JPH09321989A - Edge area detector - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide the edge area detector at a low cost with a simple configuration by reducing a time required to detect an edge area of an image more than that by a conventional edge area detector having been used for an image area separator. SOLUTION: The edge area detector that detects an edge of an image expressed in image data based on received image data is provided with a smoothing means 44 applying smoothing processing to a noted picture element of received image data by using a smoothing matrix, a density difference detection means 46 detecting a density difference of two picture elements in existence at both sides of the noted picture element based on the received image data and a discrimination means 48 discriminating whether or not the noted picture element is a picture element of an edge area of an image based on the density of the noted picture element after the smoothing and the detection result by the density detection means and providing an output of a discrimination result.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an edge area detecting device used in an image forming apparatus such as a printer, a copying machine and a facsimile.

[0002]

2. Description of the Related Art When reproducing an image in which characters and pictures are mixed in an image forming apparatus such as a digital copying machine or a facsimile machine, the character area is subjected to image processing in which resolution is prioritized and the picture area is provided with gradation. By performing the image processing with priority, the reproduced image can be improved.
For this purpose, an image area separation device for separating an image into a character area and a picture area is required, and several image area separation devices have been proposed in the past (Japanese Patent Laid-Open No. 63-142766,
(JP-A-1-132247, JP-A-5-176167, etc.).

In these image area separating apparatuses, various methods such as those utilizing the periodicity of halftone dots in the picture area are adopted. Particularly, in the prior art described in JP-A-5-176167. Makes it possible to separate the character region and the pattern region with high accuracy based on the detection results of both the detection result of the edge region of the image and the detection result of the white background region of the image. Here, the edge area detection utilizes the fact that a character has an edge where the image density changes abruptly. Conventionally, as described in JP-A-5-176167, for example, a target pixel including a target pixel is included. This is performed by performing pattern matching using a 3 × 3 matrix or the like and determining the pixel of interest as an edge area pixel when the pixel matches a specific pattern.

[0004]

However, in the conventional edge area detecting apparatus, it is necessary to successively perform pattern matching with a large number of matching patterns in order to determine whether or not the pixel of interest is a pixel in the edge area of the image. There is a problem that it takes time to make a determination. Further, not only storage means for storing a large number of matching patterns is required, but also image data needs to be stored over a plurality of lines for a predetermined time in order to perform pattern matching, so a large number of line memories are required. hand,
This has caused the device to be complicated and costly.

The present invention has been made in view of the above problems, and the purpose thereof is to compare with the prior art,
It reduces the time required to detect the edge area of the image,
An object of the present invention is to provide an edge area detection device having a simple structure and low cost.

[0006]

In order to solve the above problems, an edge area detecting device according to claim 1 detects an edge area of an image represented by the image data based on the input image data. In the edge area detection device, a smoothing unit that performs a smoothing process using a smoothing matrix on a target pixel of input image data, and two pixels existing on both sides of the target pixel based on the input image data. Based on the density of the target pixel after the smoothing and the detection result of the density detecting unit, it is determined whether or not the target pixel is a pixel in the edge region of the image. It is characterized in that a determination means for outputting a result is provided.

In the edge area detecting apparatus according to the first aspect, the density after the smoothing processing of the target pixel is obtained by the arithmetic processing using the smoothing matrix, and the density of the target pixel is calculated based on the input image data. The density difference between two pixels on both sides is detected. Then, based on the density of the pixel of interest after the smoothing process and the detected density difference,
When the density of the target pixel is higher than a predetermined value and the difference in density between two pixels on both sides of the target pixel is larger than the predetermined value, the target pixel is regarded as a pixel belonging to an edge region. judge. This eliminates the need for pattern matching, which has been performed in a conventional device.

An edge area detecting device according to a second aspect is the edge area detecting device according to the first aspect, wherein a white background area detecting means for detecting a white background area existing around a target pixel and a detection result of the white background area detecting means are used. The present invention is characterized in that a selecting means for selecting a judgment result obtained by using a plurality of corresponding smoothing matrices based on the detection result of the white background area detecting means is provided.

According to another aspect of the edge area detecting device of the present invention, the judgment result obtained by using a plurality of types of smoothing matrices corresponding to the detection result of the white background area detecting means for detecting the white background area of the image is used as the white background. The selection is made based on the detection result of the area detecting means. Here, it is empirically known that there is a high probability that a white background area exists around the pixels of the character area in the image, and that there is a high probability that a white background area exists around the pixels that belong to the pattern area. It is empirically known to be small. Therefore, the detection result of the white background area detecting means indicates whether the target pixel belongs to the character area or the picture area. Then, depending on the detection result of the white background area detecting means, when the determination result of the edge area detecting device obtained by differentiating the smoothing matrix used in the smoothing means is selected, an approximate character area and an approximate pattern area are selected. According to the above, the determination result of the edge area detecting means can be changed. Specifically, the probability of determining the target pixel as a pixel belonging to the edge region can be increased in the character region, and the probability of determining the target pixel as a pixel belonging to the edge region can be reduced in the picture region.

According to a third aspect of the present invention, in the edge region detecting apparatus according to the second aspect, the plurality of types of smoothing matrixes corresponding to the detection result of the white background area detecting means have the white background area as the detection result. It is characterized in that the matrix size corresponding to the absence of the white background area is larger than the matrix size corresponding to the case.

According to another aspect of the edge area detecting apparatus of the present invention, the size of the smoothing matrix corresponding to the detection result of the white background area not including the white background area is larger than the size of the smoothing matrix corresponding to the detection result including the white background area. To do.
As a result, the degree of smoothing of the target pixel after the smoothing processing when the detection result is that there is no white background area is larger than the degree of smoothing when the detection result is that there is a white background area. In the determination result of the determination means, the greater the degree of smoothing of the target pixel, that is, the smaller the density difference between the target pixel and the peripheral pixels, the lower the probability of being determined as an edge. Therefore, in the edge area detection device according to claim 3, when the white background area detection result is that there is no white background area, the probability that the pixel of interest is determined to belong to the edge area can be made lower than when the white background area is present. it can.

An edge area detecting device according to a fourth aspect is the edge area detecting device according to the second aspect, wherein the plurality of types of smoothing matrices corresponding to the detection result of the white background area detecting means have the same size, and Of the coefficients of the matrix, the coefficient corresponding to the pixel of interest is set larger in the matrix corresponding to the white background area than in the smoothing matrix corresponding to the case where the detection result is the white background area is not included. To do.

According to another aspect of the edge area detecting apparatus of the present invention, the plurality of types of smoothing matrices have the same size. Therefore, in the above-described edge region detecting apparatus according to claim 3, it is necessary to prepare the circuit element for the smoothing processing so as to match the size of the smoothing matrix of the maximum size. When the smoothing process is performed using the matrix, there occurs a problem that some of the circuit elements are wasted. However, in the edge area detection device according to claim 4, such a problem does not occur. Further, among the coefficients of the matrix, the coefficient corresponding to the pixel of interest is
The detection result is larger in the matrix corresponding to the white background area than in the smoothing matrix corresponding to the case where the white background area is not present. As a result, the degree of smoothing of the target pixel after the smoothing processing when the detection result is that there is no white background area is larger than the degree of smoothing when the detection result is that there is a white background area. In the determination result of the determination means, the greater the degree of smoothing of the target pixel, that is, the smaller the density difference between the target pixel and the peripheral pixels, the lower the probability of being determined as an edge. Therefore, in the edge area detection device according to claim 3, when the white background area detection result is that there is no white background area, the probability that the pixel of interest is determined to belong to the edge area can be made lower than when the white background area is present. it can.

[0014]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, the present invention is a digital copying machine as an image forming apparatus (hereinafter simply referred to as "copier").
The first embodiment applied to will be described. FIG.
1 is a schematic configuration diagram of a copying machine according to the present embodiment. The copying machine 1 according to the present embodiment is mainly composed of a scanner 2 and a printer 3. The scanner 2 is for optically reading a document image, and the printer 3 is for forming an image of the document image read by the scanner 2 on a recording medium, for example, recording paper.

The scanner 2 irradiates the original placed on the contact glass 17 which is the original placing table with the image surface facing downward in FIG. 6 with the exposure lamp 19, and reflects the reflected light from the first mirror 13 and the first mirror 13. It is guided to the imaging lens 16 by the second mirror 14 and the third mirror 15. The image forming lens 16 forms an image of the reflected light on a CCD (charge coupled device) 16. The CCD 16 photoelectrically converts the imaged reflected light to output an analog electric signal corresponding to the original image. The analog signal is converted into a digital signal by a processing circuit in the scanner 2, and shading correction processing such as variation of each element of the CCD 16 is performed to be image data and input to the image processing unit 5. . Then, the image data is further subjected to various kinds of image processing by the image processing section 5, and then sent to a writing unit 6 which is a writing means of the printer 3.

The writing unit 6 is provided with a laser diode (not shown) which generates a light beam based on the image data after the image processing sent from the image processing section 5. The light beam emitted by the laser diode is deflected by the polygon mirror 21 and guided onto the photoconductor 7 via the fθ lens and the mirror (not shown). Then, the surface of the photoconductor 7 uniformly charged by the charging device 8 is optically scanned.

The latent image formed on the photoconductor 7 by the optical scanning of the light beam is visualized by the developing device 9 and becomes a toner image. The toner image is transferred onto the recording paper conveyed from the paper supply unit 4 by a transfer device 10 provided opposite to the photoconductor 7. The recording paper on which the toner image is transferred is conveyed to the fixing device 13 by the conveying belt 12 and fixed by the fixing device 13. Thus, the image formation of the document image on the recording paper is completed. On the other hand, the cleaning device 11 collects and cleans the residual toner on the surface of the photoconductor 7 after the transfer by the transfer device is completed.

FIG. 2 is a block diagram showing the configuration of the digital copying machine according to this embodiment. As described above, the image data read by the scanner 2 of the copying machine 1 and completed with shading correction and the like is input to the image processing unit. here,
The image processing unit in the present embodiment will be further described. The image data from the scanner 2 has a two-way data flow. That is, one of them is the gamma correction circuit 20,
The flow reaches the printer 3 via the filter circuit 21, the magnification changing circuit 22, and the pseudo halftone processing circuit 23. The other is the flow input to the edge area detection device 24.

The edge area detecting device 24 is a device for detecting the edge area of the image from the input image data, and detects whether or not the pixel belongs to the edge area for each pixel of the image. , Output the result. The detection result of the edge area detection device 24 is input to each of the gamma correction circuit 20, the filter circuit 21, the scaling circuit 22, and the pseudo halftone processing circuit 23, as shown in FIG. Each circuit such as the gamma correction circuit 20 performs processing corresponding to the edge area and the non-edge area of the image based on the output signal of the edge area detection device 24. For example, for edge areas, processing with priority on resolution is performed, and for non-edge areas, processing with priority on gradation is performed.

The edge area detecting device 24 will be described below with reference to FIG. 3 which is a schematic block diagram thereof.
The edge area detection device 24 according to the present embodiment includes a first edge determination section 30, a second edge determination section 31, a white background area detection circuit 32, and a selector 33.
The selector 33 outputs the output signal E1 of the first edge determination unit 30 based on the output signal of the white background area detecting unit 32,
The output signal E2 of the second edge determination unit 31 is selected and output as separated data.

Next, the first edge determining section 30 will be described with reference to FIG. FIG. 4 is an explanatory diagram illustrating a schematic circuit configuration of the left edge determination unit. In the present embodiment, the image data from the scanner 2 is given as a so-called multi-level signal of 8 bits per pixel. The image data is input to the shift shift register including the first to fourth D flip-flops 40 to 43. The same clock signal is input to each of the D flip-flops forming the shift register. FIG.
In FIG. 1, the D flip-flop is conceptually shown by describing only one column for easy understanding, but since the image data is 8 bits, in an actual circuit, each bit of 8 bits is , There are 8 rows of D flip-flops.

In the example of FIG. 4, the second D flip-flop 4
The output Q2 of 1 indicates the data of the target pixel a (0) that is currently the target of determination as to whether or not it belongs to the edge region of the image. In the figure, a (1) indicates a pixel next to the target pixel, which is a pixel to be subjected to edge determination next to the target pixel. The data of this pixel is provided as the output Q4 of the first D flip-flop 40. The pixel a (-1) in the figure is also the pixel next to the pixel of interest, but this is the pixel of interest a
The pixel a (-2) in the figure, which has already been subjected to the edge determination before (0), is the pixel next to the pixel a (-1). The data of the a (−1) pixel is given as the output Q3 of the third D flip-flop 42, and the data of the a (−)
2) The pixel data is provided as the output Q4 of the fourth D flip-flop 43. By associating the D flip-flop with each pixel as described above, the right edge of the image is detected in the present embodiment.

A pixel of interest a (0) is added to the shift register.
When the pixels of a (1) to a (-2) including the above are set, the first calculator 44, which is a smoothing means, performs a calculation for smoothing the image data using the smoothing matrix. Be seen. The matrix used in the first edge determination unit 30 is shown in FIG. Now, let the data of the pixel of a (-2) be a, the data of pixel a (-1) be b, the data of pixel a (0) which is the target pixel be c, and the data of pixel a (1). Let d be the data b (0) after the pixel a (0), which is the pixel of interest, has been smoothed by the first computing unit 44 using the matrix shown in FIG. Required by doing.

[Formula 1] b (0) = (a × k (−2) + b × k (−1) + c × k
(0) + d × k (1)) / (k (-2) + k (-1) + k (0) + k (1))

The data of the target pixel after the smoothing process thus obtained is input to the first comparator 45. This first
The comparator 45 compares the input target pixel data after smoothing processing (density of the target pixel) with a predetermined threshold value, and when the data of the target pixel is larger than the threshold value,
"1" is output, and if it is less than or equal to the threshold, "0" is output. This threshold can be arbitrarily set in consideration of the characteristics of the printer.

On the other hand, the pixel a (-1) on both sides of the target pixel
And a (1) data is the second data as the density difference detecting means.
It is input to the calculator 46. Then, the density difference between the pixel a (-1) and the pixel a (1) is detected by the second calculator. That is, if the density difference is represented by n, the second calculator 46 calculates the density difference n by performing the calculation represented by the following equation.

## EQU00002 ## n = b-d The density difference n between the pixels on both sides of the target pixel thus obtained is input to the second comparator 47. This second comparator 4
7 compares the density difference n with a predetermined threshold value,
When the density difference n is larger than the above threshold value, "1" is output, and when the density difference n is less than or equal to the above threshold value, "0" is output.
Is output.

The output of the first comparator 45 and the output of the second comparator 47 are input to an AND circuit 48 as a judging means. The AND circuit 48 takes the logical product of the output of the first comparison circuit and the output of the second comparison circuit,
The result is output. That is, the first comparator 45
Is 1 and the output of the second comparator 47 is 1
"1" is output only when is, otherwise,
Outputs "0". (Hereinafter, margin)

Next, the second edge determining section 31 will be described. The second edge determination unit 31 in the present embodiment has substantially the same configuration as the first edge determination unit 30. The second edge determination unit differs from the first edge determination unit 30 in the second edge. The smoothing matrix used in the determining unit 31 is different from the smoothing matrix used in the first edge determining unit 30.

Specifically, the smoothing matrix used in the second edge determining section 31 is more efficient than the smoothing matrix used in the first edge determining section 30 in that the density of the pixel of interest is smoothed. Has become. this is,
For example, the coefficient of the portion of the smoothing matrix to be multiplied by the data of the pixel of interest may be smaller in the matrix used in the second edge determination unit 31 than in the matrix used in the first edge determination unit 30. Can be achieved by

In this way, by making the smoothing matrixes of the first edge determining section 30 and the second edge determining section 31 different, the first edge determining section 30 is better than the first edge determining section 30 when the same image data is input. Second edge determination unit 3
The probability of determining the pixel of interest as an edge is higher than that of 1. As a result, the first edge determination unit 3
The output signal E1 of 0 and the output signal E2 of the second edge determination unit 31 output different determination results for the same pixel of interest depending on the case.

The entire circuit configuration of the edge detection circuit 24 according to this embodiment is shown in FIG. In the illustrated example, the number of circuit elements is reduced by making the shift register formed by the D flip-flops common to the first edge determination section 30 and the second edge determination section 31. The outputs of the D flip-flops that form the shift register are similarly input to the first edge determination unit 30 and the second edge determination unit 31. As shown, the output signal E1 of the first edge determination unit 30 and the output signal E2 of the second edge determination unit 31 are
It is input to the selector 33 which is a switching means.

The selector 33 successively outputs the output signal E1 of the first edge determining section 30 and the output signal E2 of the second edge determining section 31 by the white background data which is an output signal of a white background area detecting device described later. I'm trying to switch. Specifically, when there is a white background area around the target pixel, the output signal E1 of the first edge determination unit 30 is selected as the edge separation data output from the edge area detection device, and the output signal E1 is output around the target pixel. When there is no white background area, the output signal E2 of the second edge determination unit 31 is selected and used as the edge separation data which is the output signal of the edge area detection device.

The white background area detecting means 32 will be described. The white background area detecting unit 32 synchronizes with the edge judgment of the target pixel in the first edge judging unit 30 and the second edge judging unit 31 based on the image data input to the edge region detecting unit 24, It is detected whether or not there is a white background area around the pixel of interest. Then, when there is a white background area around the pixel of interest, "1" is output as the white background data which is an output signal, and when there is no white background area, the white background data is "0".
Is output.

In this embodiment, the white area detection is 5 ×
Using the matrix of No. 5, a matching process with a pattern as shown in FIGS. 7A and 7B is performed. In FIG. 7A and FIG. 7B, “◯ (white circle)” indicates a white pixel, and an area in which 1 × 5 or 5 × 1 white pixels are continuous as shown is a white background area, It is detected whether or not this white background area exists. As the white background area detecting means, it is possible to use the white background area detecting means described in Japanese Patent Application Laid-Open No. 5-176167 already proposed by the applicant of the present application and cited in the section of the prior art.

Here, when comparing the character area and the picture area, it is empirically known that there is a high probability that a white background area exists around the pixels forming the character image, while the picture image is formed. It is empirically known that there is a low probability that a white background area exists around a pixel that has a pixel. Therefore,
As described above, when the white background area exists around the target pixel, the output signal E1 of the first edge determination unit 30 is selected,
By selecting the output signal E2 of the second edge determination unit 31 when there is no white background area around the pixel of interest,
As a result, in the edge separation data of the edge area detection device 24, the number of pixels determined as the edge area increases for the character area and the number of pixels determined as the edge area decreases for the picture area.

Then, based on the edge separation data thus obtained, in each circuit such as the γ correction circuit of the image processing unit 5 described above, the portion detected as the edge region is processed with priority on the resolution. For other parts, by performing processing with priority on gradation, processing with priority on resolution is performed on the character area of the image,
It is possible to perform a process that prioritizes gradation in the picture area. As described above, the first edge determination unit 30 in the present embodiment is positioned as a portion that performs edge region detection (edge separation) for the character region in the document, and the second edge determination unit 31 determines the pattern region in the document. That is, it is positioned as a portion for performing edge area detection (edge separation) on a photograph area and a halftone dot area other than the character area.

The operation of the selector 33 is shown in the flowchart of FIG. As shown, when the white background data from the white background area detecting means is set in the selector 33 (90
1), the selector 33 determines whether or not the white background data is 1 (902), and when the white background data is 1, the selector 33 determines the result of edge separation for the character original, that is, the output signal E1 of the first edge determination unit 30. If it is selected (903) and the white background data is 0, the result of edge separation for a photograph or halftone dot original, that is, the output signal E2 of the second edge determination unit 31 is selected (904).

Next, the edge area detecting device 24 according to the second embodiment will be described with reference to FIG. FIG. 9 is an explanatory diagram illustrating a schematic circuit configuration of the edge area detection device 24 according to the second embodiment. The edge area detection device 24 according to the second embodiment is almost the same as the edge area detection device 24 according to the first embodiment described above.

The edge area detecting device 24 according to the second embodiment is the edge area detecting device 2 according to the first embodiment.
4 is different from that of the first embodiment in the edge area detection device 24 according to the first embodiment.
The smoothing matrix used in the edge determination unit 31 is different only in its coefficient and the size (size) of the matrix is the same, but in the edge area detection device 24 according to the second embodiment, This is that the size of the smoothing matrix used by the 1-edge determining unit 30 and the size of the smoothing matrix used by the second edge determining unit 31 are different. Also, due to this difference, the configuration of the shift register configured by the D flip-flop is different.

As described above, the reason why the size of the matrix used in the first edge determination unit 30 and the size of the matrix in the second edge determination unit 31 are different is that the same image data is used, as in the first embodiment. This is to make the determination result of the first edge determination unit 30 and the determination result of the second edge determination unit different from each other. That is, in the first embodiment,
In the second embodiment, what is achieved by making the coefficients of the smoothing matrix different is intended to be made by making the size of the matrix different.

In the present embodiment, the smoothing matrix used in the first edge determination section is 1 (line) × 4.
In contrast to the (pixel) matrix, the smoothing matrix used in the second edge determination unit 31 is 2 (lines).
A matrix of × 3 (pixels) is used. In this way, the second edge determination unit 31 uses a larger smoothing matrix than the first edge determination unit because the larger the smoothing matrix, the smoother the target pixel can be. is there.

In order to perform the image data smoothing processing on the smoothing matrix, the first to fourth D flip-flops 40 are used in the edge area detecting apparatus of the second embodiment.
~ 43 to the fifth to eighth D flip-flops 50
.About.53 are used to configure a shift register for two lines. The image data input to the shift register is input to the input D1 of the first flip-flop 40 as it is, and also to the input D5 of the fifth flip-flop 50 via the FIFO (first-in first-out memory) 54. The data of the target pixel a (0) is the output Q2 of the second flip-flop 41 as in the first embodiment.

Then, the first arithmetic unit 4 of the first edge determination unit
The outputs Q1 to Q4 of the first to fourth flip-flops 40 to 43 are input to 4a, and the smoothing process using the above 1 × 4 smoothing matrix is executed. On the other hand, the first arithmetic unit 44b of the second edge determination unit 31 has the outputs Q1 to Q3 of the first to third D flip-flops 40 to 42 and the fifth to fifth outputs.
The outputs Q5 to Q7 of the seventh flip-flops 50 to 52 are input, and the smoothing process using the 2 × 3 smoothing matrix is executed.

The second computing units 46a and 46b of the first edge determining unit 30 and the second edge determining unit 31 and the first comparator 45.
Since the constituent operations of a and 45b, the second comparators 47a and 47b, the AND circuits 48a and 48b, the sector 33, and the white background area detecting means 32 are the same as those in the first embodiment, detailed description thereof will be omitted. .

The first embodiment and the second embodiment described above
In the above embodiment, only the right edge of the image is detected. In addition to the right edge, it is also possible to detect the left edge. To this end, in addition to the set of the first edge determination unit and the second edge determination unit in the above-described first embodiment or second embodiment, another set having a configuration substantially similar to the set is provided. There is a need.

FIG. 10 is a block diagram of a configuration of an edge detecting device for detecting both left and right edges. First
The edge determination unit 30 and the second edge determination unit 31 each have a left edge determination unit 34 and a right edge determination unit. Then, as shown in FIG. 10, in both the first and second edge determination sections, the logical sum of the output signal of the left edge determination section and the output signal of the right edge determination section is ORed to obtain the respective output signal E1 or E2. Output.

Here, the right edge determining unit 35 of the first edge determining unit 30 in FIG. 10 corresponds to the first edge determining unit in the above-described first embodiment or second embodiment. The right edge determination unit of the second edge determination unit 31 in FIG. 10 corresponds to the second edge determination unit in the above-described first embodiment or second embodiment. Therefore,
As the right edge determination unit, the determination unit described with reference to FIG. 4 may be used as it is.

Regarding the configuration example of the left edge determination section,
A description will be given based on FIG. FIG. 11 is an explanatory diagram illustrating a schematic circuit configuration of the left edge determination unit. The left edge determination unit 34 has almost the same configuration as the right edge determination unit (first edge determination unit in the first embodiment) described above.
The configuration is the same as that described above. In view of this, in FIG.
Elements having the same functions as the circuit constituent elements of the right edge determination section shown in FIG. 4 are designated by the same reference numerals. The difference between the left edge determination unit 34 and the right edge determination unit 35 is that 4 including the target pixel a (0) in the shift register configured by the D flip-flop described above.
The difference is in the way the two pixels are set, and the smoothing matrix used in the first calculator is correspondingly different.

From the above differences, the left edge determination unit 3
The first arithmetic unit 44 of No. 4 is different from the right edge determination unit 35,
The data b (0) after the smoothing processing of the target pixel a (0) is obtained by an operation such as the following equation. The smoothing matrix used in the calculation of the following equation is shown in FIG.

## EQU00003 ## b (0) = (a * k (-1) + b * k (0) + c * k
(1) + d × k (2)) / (k (-1) + k (0) + k (1) + k (2)) where a is a (-1) pixel data and b is a
(0) pixel and target pixel data, c represents a (1) pixel data, and d represents a (2) pixel data.

Further, the detection of the density difference n in the second computing unit 46 of the left edge determination unit is obtained by the calculation of the following equation.

## EQU00004 ## n = a-c The data (b) of the target pixel after the smoothing process thus obtained
(0)) and the density difference (n) on both sides of the pixel of interest,
Since the processes performed by the first comparator 45, the second comparator 47, and the AND circuit 48 are the same as the processes of the left edge determination unit, detailed description thereof will be omitted.

[0050]

According to the edge area detecting device of the first to fourth aspects, since it is not necessary to perform the pattern matching for detecting the edge area of the image, a storage means for storing a large number of matching patterns, Does not require a large number of line memories required for pattern matching,
The edge region detection device has a simple structure as compared with the conventional device, and it is possible to obtain a low cost edge region detection device. Further, since the processing time for pattern matching is eliminated, there is an excellent effect that the processing speed for detecting an edge region of an image can be improved.

In particular, according to the edge area detecting device of the second aspect, the probability of determining the target pixel as a pixel belonging to the edge area is increased in the character area, and the target pixel is regarded as a pixel belonging to the edge area in the picture area. Since the determination probability can be lowered, for example, when image processing such as gamma correction is performed using the output result of this edge area detection device, the resolution is prioritized for the portion determined to be the edge area. For the part that is not determined to be an edge region, the process that prioritizes gradation is performed with a simpler configuration than the conventional image area separation device, and as a result, On the other hand, it is possible to perform the processing in which the resolution is prioritized, and it is possible to perform the processing in which the gradation is prioritized in the picture area of the image.

Further, according to the edge area detecting device of the third aspect, when the white background area detection result is that there is no white background area, it is determined that the pixel of interest belongs to the edge area rather than when there is a white background area. Since the probability can be lowered, it is possible to increase the probability of determining the pixel of interest as a pixel belonging to the edge region in the character region and decrease the probability of determining the pixel of interest as a pixel belonging to the edge region in the pattern region. it can. Thus, for example, when image processing such as gamma correction is performed by using the output result of the edge area detection device, the portion determined to be the edge area is processed with priority on the resolution and the edge area is detected. For the part that is not judged, the process that prioritizes the gradation property enables the process that prioritizes the resolution for the character area of the image with a simpler configuration than the conventional image area separation device. In the picture area of the image, it is possible to perform the processing in which the gradation is prioritized.

Further, in particular, according to the edge area detecting apparatus of the fourth aspect, the sizes of the plurality of types of smoothing matrices corresponding to the detection result of the white background area are made the same.
Compared with the edge area detection device of (1), it has an excellent effect of not wasting circuit elements. Further, when the white background area detection result is no white background area, the probability that the target pixel is determined to belong to the edge area can be made lower than when the white background area is present. It is possible to increase the probability that the pixel is determined to belong to the pixel, and decrease the probability that the pixel of interest is determined to belong to the edge region in the picture area. Thus, for example, when image processing such as gamma correction is performed by using the output result of the edge area detection device, the portion determined to be the edge area is processed with priority on the resolution and the edge area is detected. For the part that is not determined, the processing that prioritizes the gradation is performed, and thus the processing that prioritizes the resolution is performed for the character area of the image with a simpler configuration than the conventional image area separation device. This makes it possible to perform the processing in which the gradation is prioritized in the picture area of the image.

[Brief description of drawings]

FIG. 1 is a schematic configuration diagram of a digital copying machine according to an embodiment.

FIG. 2 is a configuration block diagram of a digital copying machine according to an embodiment.

FIG. 3 is an explanatory diagram illustrating a schematic configuration of an edge area detecting unit 24 according to the embodiment.

FIG. 4 is an explanatory diagram illustrating a schematic circuit configuration of a first edge determination unit according to the first embodiment.

FIG. 5 is a diagram illustrating a smoothing matrix used in a first edge determination unit.

FIG. 6 is an explanatory diagram illustrating a schematic circuit configuration of the edge area detection device according to the first embodiment.

7A and 7B are explanatory diagrams illustrating a matching pattern used in a white background area detecting unit.

FIG. 8 is a flowchart illustrating the operation of the selector 33.

FIG. 9 is an edge area detection device 2 according to a second embodiment.
4 is an explanatory diagram illustrating a schematic circuit configuration of FIG.

FIG. 10 is a configuration block diagram of an edge detection device in which a left edge determination unit is added to each of a first edge determination unit and a second edge determination unit.

FIG. 11 is an explanatory diagram illustrating a schematic circuit configuration of a left edge determination unit.

FIG. 12 is an explanatory diagram illustrating a smoothing matrix used in a left edge determination unit.

[Explanation of Codes] 1 Copier 2 Scanner 3 Printer 4 Paper Feeding Unit 5 Image Processing Unit 6 Writing Unit 7 Photoconductor 8 Charging Device 9 Developing Device 10 Transfer Device 11 Cleaning Device 12 Conveyor Belt 13 First Mirror 14 Second Mirror 15 Third mirror 16 CCD 17 Contact glass 18 Exposure lens 19 Exposure lamp 20 γ correction circuit 21 Filter circuit 22 Magnification circuit 23 Pseudo intermediate processing circuit 24 Edge area detection device 30 First edge determination unit 31 Second edge determination unit 32 White background Region detection circuit 33 Selector 34, 36 Left edge determination unit 35, 37 Right edge determination unit 38, 39 OR circuit 40 First D flip-flop 41 Second D flip-flop 42 Third D flip-flop 43 Fourth D flip-flop 44a, 44b First operation Vessel 45a, 5b first comparator 46a, 46b second computing unit 47a, 47b second comparator 48a, 48b and circuit 50 first 5D flip-flop 51. 6D flip-flop 52. 7D flip-flop 53 first 8D flip-flop 54 FIFO

Claims (4)

[Claims] 1. An edge area detection device for detecting an edge area of an image represented by the image data based on the input image data, wherein a smoothing matrix is used for a target pixel of the input image data. Smoothing means for performing smoothing processing; density difference detecting means for detecting a density difference between two pixels existing on both sides of the target pixel based on the input image data; and the density of the target pixel after the smoothing and the density An edge area detection device, comprising: a determination means for determining whether or not a pixel of interest is a pixel in an edge area of an image based on the detection result of the density detection means and outputting the determination result. 2. The edge area detecting device according to claim 1, wherein a white background area detecting means for detecting a white background area existing around a target pixel, and a plurality of types of smoothing matrices corresponding to the detection result of the white background area detecting means. An edge area detecting device, comprising: a selecting unit that selects a determination result obtained by using the above-mentioned item based on the detection result of the white background region detecting unit. 3. The edge area detecting device according to claim 2, wherein the plurality of types of smoothing matrices corresponding to the detection result of the white background area detecting means are larger than the matrix size corresponding to the case where the detection result is a white background area. Also, the edge area detection device is characterized in that the matrix size corresponding to the absence of the white background area is larger. 4. The edge area detecting device according to claim 2, wherein the plurality of types of smoothing matrices corresponding to the detection result of the white background area detecting means have the same size, and among the coefficients of the matrix, An edge area detecting device characterized in that a coefficient corresponding to a pixel of interest is made larger in a matrix corresponding to a white background area than to a smoothing matrix corresponding to a case where the above detection result indicates that there is no white background area.