  FIG. 8D shows the edge determination result. The output EN of the edge fringing correction unit 105 is applied only to the portion that is regarded as the edge detection unit in FIG. 8C and is determined to have a high density by the edge correction processing selection unit 108. The image data in the case is shown. Each location of 704a and 704b will be described in detail using the edge border processing flow of FIG.

  First, in the case of the location of 704a, the input image value C input to the edge fringing process 105 is 128 (704a), and DITHER is 255 (704b). From S601 in FIG. 6, since DITHER is not 0, S603 is executed and 255 is output as the output value EN (704d).

  Similarly, in the case of the location of 705a, the input image value C input to the edge fringing process 105 is 128 (705a), and DITHER is 0 (705b). From S601 in FIG. 6, since DITER is 0, S602 is executed and 128 is output as the output value EN (705d).

  FIG. 8D shows the result of applying the output EN of the edge fringing correction unit 105 only to the portion where the edge is detected as described above. In FIG. 8D, high-resolution correction is performed by the edge trimming effect, and it is possible to compensate for data at a portion where the quantization error is large, so that jaggies can be suppressed.

  In the present embodiment, the configuration in which the input image data is added when the dither screen data is 0 is described as the edge edging correction processing, but other edging processing may be used. For example, a configuration in which input image data is directly output to the edge portion, or edge correction processing in which dither screen data and input image data are compared and the larger value is output may be performed.

  The case where the flat edge processing selection unit 109 applies the error diffusion processing output ED to the edge detection portion will be described with reference to FIG. FIG. 9A shows image data obtained by performing error diffusion processing on the input image data shown in FIG. FIG. 9B is a diagram illustrating image data when the image data subjected to the error diffusion processing illustrated in FIG. 9A is applied to the portion determined as the edge portion in FIG. 8C. FIG. 9B is a diagram showing image data when the image data subjected to the dither processing shown in FIG. 8B is applied to a portion determined as a non-edge portion. FIG. 9A shows image data when ternary error diffusion processing is performed as an example of error diffusion processing. When the error diffusion process is used, the occurrence of jaggies can be suppressed because no interruption of the edge portion occurs, but since a missing dot is randomly generated in the image, a fringe pattern that causes image degradation is generated. Therefore, by applying error diffusion processing only to the edge portion, the granularity of the flat portion can be maintained as shown in FIG. 9B, and jaggy can be suppressed.

  As described above, the edge correction processing by the edge fringing correction unit 105 and the edge correction processing by the error diffusion processing unit 104 are effective in suppressing jaggies, respectively.

FIG. 10 illustrates the edge correction processing by the edge fringing correction unit 105 and the edge correction processing by the error diffusion processing unit 104 for image data with high edge portion density. It is a figure at the time of going. FIG. 11 is a diagram in a case where edge correction processing by the edge fringing correction unit 105 and edge correction processing by the error diffusion processing unit 104 are performed on image data having a low density of the edge portion.

  The effectiveness of the processing for switching the edge correction processing by the edge fringing correction unit 105 and the edge correction processing by the error diffusion processing unit 104 according to the image density which is the characteristic part of the present case will be described with reference to FIGS.

  In FIG. 10, (a) is a diagram showing an example of image data in which the density of the edge portion is high. FIG. 10B shows image data obtained by dithering the input image data shown in FIG. FIG. 10C is a diagram showing image data obtained by performing error diffusion processing on the input image data shown in FIG. In FIG. 10D, the image data subjected to the edge correction processing by the edge edging processing unit 105 is applied to the portion determined as the edge portion, and when it is determined as the non-edge portion, the dither shown in FIG. It is a figure which shows the case where the processed image data is applied. FIG. 10E applies the image data subjected to the error diffusion processing shown in FIG. 10C to the portion determined as the edge portion, and FIG. 10B shows the case where it is determined as the non-edge portion. It is a figure which shows the case where the image data which carried out the dither process is applied. Further, the portion determined to be an edge is the same as that in FIG.

  When dither processing is performed on image information having a high density as shown in FIG. 10A, the ratio of the screen to the background increases as shown in FIG. 10B, but the occurrence frequency of jaggies decreases. , 901b, part of the edge rattling occurs.

  FIG. 10D illustrates image data obtained by performing edge correction processing on the image data illustrated in FIG. As shown in FIG. 10 (d), since the dot 901a, which is the input image, is added between the screens where the jaggy is generated in 901b, the jaggy is suppressed and the edge portion can be expressed optimally. You can see that In addition, by adding dots between the screens, the edge portion is emphasized, and an edge can be expressed with very high definition and smoothness in a character or the like whose outline is to be emphasized.

  On the other hand, FIG. 10E shows image data obtained as a result of performing edge correction processing by replacing the edge portion of the image data in FIG. 10A with data by the error diffusion processing unit 104. As shown in FIG. 10 (e), it can be seen that the jaggy of the edge portion is not completely removed at a location such as 901e. This is because, since error diffusion processing is used for the edge portion, dot dropout occurs randomly, dots do not occur at locations where jaggies are to be suppressed, and jaggies may not be completely removed.

As described above, in the case of image data having a high edge density as shown in FIG. 10A, the edge is emphasized by the correction process by the edge fringing correction unit 105 rather than the edge correction process by the error diffusion processing unit 104. Is also found to be optimal.
Further, in FIG. 11, (a) is a diagram showing an example of image data having a low density of the edge portion. FIG. 11B shows image data obtained by dithering the input image data shown in FIG. FIG. 11C illustrates image data obtained by performing error diffusion processing on the input image data illustrated in FIG. In FIG. 11D, the image data subjected to the edge correction processing by the edge edging processing unit 105 is applied to the portion determined to be an edge portion, and when it is determined to be a non-edge portion, the dither shown in FIG. It is a figure which shows the case where the processed image data is applied. FIG. 11E applies the image data subjected to the error diffusion processing shown in FIG. 11C to the portion determined as the edge portion, and FIG. 11B shows the case where it is determined as the non-edge portion. It is a figure which shows the case where the image data which carried out the dither process is applied. Further, the location determined to be an edge is the same location as in FIG.

  When dither processing is performed on image information having a low density as shown in FIG. 11A, the ratio of the screen to the background is small as shown in FIG. There will be a touch.

  FIG. 11D illustrates image data obtained by performing edge correction processing on the image data illustrated in FIG. As shown in FIG. 11D, it can be seen that the dot 1001a as the input image is added between the screens in the portion where the jaggy is generated in 1001b. However, in general, a low-density dot is easily affected by the durability of the engine and the use environment, and there is a problem that the dot reproducibility is unstable. In addition, there is a portion where the distance between the screens becomes long at the edge portion (for example, between 1002 and 1003). That is, when edge correction processing by the edge fringing correction unit 105 is performed on image data having a low density, a long distance edge is corrected using dots with unstable reproducibility (for example, 1004). portion).

  For example, when a dot of 1001d is reproduced at a lower density than desired, the edge edging process is not effective and jaggy is not improved. Also, if the 1001d dot is reproduced higher than the desired density, the edge edging process is more than necessary, resulting in an image that is unnaturally edged compared to the flat part inside the edge.

  On the other hand, FIG. 11E shows image data obtained as a result of performing edge correction processing by replacing the edge portion of the image data in FIG. 11A with the data of the error diffusion processing unit 104. The edge correction process by the error diffusion process generates dot missing at random, so that jaggies cannot be completely suppressed, but the edge part can be expressed with a period higher than the period of the screen. In addition, the dot reproducibility can be improved by using a high-density portion where the output value is stable without increasing the quantization value of error diffusion so much. That is, by performing edge correction processing by the error diffusion processing unit on low-density image data, it is possible to express the edge portion with high resolution using dots with stable reproducibility.

  Thus, in the case of image data having a low edge density as shown in FIG. 11A, it is more appropriate to express the edge at a higher resolution by the edge correction processing by the error diffusion processing unit 104. It can be seen that the correction processing by the unit 105 is also optimal.

  By performing the above processing, edge correction processing is performed by switching between edge fringing processing and error diffusion processing based on the density information of the target pixel, thereby suppressing jaggies from high density to low density edges and smoothly expressing edges. it can. In the description of FIG. 1 above, it has been described that the output (DITHER) from the dither processing unit 103 is directly input to the flat edge processing selection unit 109 and the DITER is assigned to the non-edge portion. However, the image processing of the non-edge portion is not limited to this form. For example, the output ED from the error diffusion processing unit 104 may be input instead, and the ED may be assigned to the non-edge portion. Alternatively, both the output DITER of the dither processing unit 109 and the output ED of the error diffusion processing unit 104 may be input, and either one may be selectively used depending on the image content of the non-edge portion image.

[Second Embodiment]
In the first embodiment described above, the configuration in which the edge fringing process and the error diffusion process are switched based on the density information of the target pixel has been described. However, in such image processing, in the case of an image such as a gradation, there may be a case where switching between edge fringing processing and edge correction processing by error diffusion processing is visible. For this reason, in the present embodiment, a configuration will be described in which switching is made inconspicuous by gradually switching between edge fringing processing and edge correction processing by error diffusion.

  FIG. 12 is a block diagram illustrating a functional configuration of the image processing apparatus according to the second embodiment. Description will be made centering on the difference from FIG. 1. Reference numeral 1201 denotes a mixing ratio selection unit that selects a mixing ratio of edge fringing processing and error diffusion processing according to input image data. The mixing ratio selection unit 1201 outputs a signal for selecting a ratio for mixing the edge border processing data and the error diffusion processing data as a SELECT signal according to the density of the input image. A mixing unit 1202 reads the mixing ratio from the mixing ratio table in accordance with the SELECT signal, and mixes and outputs the density of the edge fringe processing data EN and the error diffusion processing data ED at the read mixing ratio. Based on the data generated by the edge fringing correction unit 1202 (corresponding to the edge fringing processing unit) and the data generated by the error diffusion processing unit as edge correction processing by the mixing unit 1202 (data mixing unit). The data after the edge correction process is generated. Then, the edge correction data generated here can make the switching between the edge edging process and the edge correction process by the error diffusion process inconspicuous. Since other configurations are the same as those of the first embodiment, the same reference numerals are given and description thereof is omitted.

  FIG. 13 is a processing flow diagram of the mixing ratio selection unit 1201. In step S1301, the mixing ratio selection unit 1201 first reads threshold values TH1 to TH3 from a memory (not shown) in order to select a mixing ratio. Note that TH1> TH2> TH3. In the present embodiment, three threshold values are read. However, the number of the threshold values can be increased by the number of mixing ratios of one or two.

  In step S1302, the mixture ratio selection unit 1201 compares the threshold value TH1 with the input image data C. If the input image data C is a value greater than or equal to the threshold value TH1, the mixing ratio selection unit 1201 selects a signal for selecting a mixing ratio for mixing the high-density edge correction processing 100% and the low-density edge correction processing 0%. Is output as 0 (S1303). On the other hand, if the input image data C is a value smaller than the threshold value TH1, the mixture ratio selection unit 1201 shifts the processing to step S1304.

  In step S1304, the mixture ratio selection unit 1201 compares the threshold value TH2 with the image data C input to the correction processing determination unit 1201. Then, if the input image data C is greater than or equal to the threshold value TH2, the mixing ratio selection unit 1201 outputs a signal for selecting a mixing ratio for mixing the high density edge correction processing 75% and the low density edge correction processing 25%. 1 is output (S1305). On the other hand, if the input image data C is smaller than the threshold value TH2, the mixture ratio selection unit 1201 shifts the processing to step S1306.

  In step S1306, the mixture ratio selection unit 1201 compares the threshold value TH3 with the image data C input to the correction process determination unit 1201. If the input image data C is a value equal to or greater than the threshold value TH3, the mixture ratio selection unit 1201 selects a signal for selecting a mixture ratio for mixing the high density edge correction process 50% and the low density edge correction process 50%. = 2 is output (S1307). On the other hand, if the input image data C is smaller than the threshold value TH3, the mixing Hirotsu selection unit 1201 selects a mixing ratio for mixing the edge correction processing 0% for high density and the low density edge correction processing 100% in step S1308. The signal to be output is output as SELECT = 3.

  According to the flowchart of FIG. 13, the mixing ratio selection unit 1201 selects an appropriate mixing ratio and notifies the mixing unit 1202 as a SELECT signal. Accordingly, the mixing unit 1202 can increase the ratio of the edge edging process to the error diffusion process as the density of the target pixel is higher, and can increase the ratio of the error diffusing process to the edge edging process as the density of the target pixel C is lower. . As a result, the switching between the edge edging process and the edge correction process by the error diffusion process can be made inconspicuous.

  FIG. 14 is a block diagram showing a functional configuration of the mixing unit 1202. A mixing ratio table 1401 stores a ratio of mixing the density of the edge fringing process and the error diffusion process according to the value of the SELECT signal. It is stored in a memory (not shown). The mixing calculation processing unit 1402 performs a mixing calculation process on the density of the edge fringing process and the error diffusion process based on the ratio read from the mixing ratio table.

  FIG. 15 shows an example of the mixing ratio table 1401. The ratio x of the high density edge correction process and the ratio y of the low density edge correction process are stored in accordance with the value of the selection signal SELECT.

The mixing calculation processing unit 1402 acquires the high density edge correction processing ratio x and the low density edge correction processing ratio y from the mixing ratio table 1401 according to the value of the input mixing ratio selection signal SELECT. The calculation is performed according to the following expression according to the acquired ratio, and EDGE is output as an output value. Note that the output of EDGE is in units of pixels.
EDGE = x × EN + y × ED Equation 1

  In FIG. 16, (a) is a diagram showing an example of input image data. FIG. 16B is a diagram showing image data obtained by dithering the input image data shown in FIG. FIG. 16C is a diagram showing the input image data shown in FIG. 16A and the image data obtained by performing edge border correction processing by the edge border correction unit 105 from the dither image data shown in FIG. FIG. 16D is a diagram showing image data obtained by performing error diffusion processing on the input image data shown in FIG. FIG. 16E is a diagram in which the edge determination unit 110 performs edge determination on the input image data illustrated in FIG. 16A and the edge determination result is displayed as image data. FIG. 16F is a diagram in which the input image data shown in FIG. 16A is determined by the mixing ratio selection unit 1201 and a mixing ratio selection signal as a determination result is shown as image data. FIG. 16 (g) mixes the edge border correction processing data shown in FIG. 16 (c) and the error diffusion processing data shown in FIG. 16 (d) in accordance with the mixing ratio, and the mixed data is shown in FIG. It is the figure applied only to the location determined as the edge part shown by e). The mixing ratio here is read from the mixing ratio table 1401 by the mixing ratio selection unit 1201 based on the value of the mixing ratio selection signal shown in FIG.

  FIGS. 16A, 16B, 16C, 16D, and 16G show 8-bit image data. For each pixel of the image data, a pixel having a density value of 255 is set as a black pixel. The pixel image shown in 1601a is a white pixel, and the pixel image shown in 1601g is a pixel having a density value of 0. For other halftone density values, pixels with density values 1 to 63 are represented by a pixel image indicated by 1601f, pixels having density values 64 to 95 are represented by a pixel image indicated by 1601e, and pixels having density values 96 to 127 are displayed. Is represented by a pixel image represented by 1601d, pixels having density values of 128 to 191 are represented by pixel images represented by 1601c, and pixels having density values of 192 to 254 are represented by pixel images represented by 1601b. FIG. 16A shows that the input pixel groups 1604a to 1604d have the same density. The density is 146 in 1604a, the density is 109 in 1604b, the density is 73 in 1604c, and the density is 36 in 1604d. To do.

  The operations of the mixing ratio selection unit 1201 and the mixing unit 1202 will be described with reference to FIG. When the input image shown in FIG. 16A is input to the mixing ratio selection unit 1201, a mixing ratio selection signal SELECT is output according to the processing flow shown in FIG. If the threshold values of the correction processing determination unit 1201 are, for example, TH1 = 128, TH2 = 112, and TH3 = 96, the density value at the pixel position 1605a is 146, so it is equal to or higher than TH1 and SELECT = 0 is output. Further, since the density value of the pixel position 1606a is 114, it is smaller than TH1 and equal to or higher than TH2, and SELECT = 1 is output.

  FIG. 16 (f) shows the mixture ratio selection signal SELECT output by performing the mixture ratio selection of the mixture ratio selector 1201 on the input image of FIG. 16 (a). Note that a pixel with SELECT = 0 is represented by an image image 1603a, a pixel with SELECT = 1 is represented by an image image 1603b, and a pixel with SELECT = 3 is represented by a pixel image 1603c.

  As described above, when the mixing ratio selection signal SELECT is input, the mixing unit 1202 reads the ratio x of the high density edge correction process and the ratio y of the low density edge correction process with reference to the mixing ratio table 1401. For example, assume that the table of FIG. 15 is stored as the mixing ratio table.

  Since the pixel position 1605f is SELECT = 0, x = 1.0 and y = 0. Based on the read ratio, the edge correction value EDGE is calculated from the edge border processing data EN and the error diffusion processing data ED as input data. Edge border processing data EN = 146 from pixel position 1605c, and error diffusion processing data ED = 192 from pixel position 1605d. The output EDGE of the mixing unit 1202 is EDGE = x × EN + y × ED = 1.0 × 146 + 0 × 192 = 146. Further, since the pixel position 1605e is detected as an edge by the edge determination unit 110, the flat edge processing selection unit 109 outputs EDGE = 146 to 111 as output image data (1605g).

  Since the pixel position 1606f is SELECT = 1, x = 0.75 and y = 0.25. Based on the read ratio, the edge correction value EDGE is calculated from the edge border processing data EN and the error diffusion processing data ED as input data. Edge border processing data EN = 255 from pixel position 1606c, and error diffusion processing data ED = 192 from pixel position 1605d. The output EDGE of the mixing unit 1202 is EDGE = x × EN + y × ED = 0.75 × 255 + 0.25 × 192 = 239. Further, since the pixel position 1606e is detected as an edge by the edge determination unit 110, the flat edge processing selection unit 109 outputs EDGE = 146 to 111 as output image data (1606g).

  As described above, according to the present embodiment, even in the case of an image such as a gradation, high-quality edge correction processing can be performed without noticeable a switching portion between edge fringing processing and edge correction processing by error diffusion processing.

[Third Embodiment]
In the first embodiment described above, the configuration in which the edge fringing process and the error diffusion process are switched based on the density information of the target pixel has been described. However, when the edge portion and the background have a high density to some extent, the edge fringing correction may emphasize the contour too much, resulting in an unnatural contour. In the present embodiment, a configuration for switching between edge fringing processing and error diffusion processing based on edge intensity values in order to prevent excessive edge fringing will be described.

  FIG. 17 is a block diagram illustrating a functional configuration of an image processing apparatus according to the third embodiment. The correction processing determination unit 1702 switches between edge correction processing by the edge fringing correction unit 105 and edge correction processing by the error diffusion processing unit 104 based on the edge edge strength value 1061 calculated by the edge strength calculation unit 106. The operation of the correction process determination unit 1702 will be described in detail below. Since other configurations are the same as those of the first embodiment, the same reference numerals are given and description thereof is omitted.

  The correction processing determination unit 1702 outputs a signal for selecting edge correction processing for high density (edge fringing processing) if the output 1061 of the edge intensity detection unit is a value equal to or larger than the threshold Eth. If the output 1061 of the edge detection unit is smaller than the threshold value Eth, a signal for selecting the edge correction process (error diffusion process) for low density is output.

  The edge correction processing selection unit 108 can select appropriate data for edge correction processing by outputting a signal from the correction processing determination unit 1702. More specifically, the edge correction processing selection unit 108 selects the data EN generated by the edge fringing correction unit 105 when the edge intensity calculated for the target pixel C is large. The edge correction processing selection unit 108 selects the data ED generated by the error diffusion processing unit 104 when the edge intensity calculated for the target pixel C is small. Thus, high-quality edge correction processing can be performed on the edge of the image data.

  The correction process determination unit 1702 selects the edge correction process based on the edge intensity value. For example, even when the edge density is high to some extent, the edge correction process is performed by the error diffusion process. Can be suppressed.

  In the description in FIG. 17, it has been described that the edge correction processing by the edge fringing correction unit 105 and the edge correction processing by the error diffusion processing unit 104 are switched according to only the determination by the correction processing determination unit 1702. However, it is not limited to this. For example, the correction processing determination unit 107 described in FIG. 1 may be provided in the block diagram of FIG. In that case, first, the output (determination result) of the correction processing determination unit 107 and the output of the edge strength calculation unit 106 are also input to the block of the edge correction processing selection unit 108. The high density edge correction process (edge fringing process) may be selected only when both outputs indicate selection of high density edge correction. By doing so, it is possible to realize edge correction processing with higher accuracy.

[Other Embodiments]
In the embodiments described above, the edge strength is calculated using a 3 × 3 Laplacian filter such as 400 for the filter calculation unit 300 of the edge strength calculation unit 106, but if the edge strength can be calculated, this is used. Not limited to. For example, it is needless to say that the object of the present invention can be achieved even if the edge strength is calculated from the maximum / minimum difference values in a 5 × 5 Laplacian filter or other filters or 3 × 3 matrix.

  The computer may function as the above-described image processing apparatus by reading and executing a computer program that realizes the functions of the above-described embodiments. Furthermore, the present invention can also be applied to a computer-readable storage medium that stores this computer program. That is, a computer-readable storage medium that records a program code of a software computer program that implements the functions of the above-described embodiments is supplied to a system or apparatus, and the computer (or CPU or MPU) of the system or apparatus stores the storage medium. This can also be achieved by reading and executing the program code stored in. In this case, the program code itself read from the storage medium realizes the functions of the above-described embodiment, and the storage medium storing the program code constitutes the present invention.

  As a storage medium for supplying the program code, for example, a flexible disk, a hard disk, an optical disk, a magneto-optical disk, a CD-ROM, a CD-R, a magnetic tape, a nonvolatile memory card, a ROM, a DVD, or the like is used. it can.

  Further, by executing the program code read by the computer, not only the functions of the above-described embodiments are realized, but also an operating system (OS) operating on the computer based on a finger instruction of the program code. However, it is needless to say that some or all of the actual processing is performed and the functions of the above-described embodiments are realized by the processing.

  Furthermore, after the program code read from the storage medium is written to the memory provided in the function expansion board inserted into the computer or the function expansion unit connected to the computer, the function is based on the instruction of the program code. It goes without saying that the CPU of the expansion board or function expansion unit performs part or all of the actual processing, and the functions of the above-described embodiments are realized by the processing.

DESCRIPTION OF SYMBOLS 101 Image processing apparatus 105 Edge edging correction | amendment process part 106 Edge strength detection part 107 Correction process selection part 108 Edge correction selection part 109 Flat edge process selection part

Claims (9)

An image processing apparatus that performs edge correction processing on edges in an image,
Density determination means for determining the density of data of the pixel of interest;
Error diffusion processing means for performing error diffusion processing on the data of the pixel of interest;
Dither processing means for performing dither processing on the data of the pixel of interest;
Edge edging processing means for performing edge edging processing based on the data of the pixel of interest and the data generated by the dither processing means;
When the target pixel is determined to have a high density, which is a density higher than a threshold value, by the density determination unit, data generated by the edge border processing unit as data of the edge correction processing for the target pixel And when the pixel of interest is determined to be a low density that is a density smaller than a threshold value by the density determination unit, the error diffusion processing unit uses the edge correction processing data for the pixel of interest as data. An image processing apparatus comprising: selection means for selecting data to be generated.   The edge determination means for determining whether or not the target pixel is edge data based on a difference in density between the target pixel and pixels around the target pixel. The image processing apparatus described. An image processing apparatus that performs edge correction processing on edges in an image,
Density determination means for determining the density of data of the pixel of interest;
Error diffusion processing means for performing error diffusion processing on the data of the pixel of interest;
Dither processing means for performing dither processing on the data of the pixel of interest;
Edge border processing means for performing edge border processing from the data of the target pixel and the data generated by the dither processing means;
As the edge correction processing, data mixing means for generating data after edge correction processing based on the data generated by the edge edging processing means and the data generated by the error diffusion processing means,
The data mixing means increases the ratio of the edge edging process to the error diffusion process as the density of the target pixel is higher, and increases the ratio of the error diffusion process to the edge edging process as the density of the target pixel is lower. An image processing apparatus. An image processing apparatus that performs edge correction processing on edges in an image,
Edge intensity calculation means for calculating the edge intensity of the data of the pixel of interest and the edge intensity which is the density contrast with the peripheral pixels of the pixel of interest;
Error diffusion processing means for performing error diffusion processing on the data of the pixel of interest;
Dither processing means for performing dither processing on the data of the pixel of interest;
Edge edging processing means for performing edge edging processing based on the data of the pixel of interest and the data generated by the dither processing means;
When the edge intensity of the pixel of interest calculated by the edge intensity calculating means is large, the data generated by the edge edging processing means is selected, and when the result of the edge intensity calculating means is small, the error diffusion processing is performed. An image processing apparatus comprising: selection means for selecting data generated by the means. An image processing method in an image processing apparatus for performing edge correction processing on edges in an image,
A density determination step for determining the density of the data of the pixel of interest;
An error diffusion processing step of performing error diffusion processing on the data of the pixel of interest;
A dither processing step of performing dither processing on the data of the pixel of interest;
An edge edging process step of performing an edge edging process based on the data of the pixel of interest and the data generated by the dithering process;
If the target pixel is determined to have a high density, which is a density higher than a threshold value, by the density determination step, data generated by the edge edging process step as data of the edge correction process for the target pixel And when the pixel of interest is determined to be a low density that is a density smaller than a threshold value by the density determination step, the error diffusion processing step uses the edge correction processing data for the pixel of interest. And a selection step for selecting data to be generated.   6. The edge determination step of determining whether or not the target pixel is edge data based on a difference in density between the target pixel and pixels around the target pixel. The image processing method as described. An image processing method in an image processing apparatus for performing edge correction processing on edges in an image,
A density determination step for determining the density of the data of the pixel of interest;
An error diffusion processing step of performing error diffusion processing on the data of the pixel of interest;
A dither processing step of performing dither processing on the data of the pixel of interest;
An edge edging process for performing edge edging from the data of the pixel of interest and the data generated from the dithering process;
As the edge correction process, it has a data mixing process for generating data after the edge correction process based on the data generated by the edge edging process process and the data generated by the error diffusion process process,
In the mixing step, the ratio of the edge edging process to the error diffusion process is increased as the density of the target pixel is higher, and the ratio of the error diffusion process to the edge edging process is increased as the density of the target pixel is lower. An image processing method. An image processing method in an image processing apparatus for performing edge correction processing on edges in an image,
An edge strength calculation step of calculating an edge strength of the data of the pixel of interest, which is a density contrast with a peripheral pixel of the pixel of interest;
An error diffusion processing step of performing error diffusion processing on the data of the pixel of interest;
A dither processing step of performing dither processing on the data of the pixel of interest;
An edge edging process step of performing an edge edging process based on the data of the pixel of interest and the data generated by the dithering process;
When the edge intensity of the pixel of interest calculated by the edge intensity calculation process is high, the data generated by the edge edging process is selected, and when the result of the edge intensity calculation process is small, the error diffusion process is selected. And a selection step of selecting data generated by the step.   5. A computer program for causing a computer to function as each means according to claim 1.

Images