JP2002290739A - Digital image processing apparatus and method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a digital image processing apparatus and method for preventing moire patterns from being generated without deteriorating a reproduced image with a simple circuit. SOLUTION: A complementary calculation unit 200 finds seven virtual sampling points which are identical to image data in sampling frequency but different in phase by complementary calculation. When it is determined that the current pixel is an edge portion by an image edge detecting unit 230, an image complementary data selecting unit 250 outputs image data which was subjected to filter processing for emphasizing the edge by an image edge emphasizing unit 240 to a subsequent stage. When it is determined that the current pixel is not an edge portion, and only when peal point or valley point is one respectively, the point is assumed to be a representative value. In other cases, the point of actual sampling is assumed to be a representative value. When forming an image, points to be plot-outputted are switched for every pixel by a position of above representative value.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a digital image processing apparatus and method, and more particularly, to a digital image processing apparatus and method for reducing sampling moiré generated when processing a halftone original image and improving image quality. It is.

[0002]

2. Description of the Related Art It is well known that in digital copiers, a reproduced image of a halftone dot document has a characteristic density unevenness called moiré (or moiré fringe), which deteriorates image quality. . The occurrence of moire is caused by the fact that the halftone dot period of the document and the sampling period when the image is decomposed into pixels during image reading interfere with each other and affect image data. Therefore, a method of reducing moire by locally smoothing image data is well known. However, this smoothing has the disadvantage that the sharpness of the image, particularly the sharpness of the edge portion, is lost. For example, an image with little moiré can be obtained by smoothing a halftone dot portion in a document, but an image with blur or low contrast can be obtained by smoothing a line drawing portion such as a character.

For this reason, edge enhancement correction (hereinafter, MTF)
Correction) to enhance the high-frequency component of the image, especially at the edge portion.
There is a disadvantage that moire is also emphasized as a result of the F correction. Further, if the image density of reading and writing is increased, a sharp image with less moiré can be obtained. However, increasing the pixel density complicates the configuration of the apparatus and increases the difficulty in improving the processing speed, thereby increasing the cost. There is a disadvantage that it becomes expensive.

As a conventional image processing apparatus for reducing sampling moiré and improving image quality, Japanese Patent Application No. 3-196623 (Japanese Patent Application Laid-Open No. 5-41793) previously filed by the present applicant has been proposed. is there. This image processing apparatus first interpolates a new image by interpolation based on the original image data (D) at an arbitrary position between each pixel of the original image data (D) of the original image A / D converted at a predetermined sampling frequency. To form interpolation image data (D '). Then, first image data (D + D ′) having a frequency higher than a predetermined sampling frequency is formed from the interpolated image data (D ′) obtained by the interpolation and the original image data (D). Then, the first image data (D + D ′) is converted to the second image data (D ″ = (D + D ′) / 2) at a predetermined sampling frequency.
By converting to, sampling moiré is reduced.

[0005]

However, in the above-mentioned image processing apparatus disclosed in Japanese Patent Application No. 3-196623,
Since the original image data and interpolation data are averaged and replaced with the original image data, moiré components generated in the original image cannot be sufficiently removed, and There is a problem that the image is blurred. Accordingly, it is an object of the present invention to provide a digital image processing apparatus and method for preventing occurrence of moiré in a reproduced image of a halftone original without substantially deteriorating sharpness and realizing it with a simple circuit. I have.

[0006]

To achieve the above object, the present invention has the following features. According to a first aspect of the present invention, there is provided a reading unit that reads a document image as image data by sampling at a predetermined frequency, a quantization unit that quantizes the image data read by the reading unit for each pixel, and a quantization unit. Image data interpolating means for calculating a plurality of interpolated data between pixels of data quantized by, an edge detecting means for detecting an edge portion from among read points by the reading means, a plurality of interpolated data, and a read point. And selecting means for selecting pixel data to be output using the detection result of the edge detecting means.

According to a second aspect of the present invention, there is provided an edge emphasizing means for emphasizing an edge part in the data quantized by the quantizing means, and the selecting means is provided when the detection result at the read point is an edge part. , Characterized by selecting data whose edge portion has been emphasized by the edge emphasizing means.

According to a third aspect of the present invention, when the point corresponding to the vertex or the valley of the density gradation of the original image is out of the reading point, the selecting means selects from a plurality of interpolation data. Features.

According to a fourth aspect of the present invention, there is provided an image output means comprising: a recognition means for recognizing a position of a point selected by the selection means with respect to a reading point; and an image output means for outputting a plotted output on a recording sheet. Is characterized in that the plot output position within one pixel is switched for each pixel according to the position with respect to the reading point by the recognition means.

According to a fifth aspect of the present invention, there is provided a reading step of reading an original image as image data by sampling at a predetermined frequency, a quantization step of quantizing the image data read in the reading step for each pixel, An image data interpolation step of calculating a plurality of interpolation data between pixels of the data quantized by the quantization step, an edge detection step of detecting an edge portion from among read points in the reading step, and a plurality of interpolation data; And a selecting step of selecting pixel data to be output from the reading points using the detection result of the edge detecting step.

According to a sixth aspect of the present invention, there is provided an edge emphasizing step for emphasizing an edge portion of the data quantized in the quantizing step. And selecting data in which an edge portion has been emphasized by the edge emphasizing step.

According to a seventh aspect of the present invention, when the point corresponding to the vertex or the valley of the density gradation of the original image is out of the reading point, the selecting step selects from a plurality of interpolation data. Features.

The invention according to claim 8 has a recognition step of recognizing the position of the point selected in the selection step with respect to the read point, and an image output step of outputting the data on a recording paper by plot output. Is characterized in that the plot output position within one pixel is switched for each pixel according to the position with respect to the reading point in the recognition step.

[0014]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiments of a digital image processing apparatus and method according to the present invention will be described in detail with reference to the drawings.

FIG. 1 is a diagram showing an external appearance of a digital copying machine as an embodiment of the present invention. This digital copier is a general digital copying machine that exposes and scans a document set on a document table (contact glass) to read an image, performs predetermined processing on the read image, and forms an image on recording paper. This machine is provided with an image processing function for reducing moire. The document set on the document table is separated into pixels by a CCD line sensor and read. The main scanning is performed electrically by a CCD line sensor, and the sub-scanning is performed by moving the relative position between the CCD line sensor and the document. In the figure, reference numeral 5 denotes an operation unit.

FIG. 2 is a block diagram schematically showing the configuration of the digital copying machine shown in FIG. The document set on the document table with the surface to be read facing down is exposed and scanned by the image reading unit 1, and the exposure light is input to the CCD line sensor 11 of the image reading unit 1. That is, the document image is decomposed into pixels and read. Furthermore, CCD
An analog image signal output from the line sensor 11 is amplified by an amplifier (AMP) 12 and converted (quantized) into digital image data by an A / D converter 13.

The image data (quantized data) converted into digital data is input to the image processing unit 2. First,
The shading correction circuit 21 binarizes the image data into white and black and performs shading correction. Thereafter, the moiré correction circuit 22 corrects the image data to reduce the moiré. The processing in the moiré correction circuit 22 is processing according to the present invention, and will be described later in detail.

The moiré-corrected image data is subjected to spatial filtering such as smoothing in a filter circuit 23. Further, in the γ correction circuit 24, the reflectance −
The density conversion and the γ correction processing for correcting the γ characteristics in the image recording unit 3 connected to the subsequent stage of the image processing unit 2 are performed. In the image processing circuit 25, gradation processing suitable for a line image and gradation processing suitable for a photographic image are performed according to the copy mode, and further requantized into image data according to the characteristics of the image recording unit 3. You.

The image recording unit 3 prints on recording paper based on the image data output from the image processing unit 2.

The above operation is controlled by the control unit 4 in response to an input instruction from the operation unit 5. The circuits shown in FIG. 2 except for the moiré correction circuit 22 are the same as the circuits used in a general digital copying machine or the like, and are well-known technologies, and thus the details are omitted.

FIG. 3 is a diagram showing an example of image data when a halftone dot document having a uniform density is read. In the present embodiment, as an example, it is assumed that image data is quantized to 8 bits (256 gradations). Each section represents a pixel, and the numerical value in the section represents the gradation of the pixel. Possible value of one pixel is 0
２５255. x is the main scanning direction and y is the sub-scanning direction. In the following description, this image data is represented by D. When viewed locally, it is difficult to recognize the presence and presence of moiré, but moire occurs in the image data D due to interference between the sampling frequency when decomposing into pixels and the halftone frequency of the original, When the image is reproduced without correcting the moiré, the moiré is observed on the reproduced image. Moire in the main scanning direction, in which the MTF deterioration in reading is relatively small, due to the difference in the scanning method, in which the main scanning direction is electrical scanning by a CCD line sensor and the sub-scanning direction is mechanical scanning. Tends to stand out.

FIG. 4 shows a schematic configuration of the moiré correction circuit 22. The moiré correction circuit 22 includes an interpolation calculation unit 200,
An edge detection unit 230, an edge enhancement unit 240, and an interpolation data selection unit 250 are provided. Edge detector 230
Is a circuit for detecting an edge of the read image data.
(Described later). The edge emphasis unit 240 includes an edge emphasis filter circuit, and has a function of emphasizing an edge of the read image data.

The interpolation calculation unit 200 obtains seven virtual sampling points D'1 to D'7 having the same sampling frequency but different phases from the image data D by interpolation from the image data D. Assuming that the sampling period of D and D'1 to D'7 is T, the phase of D'1 is delayed from D by T / 8. That is, the phase of D′ n is delayed from D by (n × T) / 8.

FIG. 5 is a diagram for explaining the operation of the interpolation operation in the interpolation operation section 200, and shows an example of the pixel gradation at the actual sampling point (read point) and the virtual sampling point (interpolation data) in the image data. It is a graph.
Virtual sampling point D 'for arbitrary image data Dn
In obtaining n, the real sampling image data Da, Db, D of two pixels before and after the virtual sampling point D′ n
The value of D'n is calculated from the values of c and Dd and the distance from the virtual sampling point. This interpolation calculation is based on a so-called cubic function interpolation method, and more specifically, by the following calculation formula.

[0025]

(Equation 1)

Here, n is an integer of 1 to 7. The coefficients (W1n, W2n, W3n, W4n) are values determined by a cubic function of the distance from the virtual sampling point.
Table 1 shows specific values of the coefficients.

[0027]

[Table 1]

The image data D'n for the virtual sampling point is obtained by sequentially performing the interpolation according to the above equation in the scanning direction.
(N is 1 to 7) is obtained for each pixel, and eventually D and D'n are as shown in FIG. D and D'n
Have the same resolution, and in any case, for example, moire is similarly generated for a halftone original or the like. However, when D and D'n (n is 1 to 7) are combined and considered, it corresponds to reading at eight times the predetermined sampling frequency, and moire on a halftone original is almost eliminated. For example, Db, D'1, D'2, D'3, D '
When image data is formed in the order of 4, D'5, D'6, D'7, Dc,..., And the image is reproduced based on the image data, the image size is enlarged eight times in the main scanning direction. However, moire hardly occurs in the reproduced image.

However, since it is not possible to output a reproduced image that has been enlarged eight times, image data having a period T is required. Here, in the image interpolation data selection unit 250, the density values of eight points in total, including seven virtual points and one real sampling point calculated by the interpolation calculation unit 200, and output from the edge detection unit 230 (described later). Based on the edge signal, it is determined which of the eight points of image data is to be selected as an output image or image data output from an edge emphasis unit 240 (described later).

First, the edge detecting section 230 will be described. FIG. 6 is a diagram for explaining the operation of the edge detection unit 230. (1) is an example of a graph representing image data having an edge. (2) is an example of an edge detection filter. (3) is a diagram showing an example of edge detection. (4) is a diagram showing an edge signal output to the subsequent stage.

For image data having an edge as shown in FIG. 6A, the edge detector 230
An edge portion of the image is detected by an edge detection filter as in (2). Here, the numerical values shown as data examples represent the image data as 8 bits, that is, the gradation of each pixel is represented by a value of 0 to 255. FIG.
(3) is an edge detection result. Further, a threshold value on the plus side and a minus side are provided for the detected edge as shown in FIG. 6 (3), and it is determined whether the detected edge is above a certain level or below a certain level. Then, the edge detection unit 230 transmits the determination result, that is, whether the currently focused pixel is on the L level side or the H level side of the edge, or is not the edge portion, to the image interpolation data selection unit 250 at the subsequent stage. . The edge signal in this case is shown in FIG.

The edge emphasizing section 240 emphasizes the high-frequency components at the edges by the MTF correction performed by the conventional filter circuit. The MTF correction is the same as the function used in a general digital copying machine or the like, and is a known technique, so that the details are omitted.

Next, selection processing by the image interpolation data selection section 250 will be described. Image interpolation data selection unit 25
In the case of 0, when the current pixel is determined to be an edge by the edge detection unit 230, it is determined that the present pixel is image data indicating a part of a large feature of the original image, and the edge enhancement unit 240 The image data (emphasized data) that has been subjected to the filter processing for emphasizing is output to the subsequent stage.

If the image interpolation data selection unit 250 determines that the image is not an edge, the following processing is performed. FIG. 7 shows an example of an image that is not an edge portion. FIG. 7 shows three consecutive pixels (n-
1, n, n + 1) in a graph. In the related art, the sampling values of the pixels n-1, n, and n + 1 are Db, Dc, and Dd, respectively. Focusing on the center pixel n, the image data of the center pixel n is Dc as described above, but this density value does not always indicate the characteristic point of the present pixel. This is because sampling is performed at a predetermined frequency, and the value detected by chance is only Dc. If such an image is located at another location and the sampling point is shifted, the density value at a different point will be the image of the present pixel. This is because it becomes data. The characteristic point of the present pixel n is a point B in FIG. 7 which is a pixel serving as a vertex of an image in the vicinity of the pixel n. Regardless of where the image pattern shown in FIG. 7 is located on the original, if the point B is detected as image data, the original image can be faithfully reproduced, and furthermore, moire generated due to variation due to sampling. Should be possible.

For the above reasons, the main image interpolation data selecting section 250 replaces the image data at the point B with Dc in this case and sets it as a representative value of the main pixel (pixel data to be output).
Table 2 shows the relationship between the arrangement of the image data at this time and the pixels as the representative values in the arrangement.

[0036]

[Table 2]

For example, as for item 1 in Table 2, as a sequence of image data, point a has the highest density among eight density values, and point a is only one vertex. In such a case, the point a is a representative value of the pixel. Further, for example, when 5 in Table 2 is described, as a sequence of image data, e
The point has the lowest density, and point e is the only valley point. In the case of such an arrangement, the point e is the feature point, so that it is set as the representative value of the pixel. The same applies to the other cases, where the point is the only vertex at the maximum density among the eight sampling points, or the point is the only valley point at the minimum density. Is the representative value of the pixel. When this condition is not satisfied, for example, when there are a plurality of vertices, when both vertices and valleys exist, and when all eight points have the same value, the representative value of the pixel is determined at a predetermined sampling point. At this point, in FIG. 5, Dc is used as a representative value.

Here, the significance of using the method of selecting the representative value of the pixel from the virtual sampling points and the sampling points in the cycle T as described above will be described below.

In the sampling of the cycle T, as shown in FIG. 5, a change in the density of an image in one line cannot be completely detected. For example, in Dc of FIG. 5 sampled at the period T, the image data at the point A sampled in the vicinity thereof cannot be read. The image data at the point A here has the highest density in the vicinity thereof and remarkably represents the features of the main image data. If this point cannot be reproduced, it becomes one factor that deteriorates the image.

Further, when the entire original is a uniform solid image and image data having a wave of light and dark shades as shown in FIG. Tends to deviate from the cycle of At this time, the highest density value in the vicinity could be read in a certain place, but in another place, image data shifted from the image data having the highest density value as shown in FIG. 7 would be read. It can be one of the causes. According to the present invention, such a cause can be effectively corrected.

Next, the image interpolation data selection section 250 will be described. FIG. 8 is a diagram showing a block configuration of the image interpolation data selection unit 250. The image data input in the main scanning direction is latched, and five systems (a,
b, c, d, and e) as the parallel data,
51 is input data. The virtual point calculation unit 251 performs a calculation based on the input of the five systems using the virtual point calculation formula of the previous equation (1). At this time, as shown in FIG. 9, virtual points 0, 1, 2, and 3 are calculated from the actual sampling points a, b, c, and d, and virtual points 5, 6, and 5 are calculated from the actual sampling points b, c, d, and e. 7,
8 and 9 are calculated. Thus, 10 points shown in FIG. 9 are calculated together with the actual sampling points c (4 and c are the same).
Next, the image data of the ten points is compared with the virtual point comparison unit 2 in FIG.
The respective values are compared by 52 to determine the representative value of the pixel.

Next, the processing flow inside the virtual point comparing section 252 will be described with reference to FIG. 10, FIG. 11, and FIG. Figure 1
0 is a flowchart showing a first processing example for obtaining the minimum value from the above 10 points. IMG [N] is (N +
1) The density value at the point is shown. First, 255 is substituted for MIN as an initial value (step S1). When N is 9 or less (step S2; NO), the density is compared with the density value IMG [N] of each point (step S3), and the density is determined. If the value is smaller than MIN, the value is substituted for MIN (step S4). After one cycle of this operation, 1 is added to N (step S5), and the operation from step S2 is repeated again. When N = 10, since the comparison has been completed for all ten points, the flow is terminated (step S2; YES). As described above, the minimum value is stored in MIN at the end of this flow.

FIG. 11 is a flowchart showing a second processing example for obtaining the maximum value from the above ten points. IMG
[N] indicates the density value at the (N + 1) -th point.
First, 0 is substituted for MAX as an initial value (step S11), and when N is 9 or less (step S12; N)
O), the density is compared with the density value IMG [N] of each point (step S13), and when the density value is larger than MAX, the value is substituted for MAX (step S14). After one cycle of this operation, 1 is added to N (step S15), and the operation from step S12 is repeated again. When N = 10, the comparison has been completed for all 10 points.
This flow ends (step S12; YES). As described above, the maximum value is stored in MAX at the end of this flow.

FIG. 12 is a flowchart showing a third processing example for determining the number of vertices and valley points.
DIR and CMP indicate gradients of density values of two consecutive image data. In any case, a value of -1 indicates a downward tilt, a value of 0 indicates no tilt, and a value of 1 indicates upward tilt. UP and DWN are variables for counting the number of vertices and valleys, respectively. IMG [N]
Indicates the density value at the (N + 1) -th point.

First, as initial values, UP, DWN, CMP
Is set to 0 (step S21). Then the first two
The gradient of the density value of the point is obtained (step S22), and the value is substituted for DIR (steps S23, S24, S2).
5). Thereafter, when N is 9 or less (step S2)
6; NO), and sequentially determine the inclination with the next point (step S2)
7), and substitute the value into CMP (step S28, S
29, S30). If the values of DIR and CMP are different (step S31; NO), the value of N is stored as NUM (step S32). When the inclination changes from top to bottom (step S33; NO), it is determined that there is a vertex, and the
Is added to (step S34). When the inclination changes from bottom to top (step S33; YES), it is determined that there is a valley point, and 1 is added to DWN (step S33).
35). When the above operation is completed, the value of CMP is substituted for DIR (step S36), 1 is added to N (step S37), and the operation from step S26 is repeated again.
This repetition is performed from the second point to the tenth point, and the process ends (step S26; YES).

When all points have been compared as described above, the numbers of vertices and valleys are stored in UP and DWN, respectively. In addition, NUM has a vertex,
Alternatively, in the case where there is a valley point, a value indicating which of the locations 1 to 8 in FIG.

Referring to these results, "UP value is 1 and DWN value is 0" or "UP value is 0 and DWN is 0
It can be seen that a pattern conforming to Table 2 exists only in two cases where the value of N is 1 ". Virtual point comparison unit 252
Outputs the representative values shown in Table 2 as correction data to the subsequent stage.

Further, the correction data is sent to the image switching unit 25.
3 is input together with the edge-enhanced image output from the edge-enhancing unit 240, and is switched by an edge signal output from the edge detection unit 230. That is, if the edge signal is a value indicating an edge, the edge-enhanced image output from the edge emphasizing unit 240 is selected. If the edge signal is a value indicating a non-edge, the correction data is selected.
As a result, the edge portion of the image is output more sharply, and the non-edge portion can suppress deterioration in image quality due to moiré.

Next, a case where the above embodiment further includes an image output device, that is, an embodiment as an image forming apparatus will be described.

By the step S32 shown in FIG.
In M, a value indicating which region in one pixel the feature point selected as a representative value of the pixel is stored. By using the value as phase data for controlling the output device, the plot position within one pixel can be set to the left, center,
It is possible to output while switching to the right side. The relationship between the position of the feature point selected as the representative value of the pixel and the phase control signal used for controlling the output device will be specifically described below. As shown in FIG. 7, of a total of eight virtual sampling points and real sampling points in one pixel, the first two are located on the left side in the pixel, the next three are in the center, and The next three are located on the right.
Also, with reference to FIG. 9, if there is a representative value at any of points 1 and 2, plot output is performed on the left side within one pixel, and if there is a representative value at any of points 3, 4, and 5, The plot is output at the center, and if any of the points 6, 7, and 8 has a representative value, the plot is output to the right within one pixel.
Regarding the above relationship, Table 2 shows selected feature points for the arrangement of images and the phase control signal at that time.

Although the embodiment of the present invention has been described as a digital copying machine, it is not limited to a digital copying machine as long as image reading by sampling is used. In addition, a digital multifunction peripheral having other functions may be used.

[0052]

As apparent from the above description, according to the digital image processing apparatus and method of the present invention, an edge is calculated from a plurality of interpolation data calculated between pixels of image data and a reading point of a document image. By selecting the pixel data to be output using the detection result by the detection means, it is possible to perform processing under the same effect as obtaining a sampling image at a frequency higher than a predetermined frequency. This allows
Even in a reproduced image of a halftone dot document, occurrence of moire can be prevented without impairing the sharpness of the image.

Further, according to the digital image processing apparatus and method of the present invention, when the reading point is an edge portion, the edge is emphasized and output, so that the edge portion which is a characteristic of the original image is sharpened without being damaged. And output an image with reduced moiré. Furthermore, it can be realized with a simple circuit.

Further, according to the digital image processing apparatus and method of the present invention, when a point corresponding to a vertex or a valley of a density gradation of an original image is out of a reading point, a selection is made from a plurality of interpolation data. As a result, it is possible to prevent the occurrence of moiré in a reproduced image of a halftone dot document without deteriorating the sharpness of the image, and it is possible to realize the moiré with a simple circuit.

Further, according to the digital image processing apparatus and method of the present invention, the position information of the selected point with respect to the read point is obtained, and the image output unit plots and outputs one pixel within each pixel based on the position information. By controlling the position to be switched, the occurrence of moire can be prevented, and a more faithful and sharper image of the document can be output.

[Brief description of the drawings]

FIG. 1 is a diagram illustrating an appearance of a digital copying machine as an embodiment of the present invention.

FIG. 2 is a block diagram showing a schematic configuration of a digital copying machine according to the present invention.

FIG. 3 is a diagram illustrating an example of image data when a halftone dot document having a certain density is read.

FIG. 4 is a diagram showing a schematic configuration of a moiré correction circuit 22;

FIG. 5 is a diagram for explaining an operation of an interpolation calculation in the interpolation calculation unit 200, and is a graph showing an example of pixel gradation in image data.

FIG. 6 is a diagram for explaining the operation of the edge detection unit 230.

FIG. 7 is a graph showing an example of pixel gradation in an image that is not an edge portion.

FIG. 8 is a diagram showing a block configuration of an image interpolation data selection unit 250.

FIG. 9 is a diagram for explaining the operation of the virtual point comparison unit 252;
6 is a graph showing an example of how real sampling points and virtual sampling points are arranged in image data.

FIG. 10 is a flowchart illustrating a first processing example of a virtual point comparison unit 252;

11 is a flowchart illustrating a second processing example of the virtual point comparison unit 252. FIG.

FIG. 12 is a flowchart illustrating a third processing example of the virtual point comparison unit 252;

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Image reading part 2 Image processing part 3 Image recording part 4 Control part 5 Operation part 11 CCD line sensor 12 AMP 13 A / D converter 21 Shading correction circuit 22 Moiré correction circuit 23 Filter circuit 24 γ correction circuit 25 Image processing circuit 200 Interpolation calculation unit 230 Edge detection unit 240 Edge enhancement unit 250 Image interpolation data selection unit 251 Virtual point calculation unit 252 Virtual point comparison unit 253 Image switching unit

Claims (8)

[Claims] A reading unit that reads a document image as image data by sampling at a predetermined frequency; a quantization unit that quantizes the image data read by the reading unit for each pixel; Image data interpolating means for calculating a plurality of interpolated data between pixels of quantized data; edge detecting means for detecting an edge portion from reading points by the reading means; the plurality of interpolated data; Selecting means for selecting pixel data to be output using the detection result of the edge detecting means from the points. 2. The image processing apparatus according to claim 1, further comprising: an edge emphasizing unit for emphasizing an edge portion of the data quantized by the quantizing unit; wherein the selecting unit is configured to determine whether the detection result at the reading point is an edge portion. 2. The digital image processing apparatus according to claim 1, wherein said data whose edge portion is emphasized by an edge emphasizing means is selected. 3. The method according to claim 1, wherein when a point corresponding to a vertex or a valley of a density gradation of the document image is out of the reading point, the selection unit selects from the plurality of interpolation data. The digital image processing device according to claim 1. 4. An image processing apparatus comprising: a recognition unit configured to recognize a position of a point selected by the selection unit with respect to the reading point; and an image output unit configured to output a plotted image on a recording sheet. 4. The digital image processing apparatus according to claim 1, wherein a plot output position within one pixel is switched for each pixel by a position with respect to the reading point by means. 5. A reading step of reading a document image as image data by sampling at a predetermined frequency, a quantization step of quantizing the image data read by the reading step for each pixel, and An image data interpolation step of calculating a plurality of interpolation data between pixels of the quantized data; an edge detection step of detecting an edge portion from read points in the reading step; the plurality of interpolation data; And selecting a pixel data to be output using the detection result of the edge detection step from the points. 6. An edge emphasizing step for emphasizing an edge portion of the data quantized by the quantization step, wherein the selecting step includes the step of: when the detection result at the read point is an edge portion. 6. The digital image processing method according to claim 5, wherein the data whose edge portion is emphasized by the edge emphasizing step is selected. 7. When the point corresponding to the vertex or valley of the density gradation of the document image is out of the reading point, the selecting step selects from the plurality of interpolation data. The digital image processing method according to claim 5. 8. A recognition step for recognizing a position of the point selected by the selection step with respect to the reading point, and an image output step of outputting the plotted output on a recording sheet, wherein the image output step includes: The digital image processing method according to any one of claims 5 to 7, wherein a plot output position within one pixel is switched for each pixel according to a position with respect to the reading point in the step.