JP3043040B2 - Image binarization method - Google Patents

Description

Description: BACKGROUND OF THE INVENTION The present invention relates to an image for obtaining a binary image by binarizing a document image or the like in a communication device such as a facsimile, a copying machine, a document image database input device, or the like. It relates to a binarization method.

(Prior Art) Conventionally, technologies in such a field include, for example,
There was one described in JP-A-61-103372. Hereinafter, the configuration will be described with reference to the drawings.

FIG. 2 is a block diagram showing an example of the configuration of an image binarization apparatus using the conventional image binarization method described in the above-mentioned document.

This image binarization device obtains a binary image by optimally binarizing a multivalued image including halftone density pixels (dots) such as a photograph, and stores a previous pixel which operates in synchronization with a clock pulse CLK. Circuit 1, pre-pixel storage circuit 2, and pixel storage circuit 3 directly above
It has. Further, the output side of these storage circuits 1, 2, 3 is connected to a comparison / determination processing unit 10 and a smoothing circuit 20, and the output side of the smoothing circuit 20 is connected to a comparator.
Connected to 21. The output side of the comparator 21 is connected to the comparison determination processing unit 10, and the output side of the comparison determination processing unit 10 is connected to the encoding circuit 22. The comparison / determination processing unit 10 includes decision circuits 11, 12, and 13 and a logic circuit 14 connected to its output side.

In the image binarization method using this type of image binarization device, when a pixel density IN is input, the previous pixel storage circuit 1
According to the difference value between the density of the pixel of interest stored in the pixel storage circuit 2 immediately before and the pixel storage circuit 3 immediately before and the density of the pixel in the vicinity thereof, the determination circuits 11 to 13 in the comparison determination processing unit 10 determine The corresponding pixels are classified into white, black, and unknown according to the neighboring pixels and input to the logic circuit 14.

On the other hand, the density obtained by smoothing the density of the pixel of interest and the density of pixels in the vicinity thereof by the smoothing circuit 20 is converted to dither by the comparator 21 that inputs the threshold value TH, and is compared and processed by the comparison determination processing unit.
Input to the logic circuit 14 in 10.

When there is black in the output of the decision circuits 11 to 13, the logic circuit 14 outputs black as the binarized output of the pixel of interest,
When there is no black in the outputs 11 to 13 and there is white, white is output as the binarized output of the target pixel. Also, the logic circuit 14
When the outputs of the decision circuits 11 to 13 are all unknown, the output of the comparator 21 is output as a binarized output of the target pixel. The output of the logic circuit 14 is encoded by an encoding circuit 22 and output (OU
T) is done.

FIG. 3 is a flowchart for explaining the operation of the comparison / determination processing unit 10 in FIG.

As shown in FIG. 3, the comparison and determination processing unit 10 including the decision circuits 11 to 13 and the logic circuit 14
The pixel density values of M0 and the peripheral pixels M1 to M3 are input. Steps
In S2, the density difference value obtained by subtracting the density value of the target pixel M0 from the highest density value among the density values of the peripheral pixels M1 to M3 is Aw, and the density value of the peripheral pixels M1 to M3 is calculated from the density value of the target pixel M0. Ab is the density difference value obtained by subtracting the lowest density value from the values. Then, in step S3, if the value of the density difference value Ab obtained in step S2 is larger than the threshold value α, in step S5, 2
Outputs black as a quantified output.

If the value of the density difference value Ab obtained in step S2 is smaller than the threshold value in step S3, the density difference value Aw is compared with the threshold value β in step S4. If the value of the density difference value Aw obtained in step S2 is larger than the threshold value β,
In step S6, white is output as the binarized output of the target pixel. When the value of the density difference value Ab obtained in step S2 is smaller than the threshold value α and the value of the density difference value Aw is smaller than the threshold value β, the comparison is performed in step S7. The output of the unit 21 is output as a binarized output of the pixel of interest.

(Problems to be solved by the invention) However, the above-described image binarization method has the following problems.

4 (a) to 4 (c) are diagrams for explaining the problems of the prior art, wherein FIG. 4 (a) is an original image, FIG. 4 (b) is a read image, and FIG. 4 (c) is an output. Each image is shown.

In the conventional image binarization method, for example, when the original image shown in FIG. 4A is read in accordance with the read scanning pixels DA shown in FIG. 4, the density relationship for each pixel is shown in FIG. It becomes as shown in. In FIG. 4 (b), in particular, the density of the pixel at the boundary between the character pattern color (black) PB and the background pattern color (white) PW is indicated by L1 to L4 in FIG. 4 (b). Density value (intermediate density value) that takes an intermediate value between white and white W
Becomes When the read image (the density differences A, B, C, and D are all larger than the threshold values α and β) is binarized using a conventional image binarization method, L1 to L4 As shown in FIG. 4 (c), for the pixels having the intermediate density values, an output image in which the binarized output of all the target pixels is black is obtained. Results in greatly different output images, and does not necessarily improve visual intelligibility (image quality), and it has been difficult to solve them.

The present invention solves the problems of the prior art,
For example, when binarizing an image in which halftone dots and characters are mixed,
An object of the present invention is to provide an image binarization method that takes into consideration not only binarization of halftone dots but also binarization of characters as in the related art.

(Means for Solving the Problems) In order to solve the above-mentioned problems, the present invention is to input a density signal of an image including a pixel of halftone density, and to obtain the density of a target pixel and the density of surrounding pixels. And a dithered output obtained by smoothing the density of the surrounding pixels and the dithered pixel, and performing a comparison determination process to determine the black and white of the target pixel, and outputs a binary output according to the determination result. In the image binarization method, the following processing is performed in the comparison determination processing.

That is, in the comparison determination processing, the density difference value Ab obtained by subtracting the lowest value among the density values of the peripheral pixels from the density value of the target pixel, and the density difference value Ab obtained from the highest density value of the peripheral pixels. The density difference value Aw obtained by subtracting the density value of the pixel of interest and the threshold Tz
Ab> Tz and Aw ≦ Tz, or Ab> Tz, Aw>
In the case of Tz and Ab> Aw, black is output as the binarized output of the target pixel, and Ab ≦ Tz and Aw> Tz, or Ab> Tz, Aw>
When Tz and Ab ≦ Aw, white is output as the binarized output of the target pixel, and when Ab ≦ Tz and Aw ≦ Tz, the dithered output is output as the binarized output of the target pixel. ing.

(Operation) According to the present invention, since the image binarization method is configured as described above, in the comparison / determination processing for determining the black and white of the pixel of interest, the density value of the pixel of interest is calculated based on the density value of the peripheral pixel. If the density difference value Ab obtained by subtracting the lowest value is larger than the threshold value Tz, and the density difference value Aw obtained by subtracting the density value of the target pixel from the highest value among the density values of the peripheral pixels is smaller than the threshold value Tz, Alternatively, if the value of Ab is larger than the threshold Tz, the value of Aw is larger than the threshold Tz, and the value of Ab is larger than the value of Aw, black is output as the binarized output of the target pixel. If the value of Ab is smaller than the threshold Tz and the value of Aw is larger than the threshold Tz, or
If the value of Ab is larger than the threshold Tz, the value of Aw is larger than the threshold Tz, and the value of Ab is smaller than the value of Aw, 2
Outputs white as a value output. If the value of Ab is smaller than the threshold value Tz and the value of Aw is smaller than the threshold value Tz, the dithered output is output as the binarized output of the target pixel.

In other words, in the conventional method, in the comparison determination processing for determining the black and white of the pixel of interest, for example, in order to optimize the image of the halftone dot, the binarized output of the pixel of interest is blacked out for all pixels having intermediate density values. And had On the other hand, in the present invention, for example, in order to optimally binarize a character in addition to a halftone dot, whether a pixel having an intermediate density value is a pixel closer to black or a pixel closer to white Judge,
In the former case, the binarized output of the target pixel is black, and in the latter case, the binarized output of the target pixel is white. As a result, an output image closer to the original image and having visually improved intelligibility can be obtained. Therefore, the above problem can be solved.

(Embodiment) FIG. 1 is a flowchart of a comparison / determination process in an image binarization method according to an embodiment of the present invention, and FIG. 5 is an image binary image using the image binarization method according to an embodiment of the present invention. It is a block diagram of a structure of a conversion apparatus.

First, the configuration of the image binarization device shown in FIG. 5 will be described. This image binarization device obtains a binary image by optimally binarizing a multivalued image including halftone density pixels (halftone dots) such as a photograph, and obtains a binary image in synchronization with a clock pulse CLK. Storage circuit 31, pre-pre-pixel storage circuit 32 and directly above pixel storage circuit 33
It has. The previous pixel storage circuit 31 is a circuit that receives the density IN of the pixel and stores the density of the previous pixel,
Reference numeral 32 denotes a circuit for storing the density of the pixel before the previous pixel, and a pixel storage circuit 33 directly above stores a density of a pixel located immediately above the pixel of interest. The output side of these storage circuits 31 to 33 is connected to a comparison / determination processing unit 40 and a smoothing circuit 50, and the smoothing circuit 50 is input to the comparison / determination processing unit 40 via a comparator 51. Have been.

The comparison / determination processing unit 40 includes comparators 41-1 and 41-2 for comparing the density IN and the output of the storage circuits 31 to 33, and on the output side, differentiators 42-1 and 42- for performing a difference operation. 2 are connected. Further, the output side of the differentiators 42-1 and 42-2 receives the threshold value Tz and performs a comparison process or the like on the comparator 43-1.
To the comparators 43-1 to 43-3.
The logic circuit 44 is connected to the output side of the.

The smoothing circuit 50 is a circuit for smoothing the pixel density based on the density IN and the outputs of the storage circuits 31 to 33, and a comparator 51 connected to the output side of the smoothing circuit 50 sets the output of the smoothing circuit 50 to a threshold value. This circuit dithers using TH and supplies the dithered output to the logic circuit 44.

The output side of the logic circuit 44 is connected to an encoding circuit 52 for encoding the logic circuit output and outputting (OUT).

Image 2 of the present embodiment using such an image binarization device
A method of converting the value will be described.

When the density IN of the pixel is input, the comparators 41-1 and 41- in the comparison determination processing unit 40 compare the density values of the peripheral pixels stored in the pixel storage circuit 32 and the pixel storage circuit 33 immediately before. According to 2, the highest value and the lowest value are obtained from the density values of the peripheral pixels and input to the differentiators 42-1 and 42-2, respectively. The differentiators 42-1 and 42-2 output values of the comparators 41-1 and 41-2,
The difference from the density value of the target pixel stored in the previous pixel storage circuit 31 is obtained, and the difference value is input to the comparators 43-1 to 43-3.

The comparators 43-1 to 43-3 compare the respective difference values output from the differentiators 42-1 and 42-2 with the input threshold Tz, and compare the difference values output from the differentiator 42-1. The value is compared with the difference value output from the differentiator 42-2, and the magnitude determination result obtained as a result is input to the logic circuit 44.

On the other hand, the density of the target pixel and the density of peripheral pixels in the vicinity thereof are smoothed by a smoothing circuit 50, and the smoothed density is compared with a comparator.
Give to 51. The comparator 51 dithers the output of the smoothing circuit 50 using the input threshold value TH and inputs the dithered signal to the logic circuit 44.

The logic circuit 44 refers to each determination result output from the comparators 43-1 to 43-3 and the output value of the comparator 51, and makes the binarized output of the target pixel white or black. And outputs the result of the determination. The output of the logic circuit 44 is encoded by the encoding circuit 52 and output (OUT).

Next, the comparison / determination processing in the comparison / determination processing unit 40, which is a feature of the present embodiment, will be described with reference to FIG.

In the processing by the comparators 41-1 and 41-2 and the differentiators 42-1 and 42-2 in the comparison determination processing unit 40, first, in step S11, the pixel density value of the target pixel M0 and the peripheral pixels M1 to M3 is input. In step S12, the density difference value obtained by subtracting the density value of the target pixel M0 from the highest density value among the density values of the peripheral pixels M1 to M3 is set to Aw, and the density difference value of the target pixel M0 is set to Aw.
Ab is the density difference value obtained by subtracting the lowest density value among the M3 density values.
Is input to the comparators 43-1 to 43-3.

In the processing by the comparators 43-1 to 43-3, step S12
In steps S13 to S15, the values of the density difference values Ab and Aw and the input threshold value Tz obtained in step S13 to S15 are compared, and the value of the density difference value Ab and the value of Aw are compared in step S16. The result of the judgment is input to the logic circuit 44.

In the process by the logic circuit 44, in step S13,
Based on the comparison results of the comparators 43-1 to 43-3, step S1
If the value of the density difference value Ab obtained in step 2 is larger than the threshold value Tz, the process proceeds to step S14. In step S14, if the value of the density difference value Ab obtained in step S12 is larger than the threshold Tz,
In S16, the value of the density difference value Ab is compared with the value of Aw. If the value of the density difference value Ab is larger, black is output as the binarized output of the target pixel in step S17. On the other hand, when the value of the density difference value Aw is larger, step S18
Outputs white as a binarized output of the pixel of interest.

If the value of the density difference value Aw is smaller than the threshold value Tz in step S14 based on the comparison result of the comparators 43-1 to 43-3, black is output as the binarized output of the target pixel in step S19.

On the basis of the comparison result of the comparators 43-1 to 43-3, step
If the value of the density difference value Ab is smaller than the threshold value Tz in S13, the process proceeds to step S15. In Step S15, if the value of the density difference value Aw obtained in Step S12 is larger than the threshold Tz, Step S20
Outputs white as a binarized output of the pixel of interest. On the other hand, if the value of the density difference value Aw is smaller than the threshold value Tz in step S15, the output (dithered output) of the comparator 51 is output as the binarized output of the target pixel in step S21.

 This embodiment has the following advantages.

6 (a) to 6 (c) are diagrams for explaining the effect of the present embodiment. FIG. 6 (a) is an original image, FIG. 6 (b) is a read image, and FIG. An image is shown.

For example, when the original image shown in FIG. 6A is read in accordance with the read scanning pixels DA shown in FIG. 6, the relationship of the density for each pixel is as shown in FIG. 6B. In FIG. 6B, in particular, the density of the pixel at the boundary between the character pattern color (black) PB and the background pattern color (white) PW is an intermediate value between black B and white W as indicated by l1 to l4. , And the read image (density differences a, b, c, and d are all larger than the threshold value Tz) is binarized using the method of this embodiment. Is performed as shown in FIG. 6 (C). As shown in FIG. 6 (c), as shown in FIG.
For pixels with intermediate density values such as l4, pixels that are closer to black, such as l1 and l3, output black as the binarized output of the pixel of interest, and pixels that are closer to white, such as l2 and l4. Outputs white as the binarized output of the pixel of interest. Therefore, it is possible to obtain an output image closer to the original image and having visually improved intelligibility (image quality).

Further, in step S16 of FIG. 1, the magnitudes of Aw and Ab are compared using the density difference values Aw and Ab obtained in step S12. Therefore, since it is not necessary to obtain another comparison element, it is possible to simplify the operation processing and improve the operation speed accordingly.

Note that the present invention is not limited to the above embodiment, and various modifications are possible. For example, there are the following modifications.

(I) In FIG. 5, the threshold value Tz input to the comparators 43-1 and 43-3 is represented by Tz, and the threshold value input to the comparator 43-1 is represented by Tz.
1. Even if the threshold value input to the comparator 43-3 is input separately from Tz2, the same effect as in the above embodiment can be obtained.

(Ii) The circuit configuration for realizing the comparison / determination processing in FIG.
The circuit is not limited to the circuit shown in FIG. 5, but may be modified to another configuration, or may be configured to be executed by program control of a microcomputer or the like in addition to configuring each block of the circuit as an individual circuit. good.

(Effects of the Invention) As described in detail above, according to the present invention, the following processing is performed in the comparison determination processing. The density difference value Ab obtained by subtracting the lowest value of the density values of the peripheral pixels from the density value of the target pixel is larger than the threshold value Tz, and the density value of the target pixel is calculated from the highest value of the density values of the peripheral pixels. If the subtracted density difference value Aw is smaller than the threshold Tz, or the value of Ab is larger than the threshold Tz, the value of Aw is larger than the threshold Tz,
If the value of Ab is larger than the value of Aw, black is output as the binarized output of the target pixel. Ab is less than the threshold Tz,
And if the value of Aw is greater than the threshold Tz, or the value of Ab is greater than the threshold Tz, the value of Aw is greater than the threshold Tz, and Ab
Is smaller than the value of Aw, white is output as the binarized output of the target pixel. Further, the value of Ab is smaller than the threshold Tz,
If the value of Aw is smaller than the threshold value Tz, the dithered output is output as the binarized output of the target pixel.

Therefore, it is determined whether the pixel having the intermediate density value is a pixel closer to black or a pixel closer to white. In the former case, black is output as a binarized output of the target pixel.
In the latter case, white is output as the binarized output of the target pixel.

As described above, according to the present invention, the density difference value Aw is compared with the threshold value Tz, and the density difference values Ab and Aw are compared. In this case, the image is not blackened, and an output image closer to the original image and having visually improved intelligibility (image quality) can be obtained easily and accurately.

Further, in the present invention, the magnitude difference Ab> Aw is compared using the density difference values Aw and Ab once obtained. Therefore, since it is not necessary to obtain another comparison element, it is possible to simplify the operation processing and improve the operation speed accordingly.

[Brief description of the drawings]

FIG. 1 is a flowchart of a comparison / determination process in an image binarization method according to an embodiment of the present invention, and FIG.
FIG. 3 is a block diagram of a configuration of an image binarizing apparatus using a binarizing method, FIG. 3 is a flowchart of a comparison / determination process in FIG. 2, and FIGS. 4 (a) to 4 (c) are diagrams for explaining problems of the prior art. , Fifth
FIG. 6 is a block diagram showing a configuration of an image binarizing apparatus using an image binarizing method according to an embodiment of the present invention. FIGS. 6A to 6C are diagrams for explaining effects of the embodiment of the present invention. is there. 31 ... Previous pixel storage circuit, 32 ... Prior pixel storage circuit, 33 ...
... Directly above pixel storage circuit, 40... Comparison judgment processing section, 41-1, 41
-2,43-1 to 43-3 ... Comparator, 42-1,42-2 ... Differentiator, 44 ... Logic circuit, 50 ... Smoothing circuit, 51 ... Comparator, 52 ... Code Circuit.

──────────────────────────────────────────────────続 き Continued on front page (58) Field surveyed (Int.Cl. ⁷ , DB name) H04N 1/40-1/409

Claims (1)

(57) [Claims] 1. A density signal of an image including a pixel having a halftone density is inputted, and the density of a target pixel and the density of peripheral pixels thereof are smoothed, and the density of the target pixel and the density of peripheral pixels are smoothed. The image binarization method of performing black-and-white determination of the target pixel by performing a comparison determination process based on the obtained dithered output and outputting a binary output according to the determination result. A density difference value Ab obtained by subtracting the lowest value among the density values of the peripheral pixels from the density value of the target pixel, and a density obtained by subtracting the density value of the target pixel from the highest density value of the peripheral pixels. The difference value Aw is compared with the threshold value Tz, and Ab> Tz and Aw ≦ Tz, or Ab> Tz, Aw> Tz and Ab> Aw
In the case of, black is output as a binarized output of the target pixel, and Ab ≦ Tz and Aw> Tz, or Ab> Tz, Aw> Tz and Ab ≦ Aw
, White is output as a binarized output of the pixel of interest, and if Ab ≦ Tz and Aw ≦ Tz, the dithered output is output as a binarized output of the pixel of interest. Value method.