JP4154867B2 - Image area discrimination apparatus and image area discrimination program - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to an image area determination apparatus and an image area determination program for extracting a specific image area from another image area from an image in which characters, photographs, and halftone dots are mixed. More specifically, the present invention relates to an image area discriminating apparatus and an image area discriminating program capable of discriminating a character area on a halftone dot with high accuracy.
[0002]
[Prior art]
In a digital copying machine or the like that reads an image using a CCD sensor and decomposes the image into pixels to perform image processing, image attributes such as a character area, a photographic area, and a halftone dot area are determined.
Then, the image processing method for the attribute of each image is switched according to the determination result. This is because the optimum image processing method differs for each image attribute.
[0003]
Here, as one of character area discrimination techniques, for example, there is one disclosed in Japanese Patent Laid-Open No. 5-344329. In this technique, first, an input image is divided into a predetermined matrix size, and an average density is calculated within the matrix of the predetermined size. Next, assuming a matrix of a predetermined size as one pixel, a density difference between the target pixel and its surrounding pixels is calculated. If the density difference is equal to or greater than a certain value, it is determined that the target pixel belongs to the edge region. Then, a region where a predetermined number or more of pixels belonging to the edge region continues is determined as a character region. By discriminating the character area in this way, the character area and the photographic area can be distinguished.
[0004]
Also, as a representative technique for discriminating a halftone dot region, for example, focusing on isolated points (maximum value points and minimum value points) that are characteristic of the halftone dot region, an isolated point existing in a predetermined region is considered. There are some which count the number and determine the halftone dot region according to the count value of the isolated points. The character area on the halftone dot is discriminated based on the detection result by detecting an edge or an isolated point.
[0005]
[Problems to be solved by the invention]
However, the above-described conventional image area discrimination technique has a problem in that it cannot accurately discriminate a character area on a halftone dot. The reason is as follows. That is, an isolated point that constitutes a halftone dot is determined to be an edge, but when a character is small, the portion is often determined to be an isolated point. In the character region on the halftone dot, the isolated point and the edge are adjacent to each other. For this reason, it is difficult to separate the character area from the halftone dot area.
[0006]
If the character area on the halftone dot is erroneously determined to be a halftone dot area, the character on the halftone dot is blurred in the reproduced image, and thus a high-quality reproduced image cannot be obtained. This is because the smoothing process is performed on the halftone dot area, and the smoothing process is similarly performed on the character area on the halftone dot area. In other words, even though it is necessary to perform edge enhancement processing on the character area, smoothing processing with emphasis on gradation is performed to prevent moiré, so the character image on the halftone dot is blurred. It ends up.
[0007]
Therefore, the present invention has been made to solve the above-described problems, and it is an object of the present invention to provide an image area determination apparatus and an image area determination program that can determine a character area on a halftone dot with high accuracy. And
[0008]
[Means for Solving the Problems]
An image area discriminating apparatus according to the present invention, which has been made to solve the above problems, includes a first density calculating means for calculating an average density in a first area including a target pixel, and a background density in a second area including the target pixel. The second density calculation means for calculating the first density calculation means, and the calculation result of the first density calculation means according to the background density calculated by the second density calculation means and the background density. Positive or negative A character area determining means for determining whether or not the image belongs to a character area by comparing with a threshold value calculated based on a set offset amount; and based on a determination result of the character area determining means, Discriminating means for discriminating whether or not the character area belongs to.
The determining unit may determine that the target pixel belongs to the character region on the halftone dot when the average density of the target pixel calculated by the first density calculating unit is lower than the threshold value.
[0009]
In this image area discriminating apparatus, first, the average density in the first area including the target pixel is calculated by the first density calculating means. Further, the background density in the second region including the target pixel is calculated by the second density calculation unit. The concentration calculation by the first concentration calculation means and the second concentration calculation means may be performed by a known method. Further, in the first density calculation unit and the second density calculation unit, such data may be used instead of the density value as long as the data has a correlation with the density (for example, brightness).
[0010]
Subsequently, when the average density and the background density are calculated by the first density calculating unit and the second density calculating unit, the calculation result of the first density calculating unit is obtained by the second density calculating unit. Depending on the calculated background density and the background density Positive or negative It is compared with a threshold value calculated based on the set offset amount. Then, the calculation result of the first density calculation means is determined according to the background density calculated by the second density calculation means and the background density. Positive or negative The character area is determined by comparing with a threshold value calculated based on the set offset amount.
[0011]
Therefore, by performing this process only within the halftone dot region, the character region on the halftone dot can be determined. Alternatively, this processing is performed on the entire area of the input image so that it can be determined that the area is a character area, and the determined area that is a halftone dot area is a character area on a halftone dot. . In view of this, in this image area discrimination apparatus, the discrimination means discriminates whether or not the target pixel belongs to the character area on the halftone dot based on the comparison result of the density comparison means. Specifically, when the average density of the target pixel calculated by the first density calculation unit is lower than the threshold, the determination unit determines that the target pixel belongs to the character region on the halftone dot. As a result, the character area on the halftone dot can be accurately identified.
[0012]
Here, in the image region discrimination device according to the present invention, when the first density calculation unit and the second density calculation unit are calculated, the second region is preferably larger than the first region. This is because if the second area for calculating the background density is smaller than the first area for calculating the average density, the character area discrimination accuracy is lowered. In other words, it is possible to improve the discrimination accuracy of the character area on the halftone dot.
[0013]
In the image region determination device according to the present invention, the offset amount is set to a negative value when the background density calculated by the second density calculation unit is lower than a predetermined density, and the second density When the background density calculated by the calculation means is higher than the predetermined density, it is set to a positive value.
Thereby, the offset amount is set to a negative value in the case of a white background, and is set to a positive value in the case of a black background or a high-density chromatic color background. Therefore, since the offset amount changes according to the background density, the character area on the halftone dot on the black background or the high-density chromatic color background can be determined with high accuracy.
[0014]
Further, in the image region determination device according to the present invention, the dot region determination means for determining whether or not the target pixel belongs to the halftone dot region, and the target pixel is the edge region based on the density difference between the target pixel and the surrounding pixels. Edge region determining means for determining whether or not the image belongs to the image area, wherein the determining means is determined to be a halftone dot area by the halftone dot area determining means, and is determined by the edge area determining means in the edge region. It is desirable to determine that the pixel of interest belongs to a character edge region on a halftone dot when it is determined that the pixel is present and belongs to the character region by the character region determination unit.
[0015]
This image area discriminating apparatus further includes halftone dot area discriminating means and edge area discriminating means. Then, the halftone dot region discriminating unit determines whether or not the target pixel belongs to the halftone dot region. Note that the halftone dot region is determined by focusing on isolated points (maximum value points and minimum value points), counting the number of isolated points existing in a predetermined region, and performing according to the count value of the isolated points. Good. However, the determination of the dot area is not limited to this, and may be performed by other methods. Further, the edge region determination means determines whether or not the target pixel belongs to the edge region based on the density difference between the target pixel and the peripheral pixels. Then, the determination unit further considers the determination result of the halftone dot region determination unit and the determination result of the edge region determination unit, and determines whether or not the target pixel belongs to the character region on the halftone dot. Specifically, when it is determined as a halftone dot region by the halftone dot region determination unit, is determined as an edge region by the edge region determination unit, and is determined to belong to the character region by the character region determination unit In addition, it is determined that the target pixel belongs to the character region on the halftone dot. As a result, the character area on the halftone dot can be determined with higher accuracy.
[0016]
Here, the above-described image area determination device is not limited to being configured by hardware, and can also be realized by software. Therefore, as described below, it can be realized as an image area determination method or may exist as an image area determination program.
[0017]
That is, as an example of the image area determination method, the image area determination method according to the present invention includes a first density calculating step for calculating an average density in the first area including the target pixel, and a background in the second area including the target pixel. The second density calculation step for calculating the density and the calculation result of the first density calculation step according to the background density calculated by the second density calculation step and the background density. Positive or negative A character area determining step for determining whether the image belongs to a character area by comparing with a threshold value calculated based on a set offset amount, and based on the determination result of the character area determining step, the target pixel is And a determination step of determining whether or not it belongs to the character area.
In the image region determination method according to the present invention, the halftone dot region determination step for determining whether or not the target pixel belongs to the halftone dot region, and the target pixel based on the density difference between the target pixel and the peripheral pixel are the edge regions. An edge region determining step for determining whether or not the image belongs to the edge region. In the determining step, the halftone dot region determining step determines that the region is a halftone dot region, and the edge region determining step determines whether the edge region is an edge region. If it is determined that the pixel belongs to the character region by the character region determination step, it may be determined that the target pixel belongs to the character edge region on the halftone dot.
[0018]
Alternatively, as an example of the image area determination program, the image area determination program according to the present invention causes the computer to calculate a first density for calculating an average density in the first area including the target pixel, and to perform a second step including the target pixel. The second density calculation step for calculating the background density in the region and the calculation result of the first density calculation step are determined according to the background density calculated by the second density calculation step and the background density. Positive or negative A character area determining step for determining whether the image belongs to a character area by comparing with a threshold value calculated based on a set offset amount, and based on the determination result of the character area determining step, the target pixel is And a determination step of determining whether or not it belongs to the character area.
In the image region determination program according to the present invention, the halftone dot region determination step for determining whether or not the target pixel belongs to the halftone dot region, and the target pixel based on the density difference between the target pixel and the peripheral pixel are the edge region. An edge region determination step for determining whether or not the image belongs to a computer, and in the determination step, the halftone dot region determination step determines that the region is a halftone dot region, and the edge region determination step When it is determined that the pixel is an edge region and the character region is determined to belong to the character region by the character region determining step, the pixel of interest may be determined to belong to the character edge region on the halftone dot. .
[0019]
The above-described image area determination device can be realized by operating the computer by these image area determination programs, that is, by reading the software into the computer. The same effects as those of the above-described image area determination device can be obtained.
[0020]
DETAILED DESCRIPTION OF THE INVENTION
DESCRIPTION OF EXEMPLARY EMBODIMENTS Hereinafter, preferred embodiments of the invention will be described in detail with reference to the drawings. In the present embodiment, a case where the present invention is applied to an image processing apparatus will be described. First, FIG. 1 shows a schematic configuration of an image processing apparatus according to the present embodiment. The image processing apparatus 10 includes a color correction unit 11, an area determination unit 12, and an MTF correction unit 13. The operation of the image processing apparatus 10 is comprehensively controlled by the CPU 14. The CPU 14 is provided with a ROM 15 in which various programs for controlling the operation of the image processing apparatus 10 are stored.
[0021]
An image input unit 16 for acquiring image data is connected to such an image processing apparatus 10. As a result, image data is input from the image input unit 16 to the image processing apparatus 10. Here, the image input unit 16 receives reflected light obtained by scanning a document with a scanner with a CCD sensor, and photoelectrically converts it to obtain analog image data. The analog image data is converted into digital image data and then transmitted to the image processing apparatus 10. The image processing apparatus 10 is connected to a printer 17 that outputs an image subjected to image processing onto a recording medium such as paper.
[0022]
Here, the color correction unit 11 converts the input image data into a recording density signal CMYK data suitable for color reproducibility in consideration of the spectral characteristics and recording process of the four colors of toner in the printer 17. is there. The area discriminating unit 12 discriminates image attributes such as a halftone dot area and an edge area. Details thereof will be described later. The MTF correction unit 13 corrects the sharpness of the image. Specifically, the MTF correction unit 13 performs smoothing processing and edge enhancement processing on the image data output from the color correction unit 11 according to the determination result of the region determination unit 12. An image is output from the printer 17 based on the image data corrected by the MTF correction unit 13.
[0023]
Next, the details of the region discriminating unit 12 which is a characteristic part of the present invention are shown in FIG. 2, and the configuration of the region discriminating unit 12 will be described using this. The region discriminating unit 12 discriminates whether each pixel in the input image belongs to a halftone dot region, an edge region, or a character edge region within a halftone dot. In order to make such a determination, the region determination unit 12 includes a lightness data creation unit 20, a halftone dot determination unit 21, a background level determination unit 22, an average lightness calculation unit 23, an edge determination unit 24, A threshold calculation unit 25, a comparison unit 27, and an AND circuit 30 are provided.
[0024]
The lightness data creating unit 20 creates lightness data (256 gradations in the present embodiment) necessary for region discrimination based on the image data transmitted from the image input unit 16. The lightness data created by the lightness data creating unit 20 is input to each of the halftone dot discriminating unit 21, the background level discriminating unit 22, the average lightness calculating unit 23, and the edge discriminating unit 24.
[0025]
Then, the halftone dot determination unit 21 detects an isolated point from the brightness data, and determines whether or not the target pixel belongs to the halftone dot region based on the detection result. As a specific method for determining the halftone dot region, for example, a method disclosed in Japanese Patent Laid-Open No. 2000-287077 may be used. Of course, the dot area may be determined using other methods. The halftone dot area signal (_AMI) output from the halftone dot determination unit 21 is input to the AND circuit 30 and the MTF correction unit 13.
[0026]
The background level determination unit 22 creates a brightness histogram in a predetermined area from the brightness data, and calculates the background level. In the present embodiment, as shown in FIG. 3, a brightness histogram is created in a 256 × 256 size area to determine the background level. Here, an example of the brightness histogram created by the background level discriminating unit 22 is shown in FIGS. FIG. 4 shows a case of a white background, and FIG. 5 shows a case of a black background or a high-density chromatic color background. Then, as shown in FIGS. 4 and 5, the lightness having the highest frequency in the histogram is calculated as the background level. Note that th1 in FIGS. 4 and 5 represents a threshold value to be described later.
[0027]
Returning to FIG. 2, the background level calculated by the background level determination unit 22 is input to the threshold value calculation unit 25. The threshold value calculation unit 25 calculates a threshold value in the comparison unit 27. The threshold value in the comparison unit 27 is calculated based on the background level calculated by the background level determination unit 22 and a predetermined offset amount set according to the level. Specifically, the comparison unit 27 calculates the threshold value th1 by adding the offset amount to the background level. Here, the offset amount is set to a negative value when it is larger than a lightness threshold (not shown) (for example, “128” or the like), and is set to a positive value when it is smaller. Specifically, as shown in FIG. 4, a negative value is set for a white background, and a positive value is set for a black background or a high-density chromatic color background as shown in FIG. Is done.
[0028]
The average brightness calculation unit 23 calculates the average brightness around the pixel of interest V33 from the brightness data created by the brightness data creation unit 20, using a 5 × 5 size averaging filter shown in FIG. Yes. That is, the average brightness of the pixels V11 to V55 is calculated, and the brightness data is used as the average brightness data of the target pixel V33. The average brightness data for each pixel calculated by the average brightness calculation unit 23 is input to the comparison unit 27.
[0029]
As a result, the threshold value and the average brightness data are input to the comparison unit 27, respectively. Then, the comparison unit 27 compares the average brightness data with the threshold th1. In each comparison unit 27, when the threshold th1 (A) is larger than the average brightness data (B) (A> B), the signal output from the comparison unit 27 is activated (that is, “1”). .
[0030]
The signal output from the comparison unit 27 is output as a character area signal (_MOJI). The character area signal (_MOJI) output from the comparison unit 27 is input to the AND circuit 30.
[0031]
In addition, the edge determination unit 24 determines whether or not the target pixel belongs to the edge region using a differential filter or the like. Specifically, when the brightness difference between the target pixel and the surrounding pixels is a certain value or more, it is determined that the target pixel belongs to the edge region. The edge region signal (_EDG) output from the edge determination unit 24 is input to each of the AND circuit 30 and the MTF correction unit 13.
[0032]
Thus, the halftone dot area signal (_AMI), the character area signal (_MOJI), and the edge area signal (_EDG) are input to the AND circuit 30. The AND circuit 30 calculates the logical product of these signals. This logical product is output as a halftone dot character edge region signal (_EDGL) and input to the MTF correction unit 13.
[0033]
With such a configuration, the region discriminating unit 12 discriminates the image attribute of the input pixel. Specifically, the area discrimination unit 12 outputs a halftone dot area signal (_AMI), a halftone dot character edge area signal (_EDGL), and an edge area signal (_EDG), which are input to the MTF correction unit 13. . Then, as described above, the MTF correction unit 13 performs smoothing processing or edge enhancement processing based on these region determination signals. Specifically, the smoothing process is performed on the pixels in which the dot area signal (_AMI) is active. On the other hand, edge emphasis processing is performed on pixels in which either the halftone dot character edge region signal (_EDGL) or the edge region signal (_EDG) is active.
[0034]
Next, the operation of the image processing apparatus 10 having the above configuration will be described. First, the image information of the document is read by the image input unit 16. Then, the image data read by the image input unit 16 is transmitted to the image processing apparatus 10. Then, the image processing apparatus 10 performs various image processing on the image data. Then, based on the image data that has been subjected to image processing, the printer 17 outputs a reproduced image of the document on the recording medium.
[0035]
Next, the operation of the area determination unit 12 that is a characteristic part of the present invention in the image processing will be described. Here, a description will be given by taking a specific example of region discrimination of an image in which characters are drawn on a halftone dot shown in FIG. FIG. 7B shows a state of each pixel by enlarging a part of the image shown in FIG. A shaded pixel VA indicates an isolated point of the halftone dot, and a cross-hatched pixel VC indicates a target pixel when the filter processing is performed by the logic filter F. This logical filter F is scanned in the raster direction. In the following description, the processing in the region will be described by paying attention to a part of the image shown in FIG. 7A (FIG. 7B). It goes without saying that the same processing is performed in other regions.
[0036]
First, image data of the image shown in FIG. 7A is input to the brightness data creation unit 20. Then, the brightness data creation unit 20 creates brightness data (256 gradations) for each pixel. The result is shown in FIG. The character area has 0 gradation, the character edge area has 10 gradation, and the other areas have 255 gradation. The isolated point of the halftone dot has 10 gradations. In FIG. 8, since a 5 × 5 filter is used as an averaging filter, which will be described later, only two pixels are shown outside. Based on the lightness data, various processes are performed in parallel in the halftone dot determination unit 21, the background level determination unit 22, the average lightness calculation unit 23, and the edge determination unit 24.
[0037]
The halftone dot discrimination unit 21 discriminates a halftone dot region. Since the entire region of the image shown in FIG. 7A is a halftone dot region, a halftone dot region signal (_AMI) is obtained for all pixels in this region. Is activated. Further, in the background level discriminating unit 22, since this image is a white background as apparent from the lightness data shown in FIG. 8, a lightness histogram as shown in FIG. 4 is created. As a result, the ground level was calculated as “173.9”. Then, the threshold value calculation unit 25 calculates the threshold value th1. Specifically, in the case of the image shown in FIG. 8, the offset amount for calculating the threshold value is set to “−65”. Note that the offset amount varies depending on the value of the ground level, not the fixed value. Therefore, the threshold value calculation unit 25 calculates the threshold value th1 as “108.9”.
[0038]
The average brightness calculation unit 23 calculates average brightness data for each pixel by scanning the 5 × 5 size averaging filter shown in FIG. 6 in the raster direction. The result is shown in FIG. It can be seen that the average brightness of the character area is lower than the other areas. This calculation result is input to the comparison unit 27. Thus, the threshold value th1 and the average brightness data are input to the comparison unit 27.
[0039]
Then, the comparison unit 27 compares the threshold th1 with the average brightness data shown in FIG. 9 for each pixel. The comparison result is shown in FIG. It can be seen that the isolated points of the halftone dots are not extracted and only the character region is extracted. Further, the edge discriminating unit 24 discriminates the edge region by a known method. The discrimination result is shown in FIG. It can be seen that isolated points of the character edge region and the halftone dot are extracted.
[0040]
In the AND circuit 30, the halftone dot area signal (_AMI) output from the halftone dot determination section 21, the character area signal (_MOJI) output from the comparison section 27, and the edge area output from the edge determination section 24. A logical product with the signal (_EDG) is calculated. The calculation result is shown in FIG. It can be seen that only the character edge region on the halftone dot is extracted. Even if the comparison unit 27 extracts pixels that are not character areas, the AND circuit 30 can accurately extract only the character edge areas on the halftone dots. Therefore, the character area on the halftone dot can be determined with high accuracy. This is output from the AND circuit 30 as a halftone dot character edge region signal (_EDGL).
[0041]
In this way, when the area determination unit 12 performs the determination process on the image shown in FIG. 13, a determination result as shown in FIG. 14 is obtained. In FIG. 14, the cross-hatched part has been determined as the character edge area in the halftone dot. Note that the image shown in FIG. 13 is the same as the image shown in FIG. 13 and FIG. 14, it is clear that the boundary pixel VE of the character image drawn on the halftone dot is accurately determined to belong to the character edge region in the halftone dot. That is, the character region on the halftone dot is determined with high accuracy by the region determining unit 12.
[0042]
Thereafter, each region signal output as a determination result in the region determination unit 12 is input to the MTF correction unit 13, where edge enhancement processing or smoothing processing is performed on each pixel according to the state of each region signal. Specifically, edge emphasis processing is performed on pixels in which the halftone dot character edge signal (_EDGL) or edge region signal (_EDG) is active, and smoothing processing is performed on other pixels. As a result, the image output from the printer 17 has very good quality because the occurrence of moire is suppressed in the halftone area and the characters on the halftone are reproduced clearly.
[0043]
Here, in the above description, processing for an image in which normal characters are drawn on a white background has been described as an example. However, the area determination unit 12 uses a black background or a high-density chromatic background as shown in FIG. The character area on the halftone dot can be similarly determined. Note that the image shown in FIG. 15 is obtained by changing the image shown in FIG. 13 into white characters.
[0044]
When such processing is applied to such an image, the result shown in FIG. 16 is output from the comparison unit 27, and the result shown in FIG. 17 is output from the edge determination unit 24. Accordingly, the AND circuit 30 outputs the result shown in FIG.
[0045]
That is, when the discrimination process is performed on the image shown in FIG. 15, the discrimination result shown in FIG. 19 is obtained. As described above, it is determined that the outer boundary pixel VEO existing by one pixel outside the boundary pixel of the outline character image drawn on the halftone dot belongs to the character edge region in the halftone dot. Therefore, the area determination unit 12 can determine the character area on the halftone dot with high accuracy. However, as shown in FIG. 19, it is not determined that the boundary pixel of the outline character image belongs to the halftone dot character edge region, but the outside boundary pixel VEO of the outline character image belongs to the halftone dot character edge region. Therefore, the characters tend to be slightly fattened in the reproduced image.
[0046]
Further, in the above description, the case where the image area determination device according to the present invention is configured by hardware is exemplified, but such a device can also be realized by software. Therefore, the flow of processing when the above-described image processing apparatus is realized by software will be described with reference to the flowcharts shown in FIGS.
[0047]
First, software is read into a computer (for example, a copier or a printer). When the reading is completed, the computer and the software of the present invention cooperate to execute the following processing. That is, the computer first waits for input of image data (# 1). When the image data is input (# 1: YES), the computer creates brightness data for each pixel (# 2). When the brightness data is created, the computer determines pixels belonging to the edge region based on the brightness data (# 3). Then, the process of # 4 is executed for the pixels determined to belong to the edge region (# 3: YES). On the other hand, a smoothing process is performed on pixels that are determined not to belong to the edge region (# 3: NO, # 16).
[0048]
Next, in # 4, the computer determines pixels belonging to the dot area. Then, the process of # 5 is executed for the pixels determined to belong to the halftone dot area (# 4: YES). On the other hand, for the pixels determined not to belong to the halftone dot region, it is further determined whether or not they belong to the photo region (# 4: NO, # 13). At this time, the computer performs a smoothing process on the pixels determined to belong to the photographic area (# 13: YES, # 16). On the other hand, edge enhancement processing is performed on pixels that are determined not to belong to the photo area (# 13: NO, # 12).
[0049]
Subsequently, in # 6, the computer creates a brightness histogram in a 256 × 256 matrix (see FIGS. 4 and 5). Then, the background level (E) is extracted from the created brightness histogram (# 6). Further, the computer calculates the average brightness (Ave) in the 5 × 5 matrix (# 7) and determines whether the background is white (# 8). Specifically, when the background level (E) is larger than a predetermined brightness threshold value (in the case of high brightness side), it is determined that the background is a white background. On the other hand, if it is small (on the low brightness side), it is determined that the background is a black background (high density chromatic color background).
[0050]
Then, the computer sets the threshold th1 after determining the state of the background (# 9-1, # 9-2). In other words, if the computer determines that the background is white (# 8: YES), the threshold value th1 is calculated by setting the offset amount to a negative value (# 9-1). Conversely, if it is determined that the background is a black background (high density chromatic color background) (# 8: NO), the offset amount is set to a positive value and the threshold th1 is calculated (# 9-2). The offset amount is a variable set based on the background level (E).
[0051]
When the threshold value th1 is set, the computer compares the threshold value th1 with the average brightness (Ave) in # 10. At this time, if the threshold th1 is larger than the average brightness (E) (# 10: YES), it is determined that the pixel belongs to the character area on the halftone dot (# 11). Then, edge enhancement processing is performed on the pixels determined to belong to the character area on the halftone dot (# 12). On the other hand, if the threshold value th1 is smaller than the average brightness (E) (# 10: NO), it is determined that the pixel belongs to a halftone dot region (# 15). Then, smoothing processing is performed on the pixels determined to belong to the halftone dot region (# 16).
[0052]
By loading software that performs the processing of such a procedure into an image processing apparatus (corresponding to a computer) that does not have the hardware described above, it is possible to obtain the same effect as the above-described image processing apparatus. This software may exist as a program or may exist as a recording medium on which the program is recorded.
[0053]
As described above in detail, according to the image processing apparatus 10 according to the present embodiment, the area determination unit 12 includes the background level determination unit 22, the average brightness calculation unit 23, the threshold value calculation unit 25, and the comparison unit 27. And. This makes it possible to accurately determine whether the target pixel is lower or higher than the surrounding background level. In addition, since the area discriminating unit 12 includes a halftone dot discriminating unit 21 and an edge discriminating unit 24, a character area on the halftone dot is calculated by calculating the logical product of these discrimination results by an AND circuit 30. It can be determined accurately.
[0054]
In addition, white characters on the halftone dots can be accurately determined. That is, in the case of an image in which white characters are drawn on a halftone dot, the outside of the character is determined to be a character edge, so the character tends to be slightly thicker. It can be determined with high accuracy.
[0055]
It should be noted that the above-described embodiment is merely an example and does not limit the present invention in any way, and various improvements and modifications can be made without departing from the scope of the invention. For example, in the above-described embodiment, the background level is determined and the average brightness is calculated for the entire area of the input image. However, the halftone dot region is determined first, and the halftone dot region is determined. These processes can be performed only on the area.
[0056]
Further, the background level determination unit 22 determines whether the background is a white background or a black background (or a high-density chromatic color background), and inputs the determination signal to the comparison unit, whereby the calculation method in the threshold value calculation unit The circuit configuration can be simplified by switching the magnitude comparison relationship in the comparison unit. Needless to say, the specific numerical values (for example, the size of the logical filter) exemplified in the above-described embodiment are merely examples.
[0057]
【The invention's effect】
As described above, according to the present invention, there are provided an image area determination device and an image area determination program capable of accurately determining a character area on a halftone dot.
[Brief description of the drawings]
FIG. 1 is a block diagram illustrating a schematic configuration of an image processing apparatus according to an embodiment.
FIG. 2 is a block diagram illustrating a configuration of an area determination unit in FIG. 1;
FIG. 3 is a diagram illustrating a processing region of a background level determination unit in FIG. 2;
FIG. 4 is a diagram illustrating an example of a brightness histogram created by a background level determination unit.
FIG. 5 is a diagram illustrating another example of a brightness histogram created by a background level determination unit.
6 is a diagram illustrating an averaging filter used in the average brightness calculation unit in FIG. 2; FIG.
FIG. 7 is a diagram illustrating an example of an input image.
8 is a diagram illustrating an output example of a brightness data creation unit in FIG. 2; FIG.
FIG. 9 is a diagram illustrating an output example of an average brightness calculation unit.
10 is a diagram illustrating an output example of a comparison unit in FIG. 2; FIG.
FIG. 11 is a diagram illustrating an output example of an edge determination unit in FIG. 2;
12 is a diagram illustrating an output example of the AND circuit in FIG. 2; FIG.
FIG. 13 is a diagram illustrating an example of an input image.
14 is a diagram showing a result of area discrimination performed on the image of FIG. 13 by an area discrimination unit.
FIG. 15 is a diagram illustrating another example (outline character) of an input image.
16 is a diagram illustrating an output example of a comparison unit for the image of FIG. 15;
17 is a diagram illustrating an output example of an edge determination unit for the image of FIG. 15;
18 is a diagram illustrating an output example of an AND circuit for the image of FIG. 15;
FIG. 19 is a diagram illustrating a result of area determination performed on the image of FIG. 15 by an area determination unit.
FIG. 20 is a flowchart showing processing contents when the present invention is realized by software.
FIG. 21 is a flowchart showing the processing contents when the present invention is realized by software.
[Explanation of symbols]
10 Image processing device
12 Area discriminator
14 CPU
21 Halftone discrimination unit
22 Ground level discriminator
23 Average brightness calculator
24 Edge discriminator
25 Threshold calculation unit
27 Comparison part
30 AND circuit

Claims (8)

First density calculating means for calculating an average density in the first region including the target pixel;
Second density calculating means for calculating the background density in the second region including the target pixel;
The calculation result of the first density calculation means is a threshold value calculated based on the background density calculated by the second density calculation means and an offset amount set to a positive or negative value according to the background density; Character area determination means for determining whether or not it belongs to the character area by comparison,
A discriminating means for discriminating whether or not the pixel of interest belongs to a character area on a halftone dot based on a discrimination result of the character area discriminating means;
An image area discriminating apparatus characterized by comprising: In the image region discrimination device according to claim 1,
The image area discrimination apparatus, wherein the second area is larger than the first area. In the image area discrimination device according to claim 1 or 2 ,
The determination unit determines that the target pixel belongs to a character area on a halftone dot when the average density of the target pixel calculated by the first density calculation unit is lower than the threshold value. Area discriminator. In any one image region discriminating device according to claim 1 ,
Halftone dot region determining means for determining whether the pixel of interest belongs to a halftone dot region;
Edge region determination means for determining whether the pixel of interest belongs to the edge region based on the density difference between the pixel of interest and the surrounding pixels;
The discriminating means is discriminated to be a halftone dot area by the halftone dot area discriminating means, is discriminated to be an edge area by the edge area discriminating means, and is discriminated to belong to a character area by the character area discriminating means. An image region discriminating apparatus that discriminates that the pixel of interest belongs to a character edge region on a halftone dot. A first density calculating step for calculating an average density in the first region including the target pixel;
A second density calculating step for calculating the background density in the second region including the target pixel;
A threshold value calculated based on the calculation result of the first density calculation step based on the background density calculated by the second density calculation step and an offset amount set to a positive or negative value according to the background density; A character area determination step for comparing to determine whether or not the character area belongs,
A determination step of determining whether the pixel of interest belongs to a character region on a halftone dot based on the determination result of the character region determination step;
An image region discrimination method characterized by comprising: In the image region discrimination method according to claim 5 ,
A halftone dot region determining step for determining whether the pixel of interest belongs to a halftone dot region;
An edge region determination step for determining whether or not the target pixel belongs to the edge region based on a density difference between the target pixel and the surrounding pixels;
In the determining step, the halftone dot region determining step determines that the region is a halftone dot region, the edge region determining step determines that the region is an edge region, and the character region determining step belongs to the character region An image region determination method, wherein if determined, the target pixel is determined to belong to a character edge region on a halftone dot. On the computer,
A first density calculating step for calculating an average density in the first region including the target pixel;
A second density calculating step for calculating the background density in the second region including the target pixel;
A threshold value calculated based on the calculation result of the first density calculation step based on the background density calculated by the second density calculation step and an offset amount set to a positive or negative value according to the background density; A character area determination step for comparing to determine whether or not the character area belongs,
A determination step of determining whether the pixel of interest belongs to a character region on a halftone dot based on the determination result of the character region determination step;
An image area discrimination program for executing In the image region discrimination program according to claim 7 ,
A halftone dot region determining step for determining whether the pixel of interest belongs to a halftone dot region;
An edge region determination step for determining whether the pixel of interest belongs to the edge region based on a density difference between the pixel of interest and surrounding pixels;
In the determining step, the halftone dot region determining step determines that the region is a halftone dot region, the edge region determining step determines that the region is an edge region, and the character region determining step belongs to the character region An image area determination program for causing a function of determining that a target pixel belongs to a character edge area on a halftone dot when it is determined.

Images