JP3476649B2 - Image processing method and apparatus - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention reads halftone originals.
The present invention relates to an image processing method and apparatus for a digital image obtained by taking the image.

[0002]

2. Description of the Related Art Conventionally, halftone dots have been used for printing printed matter. Such a printed matter printed with halftone dots has, for example, 65 dpi to 200 dpi.
It has a resolution of about i.

FIG. 9 shows a case where a printed matter recorded with such halftone dots is read by an image input unit such as a scanner. FIG. 9 is a diagram showing a positional relationship between a halftone dot and a pixel to be read, and a halftone dot image which is an original image shows a density of 50%. That is, the hatched area shown in the figure is a portion where halftone dots are recorded. A case where such an original image is read by the pixels P1 to P8 having a size indicated by a dotted frame will be described.

One pixel size shown in FIG. 9 is approximately 87.5% of one halftone dot size. In such a case, the densities measured in the pixels P4 and P5 are 50%, which is equal to the densities shown in the original image, so that the consistency is maintained, but the densities measured in the pixels P1 and P8 are the same in the pixel region. Is larger than 50% because it includes many blackened areas of halftone dots. Actually, the density of the pixels P1 and P8 is about 6
It is 2%. In addition, the pixels P2, P3, P6, P7
The concentration measured in 1 is also higher than 50%. Thus, the densities measured in the pixels P1, ..., P8 fluctuate periodically between 50% and about 62%. This means that the image quality is deteriorated when considering that the original image has a density of 50%.

This is because interference occurs because the dot size of the original image and the pixel size to be read are different. Therefore, as the pixel size becomes smaller than the halftone dot size, the density of the halftone dot is read rather than reading the density of the original image, and the variation in the density between pixels becomes large. Further, when the pixel size is close to the halftone dot size, the fluctuating period becomes long, and it becomes visually noticeable as moire.

The above phenomenon also occurs when the relationship between the pixel size and the pattern in the general original image represented by a method other than the halftone dot and the pattern of the object is in the same state as described above. That is, moire may occur in an image obtained by scanning a transparent original or a reflective original with a scanner or an image obtained with a digital camera.

Conventionally, image processing using an image filter has been performed in order to remove such moire and prevent deterioration of image quality.

FIG. 10 is a schematic block diagram of a conventional image processing apparatus. The main signal S for the pixel to be processed (hereinafter referred to as “target pixel”) is subjected to a filter calculation process by an image filter of a predetermined size preset in the filter calculation unit 400, and a filtered signal S ′ is obtained. Is generated. FIG. 11 shows an example of the image filter used at this time. In such a conventional image processing apparatus, the horizontal size and the vertical size of the image filter are set to be about twice or more the size of each halftone dot. Therefore, for example, in the case of the example of the image filter of FIG. 11, one pixel size is about “2/5” or more of the halftone dot size. Then, in the conventional image processing apparatus, the center of the image filter is located at the pixel of interest in the filter calculation unit 400, and the weighted average of the respective density values is performed based on the weighting coefficient assigned to the pixel of interest and its neighboring pixels, and then filtered. It is output as a signal S '.

As described above, in the conventional image processing apparatus, by performing processing using the image filter as shown in FIG. 11, the weighted average of the densities is performed in the area (about 2 times) larger than the halftone dot size. Since the smoothing process is performed by the above, the frequency component of the halftone dot pattern can be removed, and the influence of the halftone dot size can be removed. Therefore,
No moiré or deterioration of image quality occurs in the filtered image.

[0010]

By the way, generally, an edge portion of an image of a printed matter recorded with halftone dots is reproduced with a resolution higher than the halftone dot size. FIG. 12 is an explanatory diagram showing a conventional method for generating one halftone dot. As shown in FIG. 12A, an area showing one halftone dot size is divided into four blocks B1, ..., B4, and each block is further subdivided into blackened areas. Then, a corresponding threshold value is set in each blackened region as shown in FIG. On the other hand, the density value obtained by reading the original image is the block B as shown in FIG.
The position corresponding to 1, B4 is "20", and the block B
When the position corresponding to 2 and B3 is “10”, the blackening is sequentially performed from the center of the halftone dot as compared with the threshold value shown in FIG. Then, the halftone dots as shown in FIG. 12C are recorded.

FIG. 13 shows a case where an edge portion of an image is recorded by such a recording method. FIG. 13 is a diagram showing an edge portion of a halftone dot image, and a dotted line in the figure is an edge of the image. The density on the right side of the dotted line is 50%, and the density on the left side is 0%. As shown in FIG. 13, halftone dots are recorded in the edge portion of the halftone image according to the density value corresponding to each of the four divided blocks, so that reproduction is performed at high resolution.

However, in the conventional image processing apparatus, when smoothing is performed using an image filter as shown in FIG. 11 as a measure against moire and deterioration of image quality, the edge of the image reproduced at the high resolution as described above. The part is also smoothed,
There is a problem of being blurred.

Therefore, the present invention has been made in view of the above problems, and provides an image processing method and apparatus for suppressing deterioration of image quality and removing moire without deteriorating the edge of the image. The purpose is to do.

[0014]

[0015]

[Means for Solving the Problems] To achieve the above object
The invention according to claim 1 is a method of subjecting an original image obtained by reading a halftone dot original document to a predetermined process for each pixel, and (a) the vicinity of a target pixel to be processed. By performing smoothing processing on each of a plurality of neighboring pixels located in the area in a predetermined area larger than each halftone dot size that constitutes the halftone dot original document around the neighboring pixel, The step of obtaining an average density value for each of the neighboring pixels; (b) the step of obtaining the contrast for the pixel of interest based on the plurality of average density values obtained from the plurality of neighboring pixels; and (c) the contrast A step of determining a weighting coefficient assigned to each component of the image filter, and (d) applying a filter operation to the main signal of the pixel of interest using the image filter to which the weighting coefficient is applied, and outputting Has a step of generating the issue, the size of the image filter applied in step (d),
It is characterized in that it is larger than the size of each halftone dot forming the halftone dot original.

The invention according to claim 2 is the method according to claim 1, wherein the step (b) is different from each other based on a plurality of average density values obtained from a plurality of neighboring pixels. It is characterized in that the method includes a step of obtaining contrasts in a plurality of directions, and the step (c) includes a step of changing respective distributions of effective weighting coefficients in a plurality of directions according to the contrasts in the plurality of directions. There is.

[0017]

According to a third aspect of the present invention, there is provided an apparatus for subjecting an original image obtained by reading a halftone dot original document to a predetermined process for each pixel, wherein (a) a target pixel to be processed. By performing smoothing processing on each of a plurality of neighboring pixels located in the vicinity in a predetermined area larger than each halftone dot size of the halftone dot original document surrounding the neighboring pixel, Means for obtaining the average density value for each of the neighboring pixels, (b) means for obtaining the contrast for the pixel of interest based on the plurality of average density values obtained from the plurality of neighboring pixels, and (c) according to the contrast. , A means for determining a weighting coefficient assigned to each component of the image filter, and (d) an image filter to which the weighting coefficient is applied, for performing a filter operation on the main signal of the pixel of interest. Comprises a means for generating an output signal, the size of the image filter applied by means (d) it is it is characterized in greater than the individual dot size constituting the dot document.

According to a fourth aspect of the present invention, in the apparatus according to the third aspect , the means (b) is provided for a plurality of directions different from each other based on a plurality of average density values obtained from a plurality of neighboring pixels. The means (c) includes means for obtaining contrast, and means (c) includes means for changing respective distributions of effective weighting coefficients in a plurality of directions according to contrasts in a plurality of directions.

[0020]

DETAILED DESCRIPTION OF THE INVENTION

<1. Overall Configuration of Device> First, an example of the overall configuration of a device to which an embodiment of the present invention is applied will be described. FIG. 1 is a schematic diagram showing the overall configuration of an apparatus to which an embodiment of the present invention is applied. The image input unit 100 optically reads a document, such as an input scanner, and generates original image data indicating multivalued density values for each pixel. Then, the generated original image data is transferred to the image processing unit 200. The image processing unit 200 is a processing unit to which the image processing device according to the embodiment of the present invention is applied. In the image processing unit 200, in addition to the image processing using the image filter according to the embodiment of the present invention, a predetermined process is performed on the original image data, and then the image output unit 300 such as an output scanner. Is output to. The image output unit 30
At 0, recording is performed on a recording medium such as a film. In order to cause the image input unit 100, the image processing unit 200, and the image output unit 300 to perform an operator's desired operation, an operation input unit 201 such as a keyboard and a mouse and an information display unit 202 such as a display display unit. And are provided.

<2. Outline of Image Processing> Next, a processing mode in the image processing apparatus according to the embodiment of the present invention will be described. In the image processing device according to the embodiment of the present invention, image processing using an image filter is performed. That is, processing such as image smoothing is performed by scanning an M × N (M and N are odd numbers of 3 or more) matrix image filter in the image plane.

FIG. 2 is an explanatory diagram showing an outline of image processing in the embodiment of the present invention. With respect to the image I shown in FIG. 2A, the X direction (horizontal direction) is the main scanning direction and the Y direction (vertical direction) is the sub scanning direction. I will go. At the time of processing by the image filter, the pixel of interest OP is positioned at the center of the image filter F as shown in FIG. Figure 2
In the example of (b), a 5 × 5 (M = N = 5) matrix image filter is used. Then, based on the weighting coefficient assigned to each component of the image filter, a weighted average of the density value of the pixel of interest OP and the density values of its neighboring pixels is derived, and the obtained value is used as the filtered signal of the pixel of interest OP. To do.

[0023] That is, image filter F is the size of the matrix as shown in FIG. 3 (2m + 1) × ( 2n + 1), the weighting coefficient of each component k _-mn to k _mn is assigned . The filtered signal S ′ _xy obtained by performing the filter calculation on the pixel of interest OP located at the coordinates (x, y) on the image plane using this image filter F is

[0024]

[Equation 1]

[0025] Here, S _xy is the density value of the pixel located at the coordinates (x, y) on the image plane of the original image.

In the image filter shown in FIG. 3,
When all the weighting coefficients corresponding to the pixel of interest and its neighboring pixels are positive numbers, it is an image filter for smoothing the image, and the frequency characteristic of the weighting coefficient is the characteristic of the image filter.

<3. Image Processing Device> Next, an image processing device according to an embodiment of the present invention will be described. Figure 4
FIG. 1 is a schematic configuration diagram showing an image processing device according to an embodiment of the present invention. The main signal S, which is the density value of the pixel of interest, is guided to the smoothing processing unit 10 and the filter calculation unit 50.

In the smoothing processing unit 10, when the main signal S for the target pixel is input, smoothing processing is performed for the pixels in the vicinity of the target pixel to obtain an average density value. The neighboring pixels of the pixel of interest for which the average density value is obtained can be four neighboring pixels such as an upper neighboring pixel, a lower neighboring pixel, a left neighboring pixel, and a right neighboring pixel. As the neighboring pixels of these target pixels, the pixels P1, P2, P3, and P4 located on the upper side, the right side, the lower side, and the left side of the target pixel as shown in FIG. Although it may be a right neighboring pixel, a lower neighboring pixel, or a left neighboring pixel, as shown in FIG. 5B, a pixel P located at a position several pixels away from each of the upper side, right side, lower side, and left side of the target pixel.
5, P6, P7, and P8 may be an upper neighborhood pixel, a right neighborhood pixel, a lower neighborhood pixel, and a left neighborhood pixel, respectively.

Then, these upper neighboring pixel, right neighboring pixel,
For each of the lower neighboring pixel and the left neighboring pixel, smoothing processing is performed in a region (smoothing region) that is larger than the halftone dot size centered on each neighboring pixel. To make the smoothing area larger than the halftone dot size, the halftone dot shape is not detected, and pixels near the center position of the halftone dot and pixels not
This is to prevent the density values from being greatly different from each other and to obtain the density value originally shown by the original image.

The smoothing processing performed here is to perform a simple average of the density values of the pixels located in the smoothed area,
Alternatively, it is performed by weighted averaging. The average density value for the upper neighbor pixel thus obtained is YU, the average density value for the lower neighbor pixel is YL, the average density value for the left neighbor pixel is XL, and the average density value for the right neighbor pixel is XR. And Then, the smoothing processing unit 10 sends the average density values YU, YL for the upper and lower neighboring pixels to the vertical direction contrast extraction unit 21, and calculates the average density values XL, XR for the left and right neighboring pixels. It is sent to the lateral direction contrast extraction unit 22.

The vertical direction contrast extraction unit 21 extracts the contrast in the vertical direction (Y direction) which is the sub-scanning direction near the pixel of interest in the image. That is, using the average density values YU and YL for the input upper and lower neighboring pixels,

[0032]

[Equation 2]

The vertical direction contrast Cy is obtained by the following.
Then, the obtained vertical contrast Cy is sent to the vertical table reference unit 31.

Similarly, the horizontal direction contrast extraction section 22 extracts the contrast in the horizontal direction (X direction) which is the main scanning direction in the vicinity of the pixel of interest of the image. That is, the average density value X for the input left neighboring pixel and right neighboring pixel X
Using L and XR,

[0035]

[Equation 3]

Then, the lateral contrast Cx is obtained.
Then, the obtained lateral contrast Cx is sent to the lateral table reference unit 32.

In general, a high contrast indicates an edge portion of an image, and a low contrast indicates a flat portion in which the density change of the image is small. Therefore, by obtaining the contrasts in the vertical direction and the horizontal direction as described above, it is possible to determine what part of the image the pixel of interest is. Further, since the smoothing processing unit 10 has performed the smoothing processing,
It does not detect the edge of the halftone dot itself.

The vertical table reference unit 31 is internally provided with a storage medium such as a memory. By referring to a table stored in advance, a reference corresponding to the input vertical contrast Cy is made. Output the value IDy.

Similarly, the lateral table reference unit 32 also has a storage medium such as a memory inside and by referring to a table stored in advance,
Reference value I corresponding to the input lateral contrast Cx
Output Dx.

FIG. 6A is a diagram showing the relationship between the input vertical contrast Cy and the output reference value IDy in the vertical table reference unit 31, and FIG. 6B is the horizontal table reference unit 32. 6 is a diagram showing a relationship between a lateral contrast Cx that is an input and an output reference value IDx in FIG. The relationship between such input and output is stored in a memory or the like as a table. As shown in FIGS. 6A and 6B, the contrast C in both the vertical direction and the horizontal direction is
The smaller y and Cx, the smaller the output values of IDy and IDx, and the larger the contrast, the larger the values of IDy and IDx. In other words, when the pixel of interest is the edge part of the image, the values of IDy and IDx are large, and when it is the flat part of the image, the values of IDy and IDx are small. The IDy and IDx obtained by the vertical table reference unit 31 and the horizontal table reference unit 32 are sent to the weighting coefficient set selection unit 40.

As shown in FIG. 6A, the relationship between the vertical reference value IDy and the vertical contrast Cy is that the reference value IDy is increased by one each time the contrast Cy shows a substantially constant increase. On the other hand, in the horizontal reference value IDx shown in FIG. 6B, the value “1” extends to the vicinity of the intermediate value of the contrast, and the value “2” has a high contrast. It extends to the area. That is, even if the horizontal and vertical contrasts are the same, the horizontal reference value IDx may be smaller than the vertical reference value IDy. This is because, when the input scanner of the image input unit 100 optically reads a document, the nature of noise differs between the main scanning direction (horizontal direction) and the sub-scanning direction (vertical direction). For example, the reading of the original by the input scanner in the main scanning direction is performed by a CCD line sensor in which a plurality of pixels are arranged in a straight line, and the reading in the sub-scanning direction is made constant by the driving mechanism such as a motor. An apparatus that is performed by moving at a speed will be described. In such a device,
Although noise appears in each pixel in the sub-scanning direction, it is possible to electrically scale in the main scanning direction. In such a case, noise included in one pixel extends in the main scanning direction. Shrink. If the noise extends in the main scanning direction, it is desirable to perform the smoothing process with a smoothing filter having a large size in the main scanning direction.
As shown in (b), a small reference value IDx may be output in the main scanning direction.

However, when there is no expansion or contraction of noise in the main scanning direction, the contrast Cx is applied to the horizontal reference value IDx as in the vertical reference value IDy of FIG. 6A.
The value of the reference value IDx may be increased by 1 every time when indicates a substantially constant increase.

Next, in the weighting coefficient set selection unit 40, the reference value ID sent from each table reference unit.
The image filter weighting coefficient k _-mn corresponding to y and _IDx
~ Select and determine k _mn . A storage unit such as a memory is provided inside the weighting coefficient set selection unit 40.
The storage unit stores the weighting coefficients corresponding to the input reference values IDy and IDx. Then, each weighting coefficient selected and determined by referring to the storage unit is sent to the filter calculation unit 50.

FIG. 7 is a diagram showing an example of an image filter obtained in the weighting coefficient set selection unit 40. As the value of the reference value IDy in the vertical direction increases, the weighting coefficient that deviates from the center of the image filter in the vertical direction decreases. That is, since the contrast is high in the vertical direction,
Since it is determined to be the edge portion of the image, the area or degree to be smoothed becomes small. Similarly, when the value of the reference value IDx becomes large, the weighting coefficient that deviates from the center of the image filter in the lateral direction becomes small. An edge portion of an image having a high contrast in the horizontal direction has a smaller area or degree of smoothing in the horizontal direction. Note that it is clear from Equation 1 that the component whose weighting coefficient assigned to the image filter is “0” is not an effective coefficient for processing by the image filter.

For example, in the case where the reference value IDy obtained by the vertical table reference unit 31 is "3" and the reference value IDx obtained by the horizontal table reference unit 32 is "1", FIG. The image filter F13 shown in is selected. The image filter F13 has a large smoothed area in the horizontal direction but a small smoothed area in the vertical direction. This is because the contrast in the vertical direction is large and the contrast in the horizontal direction is small. When the contrast is high in both the vertical and horizontal directions (IDy = 3, I
In the case of Dx = 3), the image filter F33 shown in FIG. 7 is applied, and the smoothed region becomes the smallest. This makes it possible to minimize the deterioration of the edge portion of the image. When the contrast is small in both the vertical and horizontal directions (when IDy = 1 and IDx = 1), the image filter F11 shown in FIG. 7 is applied and smoothing is performed on the entire 5 × 5 region.

The image filter shown in FIG. 7 shows an example in which the size of the image filter is 5 × 5, but the size is not limited to this. However, the size of the image filter needs to be larger than the size of each halftone dot of the original image in order to satisfactorily remove the deterioration of image quality and the occurrence of moire due to the effect of halftone dots. It is preferable that the size is about twice or more.

Then, the filter calculation section 50 performs the calculation processing shown in the equation 1 on the basis of each weighting coefficient sent from the weighting coefficient set selection section 40 to generate a filtered signal S '. This filtered signal S'becomes an output signal of the image processing apparatus of this embodiment. When the filtered signal S ′ is obtained in this way, the processing for the pixel of interest is completed. Then, the same process is performed with the pixel to be processed next in the next scanning order as the pixel of interest. By repeating this, the process is completed for the entire image.

Here, the data output from the weighting coefficient set selecting section 40 to the filter calculating section 50 may be the weighting coefficient of each component, but is not limited to this.

If a plurality of weighting factors are stored in advance in the storage unit such as RAM in the filter calculation unit 50,
The weighting coefficient set selection unit 40 can be omitted. That is, the vertical table reference unit 31 and the horizontal table reference unit 32 have the contrast C in each direction.
If the offset value of the address for accessing the RAM of the filter calculation unit 50 is output as the reference values IDy and IDx according to y and Cx, the filter calculation unit 50 will output the reference value ID.
By accessing the RAM based on the offset values indicated by y and IDx, it is possible to directly obtain each weighting coefficient corresponding to the contrast in the vertical direction and the contrast in the horizontal direction.

As described above, in the image processing apparatus of this embodiment, the image filter that does not deteriorate the edge is applied to the edge portion of the original image, and a larger size is applied to the flat portion of the image. An image filter is applied, so without degrading the edges of the image,
It is possible to satisfactorily suppress the deterioration of image quality and remove moire.

Further, since each weighting coefficient of the image filter to be applied is selected and determined based on the contrast in the vertical direction and the horizontal direction, it is possible to detect the directionality of the edge of the original image and save the edge. Image filters can be obtained. Therefore, it is possible to apply an image filter that preserves edges and suppresses moire and image quality deterioration.

<4. Modification> The size of the image filter applied by the filter calculation unit 50 shown in FIG. 4 needs to be larger than the halftone dot size of the original image in order to satisfactorily remove the occurrence of moire and the deterioration of image quality. That was already explained. Therefore, if the halftone dot size of the original image is changed, it may be necessary to change the size of the image filter applied by the filter calculation unit 50. Even in this case,
In order to properly perform the processing, the following processing may be performed.

That is, the image input section 100 (see FIG. 1)
If the halftone dot size can be detected when the original image is read in, the detected halftone dot size is input to this image processing apparatus. Then, in the image processing apparatus, the weighting coefficient set selection unit 40 may change the size of the image filter to an appropriate size when selecting the weighting coefficient. Further, when the halftone dot size is changed, the operator may set and input the halftone dot size from the operation input unit 201 and input the set and input halftone dot size to the image processing apparatus. With this configuration,
An image filter of an appropriate size is always applied according to an arbitrary halftone dot size, and the image processing apparatus can suppress the deterioration of the image quality satisfactorily without degrading the edge of the image regardless of the original image. And it is possible to remove moire.

Next, the contrast extraction described in the embodiment is performed in two directions, the vertical direction and the horizontal direction, but it is more preferable to extract the contrast in the diagonal direction. For example, it is preferable to apply the image filter so that the contrast is extracted in the diagonal direction to the right and diagonal direction to the left of the pixel of interest, and when the contrast is high in each direction, the edge in that direction is saved.

Further, the contrast may be obtained in directions other than the vertical, horizontal and diagonal directions. That is, in general, the directions may be different from each other in the image filter having a two-dimensional spread.

Next, in the above description, the respective reference values IDy and IDx obtained by the vertical table reference unit 31 and the horizontal table reference unit 32 are as shown in FIGS. 6 (a) and 6 (b). There were three values of "1", "2", "3",
As shown in FIGS. 8 (a) and 8 (b), it is possible to take a multi-level value of three or more values, and calculation is performed based on nine image filters as shown in FIG. The image filter may be derived. When such processing is performed, the accuracy of image processing is increased.

[0057]

[0058]

As described above, according to the first aspect of the present invention, for each of a plurality of neighboring pixels located in the vicinity of the target pixel to be processed, a network around the neighboring pixel is provided. Rather than the individual dot sizes that make up the dot manuscript
By performing the smoothing process in a large predetermined area, the average density value is obtained for each of the plurality of neighboring pixels, and the contrast for the target pixel is obtained based on the obtained plurality of average density values. Then, according to the contrast, the weighting coefficient assigned to each component of the image filter is determined, and the image signal to which the weighting coefficient is applied is subjected to a filter operation on the main signal of the pixel of interest to output the output signal. Is generated, but in this filter operation
The size of the applied image filter also depends on the halftone original.
It is larger than the size of each halftone dot.
Therefore, it is possible to favorably suppress the deterioration of the image quality and remove the moire without deteriorating the edge of the image.

[0059]

According to the second aspect of the present invention, when obtaining the contrast for the pixel of interest, the contrast is obtained for a plurality of different directions based on a plurality of average density values obtained from a plurality of neighboring pixels. , The step of changing the distribution of effective weighting factors in a plurality of directions according to the contrasts in a plurality of directions when determining the weighting factors assigned to the respective components of the image filter. It is possible to obtain an image filter according to the directionality of the edge. Therefore, it is possible to apply an image filter that preserves edges and well suppresses moire and image quality deterioration.

According to the third aspect of the invention, for each of the plurality of neighboring pixels located in the vicinity of the target pixel to be processed, a predetermined halftone dot original document around the neighboring pixel is provided.
By performing a smoothing process in an area larger than the individual halftone dot sizes that form the , the average density value is obtained for each of a plurality of neighboring pixels, and for the pixel of interest based on the obtained plurality of average density values. Find the contrast of.
Then, according to the contrast, the weighting coefficient assigned to each component of the image filter is determined, and the image signal to which the weighting coefficient is applied is subjected to a filter operation on the main signal of the pixel of interest to output the output signal. The image file to generate but apply in this filter operation
The size of the data also depends on the individual halftone dots that make up the halftone dot original.
It is larger than the size. Therefore, it is possible to favorably suppress the deterioration of the image quality and remove the moire without deteriorating the edge of the image.

[0062]

According to the fourth aspect of the present invention, when obtaining the contrast for the pixel of interest, the contrast for different directions is obtained based on the average density values obtained from the neighboring pixels. , The step of changing the distribution of effective weighting factors in a plurality of directions according to the contrasts in a plurality of directions when determining the weighting factors assigned to the respective components of the image filter. It is possible to obtain an image filter according to the directionality of the edge. Therefore, it is possible to apply an image filter that preserves edges and well suppresses moire and image quality deterioration.

[Brief description of drawings]

FIG. 1 is a schematic diagram showing an example of the overall configuration of an apparatus to which an embodiment of the present invention is applied.

FIG. 2 is an explanatory diagram showing an outline of image processing according to the embodiment of the present invention.

FIG. 3 is a diagram showing weighting coefficients of respective components of the image filter.

FIG. 4 is a schematic configuration diagram showing an image processing apparatus according to an embodiment of the present invention.

FIG. 5 is an explanatory diagram for explaining upper neighboring pixels and the like.

FIG. 6 is a diagram showing a relationship between inputs and outputs in a vertical table reference unit and a horizontal table reference unit.

FIG. 7 is a diagram showing an example of an image filter obtained by a weighting coefficient set selection unit.

FIG. 8 is a diagram showing a relationship between inputs and outputs in a vertical table reference unit and a horizontal table reference unit.

FIG. 9 is a diagram showing a positional relationship between halftone dots and pixels to be read.

FIG. 10 is a schematic configuration diagram showing a conventional image processing apparatus.

FIG. 11 is a diagram showing an example of a conventional image filter.

FIG. 12 is an explanatory diagram showing a conventional method for generating one halftone dot.

FIG. 13 is a diagram showing an edge portion of a halftone image.

[Explanation of symbols]

10 Smoothing processing unit 21 Vertical direction contrast extraction unit 22 Horizontal contrast extraction unit 31 Vertical table reference section 32 Horizontal table reference section 40 Weighting coefficient set selection unit 50 Filter calculator 100 image input section 200 Image processing unit 201 Operation input section 202 Information display section 300 image output section S main signal S'filtered signal

─────────────────────────────────────────────────── ─── Continuation of front page (58) Fields investigated (Int.Cl. ⁷ , DB name) H04N 1/40-1/409

Claims (4)

(57) [Claims] 1. Obtained by scanning a halftone dot original
It is a method to perform a predetermined process for each pixel on the original image.
Therefore , (a) a plurality of pixels located in the vicinity of the target pixel to be processed
For each of the neighboring pixels, before the neighborhood of the neighboring pixel
Larger than the size of each halftone dot in the halftone dot original
By performing smoothing processing in a predetermined area,
The average density value for each of the neighboring pixels of
And (b) the plurality of averages obtained from the plurality of neighboring pixels
Obtain the contrast for the pixel of interest based on the density value
And (c) each of the image filter components depending on the contrast.
Determining the weighting factors assigned to the minutes, and (d) using an image filter to which the weighting factors are applied.
Filter operation is performed on the main signal of the pixel of interest.
And, and a step of generating an output signal, the size of the image filter applied in the step (d)
Is the size of each halftone dot that composes the halftone dot original.
An image processing method characterized by being large. 2. The method according to claim 1, wherein the step (b) is performed by using a plurality of pixels obtained from the plurality of neighboring pixels.
A plurality of directions different from each other based on the average density value of the number
For the contrast in the plurality of directions.
Effective weighting for the multiple directions depending on the strike
It is characterized by including the step of changing the distribution of each coefficient
Image processing method. 3. Obtained by scanning a halftone dot original
A device that performs predetermined processing for each pixel on the original image
Therefore , (a) a plurality of pixels located in the vicinity of the target pixel to be processed
For each of the neighboring pixels, before the neighborhood of the neighboring pixel
Larger than the size of each halftone dot in the halftone dot original
By performing smoothing processing in a predetermined area,
Means for obtaining the average density value for each of the neighboring pixels of
And (b) the plurality of averages obtained from the plurality of neighboring pixels
Obtain the contrast for the pixel of interest based on the density value
And (c) each component of the image filter according to the contrast.
A means for determining a weighting factor assigned to a minute, and (d) using an image filter to which the weighting factor is applied.
Filter operation is performed on the main signal of the pixel of interest.
And, and means for generating an output signal, the size of the image filter applied by said means (d)
Is the size of each halftone dot that composes the halftone dot original.
An image processing device that is also large. 4. The apparatus according to claim 3, wherein the means (b) is a plurality of pixels obtained from the plurality of neighboring pixels.
A plurality of directions different from each other based on the average density value of the number
And a means for obtaining a contrast for the plurality of directions.
Effective weighting for the multiple directions depending on the strike
Characterized by including means for changing each distribution of the coefficient
Image processing device.