JPH06284288A - Processing method for picture information - Google Patents

Abstract

PURPOSE:To obtain an always stable and excellent picture regardless of line density of picture information and outputted picture density when the density difference between a noted picture element and an adjacent picture element is detected and it is compared with a threshold level so as to discriminate the surface part of the picture information. CONSTITUTION:A picture inputted from a scanner 1 not being a surface part is subjected to emphasis processing by revising a threshold level or a picture element position at which density difference from a noted picture element is detected depending on linear density of picture information. Or in the other invention, when an outputted picture density is selected by a selector 5 as 'usual' or 'darker', the threshold level or the picture element position is revised similarly to the above mentioned depending on the selection state and a surface is discriminated by a surface discrimination device 4. Thus, when there is density change of a constant level or over at all times regardless of linear density of picture information, it is discriminated not to be the surface and emphasis processing is executed. Or in the other invention, when 'darker' picture density is selected, the output of surface noise is prevented an even when 'usual' density is selected, a thin original picture is outputted clearly.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method of processing image information which determines whether or not each pixel of image information is a background portion of a document and emphasizes the shade of the pixel when the pixel is not the background portion.

[0002]

2. Description of the Related Art An image processing apparatus for reading and processing a document image determines whether or not each pixel of image information read from the document is a background portion of the document image, and when it is not a background portion, MTF (Modulation Tracing).
nsfer Function) correction is often performed. The MTF correction is a process of increasing the density difference between adjacent pixels, and by doing so, for example, a character image of a character original having a low density can be output clearly.

As one method of determining the background portion of image information, a method of detecting a density difference between a pixel of interest and an adjacent pixel is well known. In this method, nine pixels A to I are sequentially extracted in a 3 × 3 matrix form from the image information of one page as shown in FIG. Then, the pixel E at the center is noticed, and the density difference between the noticed pixel E and the adjacent pixels B, D, F, H is detected. Then, when each of the detected density differences is less than or equal to a certain threshold value, it is determined that the target pixel E is a background portion.

That is, in this determination method, the pixels D to F are
Is used to determine the magnitude of the density change in the main scanning direction, and the pixels B, E, and H are used to determine the magnitude of the density change in the sub-scanning direction.

By the way, for example, the linear density of an image of a G3 facsimile apparatus is expressed in the form of main scanning direction × sub scanning direction, 8 × 7.7 (dots / mm), 8 × 1.
There are three types, 5.4 (dots / mm) and 16 × 15.4 (dots / mm). The user arbitrarily uses these linear densities.

Now, as shown in FIG. 10, an original image whose density changes with a constant gradient in the sub-scanning direction is
It is assumed that reading is performed at a linear density of × 7.7 (dots / mm). When executing the determination processing of the background portion, pixels B and E,
The density difference between the E and H pixels is detected. In this case, the density difference d corresponding to the sub-scanning pitch of 1 / 7.7 mm is detected, and the density difference d is compared with a fixed threshold value.

Next, the same original document is printed in 8 × 15.4 (dot /
It is assumed that the reading is performed with a linear density of mm. In this case, since the sub-scanning pitch is 1 / 15.4 mm, the detected density difference is ½ of the density difference d.

Despite the change in the detected density difference depending on the linear density of the image information as described above, conventionally, the detected density difference is always compared with a constant threshold value. Therefore, when the same portion of the same document is read at a low linear density, it is determined that it is not a background and is emphasized, but when read at a high linear density, it is determined as a background portion and an emphasis process is executed. There was a thing that wasn't done. in this way,
The determination condition of the background and the processing method of image information were changed depending on the read line density.

By the way, for example, in the case of a facsimile machine, the background of the image information is determined as described above, and the portion of the image information which is not the background is subjected to MTF correction and then binarized. Then, the binarized image information is recorded on the recording paper or transmitted.

In addition, in a normal facsimile apparatus, when recording or transmitting image information, the image density output by the operator can be arbitrarily switched, such as "dark" or "normal". . The switching of the image density is performed by changing the threshold value when binarizing the image information to high or low.

Now, suppose that the image information read by the scanner section includes background noise N as shown in FIG. The background noise N is generated by reading the background dirt of the original image.

When the background judgment is executed on this image information, if the threshold value for the background judgment is set small, it is judged that the background noise N is not the background, as shown in FIG. In addition, the background noise N is also emphasized by the MTF correction.

When binarizing this image information, a constant threshold value S1 is set when the image density is "normal", and a threshold value lower than the threshold value S1 when the image density is "dense". S
2 is set.

It is assumed that the operator selects the image density "normal" and binarizes it with the threshold value S1.
In this case, in this case, the level of the background noise N has not reached the threshold value S1, so that the binarized image information includes
Background noise does not occur. Further, in this case, since the background judgment threshold value is set small, even a small density change portion is emphasized by MTF correction. Therefore, a clear output image can be obtained even with a thin character image.

However, it is assumed here that the image density "dark" is selected and binarized by the threshold value S2. In this case, since the level of the background noise N exceeds the threshold value S2, the background noise occurs in the binarized image information, and the image quality deteriorates.

Conventionally, in order to prevent such deterioration of image quality, the threshold value for determining the background has been set large. As a result, the background noise N of FIG.
Can be determined as the background, and the background noise N can be prevented from being emphasized by the MTF correction. Therefore, the image information does not reach the threshold value S2, and the background noise does not occur in the binarized image information.

Conventionally, the threshold value for determining the background has always been set to a constant value regardless of the image density to be output. Therefore, if it is attempted to prevent the background noise N when the image density “dark” is selected, the thin portion of the original image cannot be emphasized, and the image quality of the output image is deteriorated.

[0018]

As described above, in the related art, since the judgment condition of the background portion and the processing method of the image information are changed depending on the line density of the image information, a stable image cannot be obtained,
In addition, there is a problem that the image quality is deteriorated such that background noise appears or the image becomes lighter depending on the density at which the image is output.

SUMMARY OF THE INVENTION It is an object of the present invention to solve the above problems and to provide a method for processing image information which can always obtain a stable and good image regardless of the linear density of the image information and the image density to be output. And

[0020]

To this end, the first aspect of the present application
In the invention, when the background judgment of the image information is executed, the density difference between the target pixel and the adjacent pixel is detected as in the conventional case, and the threshold value to be compared with the detected density difference is set as the read line density. The higher the value, the smaller the value.

In the second aspect of the invention, the pixel position for detecting the density difference between the pixel of interest and the pixel of interest is set to a position adjacent to the pixel of interest when the read line density is low. The position is set apart from the pixel of interest.

According to the third aspect of the present invention, the image information is binarized and output, and the density of the output image is switched by changing the threshold when binarizing the image information, and the background judgment of the image information is executed. In this case, when outputting a dark image, the threshold value for determining the background is set large, while when outputting a light image, the threshold value for determining the background is set small. .

According to the fourth aspect of the present invention, when a background image of image information is executed by the same device as described above, when outputting a dark image, each pixel position for detecting the density difference from the target pixel is set to the target pixel. On the other hand, when outputting a thin image, each pixel position is set to a position distant from the pixel of interest.

[0024]

In the first and second aspects of the present invention, even if the reading linear density of the original image changes, it is possible to always determine the background portion when the density change is less than a certain value, so that a stable image can be obtained. To be

Further, in the third and fourth aspects of the invention, when a dark image is output, the background noise is not emphasized because the background portion is determined even if the density change is relatively large. Noise can be prevented from appearing. Further, when outputting a thin image, even if the density change is relatively small, it is determined that it is not the background portion, and the image information is emphasized. Therefore, even a thin original image can be output clearly. As a result, a good image is always obtained.

[0026]

Embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

FIG. 1 is a block diagram showing the arrangement of an image processing apparatus according to the first embodiment of the present invention. In the figure, a scanner 1 reads a document image in multiple gradations. In this embodiment, the number of gradations read by the scanner 1 is 32, and the read line density is 8 × 7.7 (dots / m).
m) and 8 × 15.4 (dots / mm).

The matrix memory 2 sequentially extracts a plurality of pixels having a fixed matrix size from the read image information. The MTF correction unit 3 performs MTF correction on a pixel of interest of image information. The background determination unit 4 determines whether or not the pixel of interest is the background portion of the original image.

The selection circuit 5 selects, in pixel units, whether to take out the target pixel as it is or to take it out after performing the MTF correction. The binarization unit 6 changes the image information of 32 gradations into the image information of monochrome 2 gradations. The plotter 7 records image information on recording paper. Transmitter 8
Is for transmitting image information to another device.

In the case of using the image processing apparatus of the present embodiment with the above configuration, the operator sets the original on the scanner 1 and 8 × 7.7 (dots / mm) or 8 × 15.4.
The read line density of (dots / mm) is selected and the device is activated.

When the image processing apparatus is activated, it reads a document image and outputs image information. The matrix memory 2 sequentially extracts and outputs nine pixels A to I with a matrix size of 3 × 3 from the output image information, as shown in FIG. Here, the extracted pixel E becomes the target pixel.

The MTF correction unit 3 calculates the target pixel E according to the following equation.
Is corrected to the density E '. In addition, the pixel code of the following mathematical formula shall show a pixel density. E ′ = 3 · E− (B + D + F + H) / 2 As a result, the target pixel E is emphasized.

The background determination unit 4 determines whether the target pixel E is a background portion of the original image. In this determination process, the target pixel E and each pixel B, D, F, H are referred to in the same manner as above. In this case, as shown in FIG. 2, the read linear density is 8 ×.
In the case of 7.7 (dots / mm), the interval between each pixel is 1/8 mm in the main scanning direction and 1 / 7.7 mm in the sub scanning direction.
In the case of 8 × 15.4 (dots / mm), the main scanning direction is the same as above, and the sub scanning direction is 1 / 15.4 mm.

The background determination unit 4 has a density difference | B- between the target pixel E and the pixels B, D, F, and H as shown by the determination formula in FIG.
E |, | DE |, | FE |, | HE | are detected. Then, both the density differences | BE- and | HE- are smaller than the threshold value J1, and the density differences | DE- and | F-.
If both E | and the threshold value J2 are smaller than the threshold value J2, it is determined that the target pixel E is a background portion. At this time, when the read linear density is 8 × 7.7 (dots / mm), the threshold value J1,
Both J2 are set to "4/32", and in the case of 8 x 15.4 (dots / mm), the threshold J1 is set to "2/32" and the threshold J2 is set to "4/32."

The selection circuit 5 takes out the image information of the target pixel E output from the matrix memory 2 as it is when the target pixel E is determined to be the background portion by the background determination unit 4. On the other hand, if it is not determined that the background portion, MTF
The image information of the target pixel E corrected to the density M ′ by the correction unit 3 is extracted.

The binarizing unit 6 compares the density of the extracted image information with a fixed threshold value to convert the image information into black and white two-tone image information. The plotter 7 records the image information on a recording sheet, and the transmission unit 8 transmits it to another device.

As described above, in the present embodiment, the read line density in the sub-scanning direction is switched in two steps. However, the density difference | BE |, | H- detected between the pixels in the sub-scanning direction. The threshold value J1 to be compared with E | is set to be large when the reading line density is low, and is set to be small when the reading line density is high.

Therefore, even if the read line density changes, if the change in the density of the original image is less than a certain value, it can be determined that it is a background portion. As a result, the MTF correction of the image information can always be executed under the same condition, so that a stable image can be obtained.

Next, a second embodiment of the present invention will be described.

The image processing apparatus of the present embodiment has the same configuration as that of FIG. 1, and the matrix memory 2 has a size of 5 × as shown in FIG.
It is assumed that 25 pixels A to Y are extracted with a matrix size of 5.

Then, in the present embodiment, the background determination unit 4 executes the background determination process with the central pixel M as the pixel of interest. For example, the reading linear density is 8 x 7.7 (dots / mm)
In this case, as shown in FIG. 4, with reference to the target pixel M and the adjacent pixels H, L, N, and R, the respective density differences | HM
|, | LM |, | NM |, | RM | are detected.
Then, when all the density differences are smaller than the threshold value J, it is determined that the pixel of interest M is a background portion.

When the read line density is 8 × 15.4 (dots / mm), the target pixel M and the adjacent pixels L, N
, And the pixels C and W separated by two pixels, the respective density differences | LM-, | NM-, | CM-, | WM-
Detects |. Then, when all the density differences are smaller than the threshold value J, it is determined that the pixel of interest M is a background portion.

In this embodiment, the threshold value J is always set to "4/32" regardless of the read line density. And
Similar to the above-described embodiment, MTF correction is performed on image information that is not the background portion. In this case, the density of the target pixel M is corrected to the density M ′ by the following equation. M ′ = 3 · M− (H + L + N + R) / 2 Then, the image information is binarized and recorded or transmitted.

As described above, in this embodiment, the threshold value J for judging the background is always set to a constant value. And
In the sub-scanning direction that is switched in two steps, the pixels for detecting the density difference from the target pixel M are set to the adjacent pixels H and R when the reading line density is low, and 1 when the reading line density is high. Pixels C and W separated from each other are set.

As a result, like the first embodiment,
Even if the read line density changes, if the change in the density of the original image is less than a certain value, it can be determined as the background portion, and a stable image can be obtained.

In each of the above embodiments, the case where the read line density changes only in the sub-scanning direction is taken as an example. Therefore, only the threshold value for determining the background and the reference pixel position correspond to the sub-scanning direction. Although it is changed, if the main scanning direction also changes, the threshold value for background determination and the reference pixel position corresponding to the scanning direction may be similarly changed.

Next, a third embodiment of the present invention will be described.

The image processing apparatus of this embodiment has the same configuration as that of FIG. 1, and the reading linear density of the scanner 1 is fixed at 8 × 7.7 (dots / mm). Further, in this embodiment, the density of the image information to be output is set to "normal" or "dark".
As described above, it is assumed that the operator has a function of arbitrarily selecting. Then, as shown in FIG. 5, the selection result is input to the background determination unit 4 and the binarization unit 6, and the operation condition is changed. The matrix memory 2 is assumed to extract nine pixels A to I with a matrix size of 3 × 3 as shown in FIG.

In this embodiment, the operator arbitrarily selects the density of image information to be output and activates the apparatus.

As a result, the image processing apparatus starts a predetermined operation. In this case, the background determination unit 4 refers to four pixels B, D, F, and H adjacent to the target pixel E as shown in FIG. 6, and the respective density differences | BE |, | D- E |,
| FE | and | HE | are detected. Then, when all the density differences are smaller than the threshold value J, it is determined that the pixel of interest E is a background portion. At this time, as shown in FIG. 7, the threshold value J is set to “4/32” when the image density “normal” is selected, and is set to “6/32” when “dark” is selected. Set to.

In this way, the background portion is determined, and the MTF correction is performed on the image information that is not the background portion.

When the image density "normal" is selected, the binarizing unit 6 binarizes the image information by using, for example, half the maximum density level of the image information as a threshold value. When the image density is “dark”, the image information is binarized with a threshold value lower than the threshold value.

In this way, the binarized image information is recorded or transmitted.

As described above, in this embodiment, the background difference is always detected by detecting the density difference between the target pixel E and the adjacent pixels B, D, F and H. Then, when outputting a dark image, the threshold value J for determining the background is set to be large, so that the background portion is determined even if the density change is relatively large, and therefore background noise is not emphasized. , It is possible to prevent the background noise from appearing in the output image. Further, when outputting a light image, the threshold value J is set to a small value. Therefore, even a light image portion with a small change in density is determined not to be the background, and the gradation is emphasized, so a clear image is output. To be done. As a result, a good image can always be obtained.

Next, a fourth embodiment of the present invention will be described.

The image processing apparatus of this embodiment has the same configuration as that of the third embodiment, and the matrix memory 2 extracts 25 pixels A to Y as shown in FIG.

In this embodiment, when the image density is selected as "normal" as shown in FIG.
By referring to the pixel of interest M and the pixels C, K, O, and W separated by two pixels, the respective density differences | C−M |, | K−M |,
| OM |, | WM | is detected. Then, when all the density differences are smaller than the threshold value J, it is determined that the pixel of interest M is a background portion.

When the image density is selected to be "dense", the density difference | HM− ||| of the pixel of interest M and the adjacent pixels H, L, N and R are referred to. LM
|, | NM |, | RM | are detected. Then, when all the density differences are smaller than the threshold value J, it is determined that the pixel of interest M is a background portion.

In this embodiment, the threshold value J is always set to "4/32" regardless of the image density. Then, as in the above-described embodiment, M is applied to image information that is not the background portion.
Perform TF correction. Then, the image information is binarized and recorded or transmitted.

As described above, in this embodiment, the threshold value J for determining the background is always set to a constant value. Then, when outputting a dark image, the pixels for detecting the density difference from the target pixel M are set to the adjacent pixels L, H, N, and R, while when outputting a light image, they are separated by two pixels. Pixels C, K, O, and W are set.

As a result, similarly to the third embodiment,
When a dark image is output, even if the change in density is relatively large, it is determined that it is a background portion, and background noise in the output image is prevented. Also, when outputting a thin image,
Even if the density change is relatively small, it is determined that it is not the background portion, and a thin image becomes clear. This allows
A good image can always be obtained.

In each of the above embodiments, the number of read gradations of image information is 32, but it goes without saying that it can be set arbitrarily. Also, it goes without saying that the threshold value for determining the background can be arbitrarily set to an appropriate value according to the number of gradations.

[0063]

As described above, according to the first invention of the present application,
The threshold value to be compared with the density difference detected between the pixel of interest and the adjacent pixel is changed to a smaller value as the read line density becomes higher. Can be determined and a stable image can be obtained.

In the second aspect of the invention, the pixel position for detecting the density difference between the pixel of interest and the pixel of interest is changed according to the read line density. Therefore, similarly to the above, regardless of the read line density, A stable image can be obtained.

According to the third aspect of the invention, by changing the threshold value for determining the background in accordance with the image density to be output, when outputting a dark image, the background noise is prevented from being output and a thin image is output. In such a case, since an image with a small change in density is emphasized, a good image can always be obtained.

Further, according to the fourth aspect of the invention, by changing the pixel position for detecting the density difference from the target pixel according to the output image density, the background noise is prevented from being output and the density change is carried out in the same manner as described above. Since a small image is emphasized, a good image can always be obtained.

[Brief description of drawings]

FIG. 1 is a block configuration diagram of an image processing apparatus according to a first embodiment of the present invention.

FIG. 2 is an explanatory diagram of a background determination process of the above embodiment.

FIG. 3 is an explanatory diagram of extracted pixels of a matrix memory in the image processing apparatus according to the second embodiment of the present invention.

FIG. 4 is an explanatory diagram of a background determination process of the above embodiment.

FIG. 5 is a block diagram showing a part of an image processing apparatus according to a third embodiment of the present invention.

FIG. 6 is an explanatory diagram of a background determination process of the above embodiment.

FIG. 7 is an explanatory diagram of a threshold setting method in the background determination processing.

FIG. 8 is an explanatory diagram of a background determination process in the image processing apparatus according to the fourth embodiment of the present invention.

FIG. 9 is an explanatory diagram of extracted pixels in a conventional background determination process.

FIG. 10 is an explanatory diagram of a document image density detection operation at one reading linear density.

FIG. 11 is an explanatory diagram of a document image density detection operation at another reading linear density.

FIG. 12 is an explanatory diagram illustrating an operation example in which background noise is output.

[Explanation of symbols]

 1 Scanner 2 Matrix Memory 3 MTF Correction Unit 4 Background Texture Determination Unit 5 Selection Circuit 6 Binarization Unit 7 Plotter 8 Transmission Unit

Claims (4)

[Claims] 1. An original image is read at an arbitrarily set reading linear density, attention is paid to each pixel of the obtained image information, the density difference between the pixel of interest and each of its surrounding pixels is detected, and the detected density difference is detected. And a constant threshold value are compared to determine whether or not the pixel of interest is the background portion of the document, and when it is not the background portion, in the image information processing method of emphasizing the pixel of interest, the density difference from the pixel of interest is calculated. The image information processing method is characterized in that each pixel position to be detected is set at a position adjacent to the pixel of interest, and the threshold value is changed to a smaller value as the read line density becomes higher. 2. An original image is read at an arbitrarily set reading linear density, attention is paid to each pixel of the obtained image information, a density difference between the pixel of interest and each of its surrounding pixels is detected, and the detected density difference is detected. And a constant threshold value are compared to determine whether or not the pixel of interest is the background portion of the document, and when it is not the background portion, in the image information processing method of emphasizing the pixel of interest, the density difference from the pixel of interest is calculated. Each pixel position to be detected is set to a position adjacent to the pixel of interest when the read line density is low, and is set to a position distant from the pixel of interest as the read line density increases. Processing method. 3. Focusing on each pixel of the image information, detecting the density difference between the pixel of interest and each of its surrounding pixels, comparing the detected density difference with a threshold value for background judgment, and the pixel of interest is the original document. If it is not the background portion, the pixel of interest is emphasized, while the processed image information is binarized and output, and the threshold at the time of binarization is changed to high or low. In the image information processing method of switching the image density to be output, the threshold value for the background judgment is set large when a dark image is output, while the background judgment is performed when a thin image is output. A method of processing image information, characterized by setting a small threshold value. 4. A pixel of interest is an original document by focusing attention on each pixel of image information, detecting a density difference between the pixel of interest and each of its surrounding pixels, and comparing the detected density difference with a threshold value for background judgment. If it is not the background portion, the pixel of interest is emphasized, while the processed image information is binarized and output, and the threshold at the time of binarization is changed to high or low. In the image information processing method of switching the image density to be output by outputting the image, when outputting a dark image, each pixel position for detecting the density difference with the target pixel is set to a position adjacent to the target pixel, while When outputting an image,
A method for processing image information, wherein each pixel position for detecting the density difference is set at a position apart from the pixel of interest.