JP3472289B2 - Image processing device - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing apparatus for binarizing multi-valued image data obtained by raster scanning and a document image in which a binary image and a grayscale image are mixed in one page. , MTF correction and simple binarization with a fixed threshold value are performed on the obtained binary image portion of the multivalued image information, while gamma correction and pseudo halftone are performed on the grayscale image portion. The present invention relates to an image processing device that executes binarization.

[0002]

2. Description of the Related Art For example, in an image processing apparatus for processing image data such as a facsimile machine, a binary image obtained by reading an original image as multi-valued image data and binarizing the multi-valued image data. Image data is transmitted or stored.

When binarizing multi-valued image data, as a matter of course, since the amount of information is lost, the received original obtained by recording and outputting the binary image data and the stored binary The image quality of the output image when the image data is extracted and output is deteriorated as compared with the original document image.

Since the image quality deteriorates when the multi-valued image data is converted into the binary image data as described above, various processes have been conventionally performed so as to prevent the influence thereof as much as possible. For example, a spatial frequency correction (MTF correction) calculation process is applied to a non-halftone image to correct a blurred image, or a halftone image may be binarized by an error diffusion method that can further improve the image quality. Has been done.

Further, a grayscale image (halftone image) such as a photograph
In the case of reading, the gamma correction processing or the like is performed on the obtained multivalued image information, and various processing such as pseudo halftone is executed on the corrected multivalued image information. The gamma correction is a process of correcting the deviation of the gradation characteristic between the image reading unit and the image output unit.

By the way, when a binary image and a grayscale image are mixed in one page of the original image, the above-mentioned various correction processes are executed for each image.

FIG. 23 shows an example of an image processing apparatus for executing such correction processing. That is, the multi-valued image information read from the original line by line is input to the line buffer 1001 and the gamma correction unit 1002. The line buffer 1001 temporarily stores the image information for a certain number of lines. The MTF correction unit 1003 performs MTF correction on the stored image information, and the binarization unit 1004
Binarizes the corrected image information.

On the other hand, the gamma correction unit 1002 gamma-corrects the read multi-valued image information, and the line buffer 1
005 temporarily stores the corrected image information for a certain number of lines. The pseudo halftone processing unit 1006 binarizes the stored image information by pseudo halftone processing. The image synthesizing unit 1007 outputs the image information of the binary image area in one page from the binarizing unit 1004 among the image information binarized by the pseudo halftone processing unit 1006 and the binarizing unit 1004. Image information of the grayscale image is input from the pseudo halftone processing unit 1006 to form a one-page image.

[0009]

However, these conventional methods have a disadvantage that the image quality of a binary image cannot be sufficiently improved.

Further, in the above-mentioned conventional apparatus, when the predetermined correction processing is executed for the binary image and the grayscale image in one page, two line buffers 1001 are provided for temporarily storing the image information. Since 1005 is required, there is an inconvenience that the device cost becomes high.

The present invention has been made in order to eliminate such inconvenience, and an object of the present invention is to provide an image processing apparatus capable of improving the image quality of a binary image. Also, 1
An object of the present invention is to provide an image processing apparatus capable of reducing the apparatus cost when performing various predetermined correction processes on a binary image and a grayscale image in a page.

[0012]

According to the present invention, a document image in which a binary image and a grayscale image are mixed in one page is read, and the binary image portion of the obtained multivalued image information is read. In the image processing apparatus, which performs gamma correction, MTF correction, and simple binarization with a constant threshold value, while performing gamma correction and binarization with pseudo halftone on a grayscale image portion, an original image Input the multi-valued image information obtained by scanning, and in the low and high density areas of the multi-valued image information, the same L
The gamma correction means for performing gamma correction by the input / output characteristic correction characteristic that is non-linear in the og curve and linear in the intermediate density range, a buffer memory for temporarily storing the gamma-corrected multivalued image information, and the stored Binarized image processing means for performing MTF correction and simple binarization on multi-valued image information, and 2 by pseudo halftone for the grayscale image of the multi-valued image information stored in the buffer memory. And a grayscale image processing means for performing the binarization.

Further, the gamma correction means comprises means for inputting multivalued image information as a digital signal and outputting gamma-corrected multivalued image information as a digital signal.

Further, the gamma correction means comprises analog-digital conversion means for inputting multi-valued image information as an analog signal and outputting gamma-corrected multi-valued image information as a digital signal.

[0015]

Therefore, since the gamma correction is performed with the linear characteristic in the intermediate density area of the multivalued image information, the multivalued image information holding the intermediate density information can be obtained. In a binary image,
Since the MTF correction and the simple binarization are performed on the multi-valued image information, it is possible to obtain the simple binarized image information equivalent to the conventional one without impairing the effect of the MTF correction. Further, in the grayscale image, since the gamma-corrected multivalued image information is binarized by the pseudo halftone, it is possible to obtain the same grayscale image information as the conventional one. In this case, since only one buffer memory is required, the device cost is reduced.

[0016]

DETAILED DESCRIPTION OF THE INVENTION Referring to the accompanying drawings,
Embodiments of the present invention will be described in detail.

In the binarization processing by the error diffusion method, for example, as shown in FIG. 1, according to the result of binarizing the target pixel A which is the target of the binarization processing, the original attention is given. An error between the density of the pixel A and the density assigned to the target pixel A is calculated based on the binarization result, and the error is averaged to the adjacent pixels B, C, D, and E appearing after the target pixel A. It is intended to be distributed to.

The error diffusion method is described by the following equation.

F '(xy) = f (xy) + (1 / Σa
(Ij)) Σa (ij) e (x + ii + j)

[0020] e (xy) = f '(xy) -B (f' (xy) ≧ T)

[0021] e (xy) = f '(xy) (f' (xy) <T)

Here, f (xy) is an input value (input density), f '(xy) is a correction value (correction density), and a (ij).
Is a diffusion coefficient, e (xy) is a binarization error, T is a threshold, and B is a black density level.

For example, when the number of bits of the input value is 4 bits and the input value of the target pixel A is the intermediate density 7, if the value of the threshold T is set to the intermediate value (8), the binarization error e at this time is set.
The value of (xy) becomes 7, and the binarization error is reduced to the adjacent pixel B,
When averaged to C, D, and E, the error value diffused to each adjacent pixel B, C, D, and E is (7 /
4).

Here, for example, in order to simplify the configuration of the circuit that performs the error diffusion processing operation, if the error diffusion processing operation is performed by an integer operation, the value of this diffused error (7 /
In 4), the value after the decimal point is rounded down, that is, 1.

Therefore, each adjacent pixel B, C,
The value of the error diffused in D and E becomes 1, and the sum of them is 4. Therefore, the error value (7) of the target pixel A does not match, and the image quality of the binarized image is deteriorated, which has been a conventional problem.

Therefore, in the present invention, the bit number of the multi-valued image data obtained by the raster scan is increased, and the error diffusion calculation process is performed by the integer calculation circuit in that state, whereby the image quality of the binary image is obtained. Is prevented from deteriorating.

FIG. 2 shows an image processing apparatus according to an embodiment of the present invention.

In FIG. 1, an image input section 1 reads an image of a read document (not shown) by raster scanning, and an analog image signal AV obtained by the raster scanning is obtained by an analog / digital converter 2. It is converted into a 4-bit digital image signal PD and added to the bit number conversion unit 3.

The bit number conversion unit 3 converts the 4-bit digital image signal PD into a 6-bit extended image signal DD, and the extended image signal DD is added to the error correction unit 4.

The error correction unit 4, the binarization unit 5, and the error calculation unit 6 apply the above-described error diffusion calculation process to the 6-bit extended image signal DD, and the calculation is an integer calculation. It is realized by. The binarized image signal BW output from the binarization unit 5 is output to the data output unit 7.
Via an external device (not shown).

FIG. 3 shows an example of the bit number converter 3.

In the bit number converter 3, the 4-bit digital image signal PD is set as it is in the upper 4 digits of the 6-bit extended image signal DD, and the data B is set in the lower 2 digits of the extended image signal DD. '00 (B 'represents a binary number; the same applies below) is fixedly set.

Therefore, as shown in FIG. 4, the value of the extended image signal DD changes according to the value of the digital image signal PD.

Here, consider a case where the error diffusion processing is executed under the same conditions as described above. Under the conditions described above, the input value of the target pixel A is 7, and the corresponding extended image signal D
The value of D is 28 (see FIG. 4).

If the binarization threshold value of the binarization unit 5 is set to the median value (32), the density of this pixel of interest A is smaller than the threshold value, and the error value is 28. This error value (2
8) is equally distributed to the adjacent pixels B, C, D and E,
The value is 7 = (28/4). Since this value (7) is an integer value, all the density errors of the target pixel A are diffused to the adjacent pixels B, C, D and E.

In this way, since the calculation error of the error diffusion processing calculation does not appear when the target pixel A has the value of the intermediate density, the deterioration of the image quality can be prevented as a result when the entire image is considered. .

In this way, in this embodiment, the image quality of the binary image obtained by the error diffusion processing can be improved.

FIG. 5 shows another example of the bit number converter 3.

In this case, the 4-bit digital image signal PD
Is set to the upper 4 bits of the 6-bit extended image signal DD, and the 2-bit data output from the random number generator 3a is
It is set to the lower 2 bits of the extended image signal DD.

In this case, unlike the one shown in FIG. 3, the data of the lower 2 bits becomes the data generated by the random number generator 3a, and is not fixed to B'00. In the case of FIG. 3, when the value of the black density B in the error diffusion processing is set to (63), the maximum density of the extended image signal DD is (60), so that the density of the binary image becomes low overall. , A blank image may appear on the black solid part. On the other hand, in the case of FIG. 5, since the data B′111111 (= 63) is stochastically generated, such a situation can be avoided.

FIG. 6 shows still another example of the bit number converter 3.

In this case, the 4-bit digital image signal PD
Is set to the upper 4 digits of the 6-bit extended image signal DD, and the AND circuit 3b forms a logical product of each bit of the digital image signal PD, and the output of the AND circuit 3b is
It is set in the lower two digits of the extended image signal DD.

Therefore, in this case, as shown in FIG. 7, the extended image signal DD is changed according to the value of the digital image signal PD.
The value of changes. That is, when the value of the digital image signal PD is B'1111, the value of the extended image signal DD is B'1.
It becomes 11111, and the black solid part can be reproduced.

FIG. 8 shows still another example of the bit number converter 3.

In this case, the 4-bit digital image signal PD
Is set to the upper 4 digits of the 6-bit extended image signal DD, and the value of bit 3 of the digital image signal PD is set to bit 1 of the extended image signal DD and bit 2 of the digital image signal PD.
Are set in bit 0 of the extended image signal DD.

Therefore, in this case, as shown in FIG. 9, according to the value of the digital image signal PD, the extended image signal DD
The value of changes. That is, the values of the digital image signal PD are 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 1.
When 1, 12, 13, 14, and 15, the values of the extended image signal DD are 0, 4, 8, 12, 17, 21, 21, 25, and 2, respectively.
9, 34, 38, 42, 46, 51, 55, 59, 63
The value of the extended image signal DD reflects the value of the original digital image signal PD well.

In each of the embodiments described above, the value of the 4-bit digital image signal PD is converted into the 6-bit extended image signal DD, but the number of bits of these image signals is not limited to this. There is no such thing.

By the way, when a non-halftone image is converted into a binarized image, a spatial frequency correction process is performed in order to eliminate the blurring of the image that is caused by the optical characteristics of the reading optical system and the like.

In this spatial frequency (MTF) correction process, for example, a pixel matrix as shown in FIG. 10 is considered, and the density of the target pixel E to be binarized is set to the adjacent pixels B, D, F, and The concentration of H is used for correction by the calculation shown in the following formula (I).

[0050] E '= 3E- (B + D + F + H) / 2 (I)

To increase the degree of correction in this spatial frequency correction processing, the calculation shown in the following equation (II) is applied.

[0052] E '= 5E- (B + D + F + H) (II)

Generally, when the degree of spatial frequency correction is increased, the edge of a character image or the like becomes clear and the resolution is improved. However, when the correction degree is increased too much (overcorrection), background noise is generated in the image. It appears and the image quality deteriorates. Further, when reading fine characters or fine patterns on a document, the effect of image blurring is large, and moire occurs in image data, resulting in deterioration of resolution.

Therefore, the calculation of the above formula (II) is applied to the fine part of the image to enhance the degree of spatial frequency correction, and the calculation of the above formula (I) is applied to the other part including the background part. If the degree of spatial frequency correction is not increased so that an image with good image quality can be obtained.

FIG. 11 shows an image processing apparatus according to another embodiment of the present invention. In the figure, the same parts as those in FIG. 2 and corresponding parts are designated by the same reference numerals.

In the figure, the 4-bit digital image signal PD output from the analog / digital converter 2 is M
It is added to the TF (spatial frequency) correction unit 10 and the MTF mode determination unit 11. The MTF mode determination unit 11
The mode of MTF calculation is determined based on the density of the target pixel E, and the determination result is used as the mode setting signal MD.
Is output to the MTF correction unit 10.

The MTF correction unit 10 operates in the mode 1 by the mode setting signal MD added from the MTF mode determination unit 11.
Is set, the calculation of the above formula (I) is executed, and when the mode 2 is set by the mode setting signal MD, the calculation of the above formula (II) is executed to set the target pixel E. The density is corrected, and the correction result is added to the binarization unit 5 as a corrected image signal CD.

The binarization unit 5 binarizes the corrected image signal CD output from the MTF correction unit 10 with a predetermined threshold value, and the processing result is the binarized image signal BW as data. It is output to an external device via the output unit 7.

FIG. 12 shows an example of processing of the MTF mode determination section 11.

First, the MTF judging section 11 checks whether or not the density of the target pixel E is within the range of 5 or more and 10 or less (decision 101), and the result of the judgment 101 is YE.
When it becomes S, it is an intermediate density, and in this case, since it is a portion of the original image such as fine characters, mode 2 is set in order to increase the degree of MTF correction (Processing 10).
2) When the result of judgment 101 is NO,
Since it is the other part, the mode 1 is set so that the degree of MTF correction is not so large (process 103).

With the above configuration, when the image input unit 1 is reading a portion of a fine image, the analog image signal AV
Since the value of is in the intermediate area, the MTF mode determination unit 1
1 sets the MTF correction mode to the mode 2 by the above-mentioned processing
The mode setting signal MD is notified to the MTF correction unit 10.

As a result, the MTF correction unit 10 executes the calculation shown in the above equation (II) to correct the density of the target pixel E, and the correction result is sent to the binarization unit 5 as the corrected image signal CD. Output.

On the other hand, when the image input unit 1 is reading a portion other than a fine image, the area where the value of the analog image signal AV is small or large becomes large. Therefore, the MTF mode judgment unit 11 performs the MTF correction by the above-described processing. The mode is set to the mode 1, and the MTF correction unit 10 is notified by the mode setting signal MD.

As a result, the MTF correction unit 10 executes the calculation shown in the above formula (I) to correct the density of the target pixel E, and the correction result is used as the corrected image signal CD in the binarization unit 5.
Output to.

In this way, in this embodiment, since the MTF correction calculation according to the image being read is performed, the image quality of the image of the binary image signal BW can be improved.

FIG. 13 shows another processing example of the MTF mode determination section 11.

In this case, the MTF mode determining section 11 determines the target pixel E and its adjacent pixels A, B, C, D, F, G, and
Average value of H and I concentrations J (= (A + B + C + D + E + F
+ G + H + I) / 9) is calculated (process 201), and it is checked whether or not the value of the average value J is within the range of 5 or more and 10 or less (decision 202).

When the result of the determination 202 is YES, the density is intermediate, and in this case, since it is a portion of the original image such as fine characters, mode 2 is set in order to increase the degree of MTF correction (process 203). , Again, judgment 20
When the result of 2 is NO, since it is the other part, the mode 1 is set so that the degree of MTF correction is not so large (process 204).

Further, in order to judge whether or not the read portion is a fine image, as described above, the density of the target pixel E, the target pixel E and its adjacent pixels A, B,
There is a method other than the method of examining the average value J of the concentrations of C, D, F, G, H, and I.

For example, when a specific pattern appearing in a fine image is examined and such a pattern is detected,
It is also possible to determine that the image being read is a fine image portion. FIG. 14 shows a processing example executed by the MTF mode determination unit 11 in this case.

First, the target pixel E and its adjacent pixels A,
Matrix data of B, C, D, F, G, H, and I is input (process 301), and it is checked by pattern matching whether the matrix data matches a pattern (not shown) peculiar to a fine image. (Processing 302), it is determined whether or not any of the patterns matches (determination 30).
3).

When the result of judgment 303 is YES, since it is a fine image portion, mode 2 is set in order to increase the degree of MTF correction (process 304), and when the result of judgment 303 is NO, it is set. Since it is a part other than, in order not to increase the degree of MTF correction too much,
Mode 1 is set (process 305).

By the way, some originals read by a facsimile machine include a mixture of halftone images such as photographs and non-halftone images such as characters. As described above, in a mixed image in which a halftone image and a non-halftone image are mixed, if the MTF correction calculation is executed for all areas, there is a disadvantage that the image quality of the halftone image part is deteriorated, and For example, when the binarization processing of the halftone image mode such as the error diffusion processing is executed for the area of
There is an inconvenience that the image quality of the non-halftone image part is deteriorated.

FIG. 15 shows an image processing apparatus according to another embodiment of the present invention which can eliminate such inconvenience. In the figure, the same parts as those in FIGS. 2 and 11 and corresponding parts are designated by the same reference numerals.

In the figure, the 4-bit digital image signal PD output from the analog / digital converter 2 is M
TF correction unit 10, halftone mode binarization unit 15, and
It is added to the image area determination unit 16.

The MTF correction unit 10 executes the calculation of the above formula (I) when the mode 1 is designated by the mode setting signal MM supplied from the mode setting unit 17, and
When the mode 2 is designated by the mode setting signal MM, the calculation of the above formula (II) is executed to correct the density of the target pixel E, and the correction result is binarized as the corrected image signal CD. Added to Part 5. Binarization unit 5
Is a binarized correction image signal CD based on a predetermined threshold,
The binarization result is used as a binary image signal BW, and the image composition unit 18
Output to one input terminal of.

The halftone mode binarization unit 15 executes the above-mentioned error diffusion calculation based on the inputted digital image signal PD, and the binary image signal B obtained thereby.
X is added to the other input end of the image composition unit 18. Note that the delay time for the processing of the halftone mode binarization unit 15 and the delay time for the processing of the MTF correction unit 10 and the binarization unit 5 match.

Based on the input digital image signal PD, the image area determination unit 16 determines whether the pixel corresponding to the digital image signal PD belongs to the halftone area or the non-halftone area. A well-known image area determination calculation is performed, and the image area determination signal KK obtained by this is output to the image synthesizing unit 18. Note that in this case, the image area determination unit 16 determines that only the vicinity of the character image of the image is
That is, it is determined as a non-halftone area.

In the state where the mode 2 is set by the mode setting signal MM, the image synthesizing section 18 determines the image area determining section 1
When the image area determination signal KK added from 6 represents the halftone area, the binary image signal BX output from the halftone mode binarizing unit 15 is selected, and
When the image area determination signal KK represents a non-halftone area, the binary image signal BW output by the binarizing unit 5 is selected, and the selected binary image signal is output to the data output unit 7. It is output to an external device via the. Further, the image synthesizing unit 18 always selects the binary image signal BW output from the binarizing unit 5 and outputs it to the data output unit 7 in the state where the mode 1 is set by the mode setting signal MM. ing.

The mode setting unit 17 controls the value of the mode setting signal MM based on the image type signal SS supplied from an external control device (not shown), and the mode setting signal MM is the MTF correction unit. 10 and the image composition unit 18. In this case, when the image type signal SS designates a mixed image original, the contents indicating the mode 2 are set in the mode setting signal MM, and the image type signal SS indicates a character image original. If it is specified that the mode 1 is set, the content indicating the mode 1 is set in the mode setting signal MM.

With the above configuration, when a mixed image original is designated by the image type signal SS, the mode setting section 17 outputs the mode setting signal MM representing the mode 2, whereby the MTF correcting section 10 is caused. The mode 2 MTF correction calculation is applied to the original.

When the image reading operation of the image input unit 1 is started, the analog image signal AV output from the image input unit 1 is converted by the analog / digital converter 2 into the corresponding digital image signal PD, and the MTF correction unit. 10, the halftone mode binarization unit 15, and the image area determination unit 16.

In this case, the MTF correction unit 10 executes the MTF correction calculation of mode 2 to form the corrected image signal CD, and the binary image signal BW corresponding to this corrected image signal CD is converted into the binarization unit. 5 are sequentially output and added to one input end of the image synthesizing unit 18.

Further, the halftone mode binarization unit 15 executes the error diffusion processing operation based on the digital image signal PD, and the binary image signals BX obtained thereby are sequentially output to the image synthesis unit 18. It is added to the other input.

In this state, the image area determination unit 16 executes the image area determination processing to determine whether the image being read is a halftone area.
Alternatively, the image area determination signal KK indicating whether it is a non-halftone area is output.

Therefore, in this case, since the mode 2 is designated by the mode setting signal MM, the image synthesizing section 18 determines from the halftone mode binarizing section 15 for the area determined to be the halftone area. The binary image signal BX to be output is selected, and for the region determined to be the non-halftone region, the binary image signal BW output from the binarizing unit 5 is selected to output the data output unit. Output to 7.

Further, when the original of the character image is designated by the image type signal SS, the mode setting section 17
A mode setting signal MM representing the mode 1 is output, whereby the MTF correction unit 10 executes the MTF correction calculation of the mode 1 and outputs the corrected image signal CD.

In addition, in this case, since the mode 1 is designated by the mode setting signal MM, the image synthesizing unit 18 outputs the binary image signal BW output from the binarizing unit 5.
Is selected and output to the data output unit 7.

As described above, in this embodiment, the vicinity of the character image in the mixed image is the MTF of mode 2.
Since the correction calculation is applied, the edge portion becomes clear and the image quality becomes very good. In this case, since the image of the other part is determined as the halftone area,
Binary image signal B output from the halftone mode binarization unit 15
Since X is selected, it is possible to prevent the output of an image in which the background stain is conspicuous. If the image is not a mixed image, the MTF correction calculation in Mode 1 is applied, so that an image with an appropriate image quality is output.

By the way, in this embodiment, the edge portion of the character image can be reproduced clearly by performing the MTF correction operation of mode 2 in the vicinity of the character image in the mixed image. Even if the threshold value is set to a small value, the same effect can be obtained.

FIG. 16 shows an image processing apparatus according to still another embodiment of the present invention. In the figure,
The same parts as those in FIG. 15 and corresponding parts are designated by the same reference numerals.

In the figure, the MTF correction unit 10 always executes the above-described mode 1 MTF correction calculation, and the corrected image signal CD obtained by the MTF correction calculation is the binarization unit 5.
Has been added to.

When the mode 1 is set by the mode setting signal MM, the binarizing unit 5 applies the reference threshold value (for example, 8) to binarize the corrected image signal CD and set the mode. When mode 2 is set by the signal MM, a small threshold value (for example, 6) is applied,
The corrected image signal CD is binarized, and the processing result is added to the image synthesizing unit 18 as a binary image signal BW.

With the above configuration, when the original of the mixed image is designated by the image type signal SS, the mode setting section 17 outputs the mode setting signal MM representing the mode 2, whereby the binarization section 5 is outputted. Applies a small threshold for the original.

When the image reading operation of the image input section 1 is started, the analog image signal AV output from the image input section 1 is converted by the analog / digital converter 2 into the corresponding digital image signal PD, and the MTF correction section. 10, the halftone mode binarization unit 15, and the image area determination unit 16.

The MTF correction section 10 executes the MTF correction calculation of mode 1 to form the corrected image signal CD, and the binarization section 5
Applies a small threshold value to the corrected image signal CD to form a binary image signal BW.
Has been added to one input end.

Further, the halftone mode binarization unit 15 executes the error diffusion processing operation based on the digital image signal PD, and the binary image signals BX obtained thereby are sequentially output to the image synthesis unit 18. It is added to the other input.

In this state, the image area determination unit 16 executes the image area determination processing to determine whether the image being read is a halftone area.
Alternatively, the image area determination signal KK indicating whether it is a non-halftone area is output.

Therefore, in this case, since the mode 2 is designated by the mode setting signal MM, the image synthesizing section 18 determines from the halftone mode binarizing section 15 for the area determined to be the halftone area. The binary image signal BX to be output is selected, and for the region determined to be the non-halftone region, the binary image signal BW output from the binarizing unit 5 is selected to output the data output unit. Output to 7.

Further, when the original of the character image is designated by the image type signal SS, the mode setting section 17
A mode setting signal MM representing the mode 1 is output, whereby the binarization unit 5 uses the standard threshold to correct the corrected image signal C.
Binarize D.

In addition, in this case, since the mode 1 is designated by the mode setting signal MM, the image synthesizing unit 18 outputs the binary image signal BW output from the binarizing unit 5.
Is selected and output to the data output unit 7.

As described above, in the present embodiment, since the small threshold value is applied by the binarizing unit 5 in the vicinity of the character image in the mixed image, its edge portion becomes clear and the image quality becomes very good. Become. In this case, since the image of the other portion is determined as the halftone area, the binary image signal BX output from the halftone mode binarization unit 15 is selected, and thus the background stain is conspicuous. It is possible to prevent such an image from being output.

Although the number of bits of the digital image signal PD is set to 4 in the above embodiment, the number of bits is not limited to this.

Further, in the above-described embodiment, the halftone mode binarization unit performs the binarization process using the error diffusion method. However, other halftone mode binarization processes, for example, dither. Binarization processing using a matrix can also be applied.

FIG. 17 is a block diagram showing the arrangement of an image processing apparatus according to another embodiment of the present invention. In the figure, a line sensor 1101 reads an original image line by line and extracts an image signal of an analog signal. A / D (analog / digital) converter 1102
Is for converting the image signal into multi-valued image information which is a digital signal. The gamma correction unit 1103 performs gamma correction on multi-valued image information, and the gamma table 1104
The input / output characteristic in the gamma correction is stored.

The line buffer 1105 temporarily stores multi-valued image information for a fixed number of lines. MTF correction unit 1
Reference numeral 106 denotes MTF correction of the multi-valued image information, and the binarization unit 1107 simply binarizes the multi-valued image information with a constant threshold value. Pseudo halftone binarization unit 1108
Is for binarizing multi-valued image information by known processing such as dither processing or error diffusion processing.

The image area discriminating section 1109 discriminates whether each pixel of the multivalued image information is a binary image area or a grayscale image area of the original image. The image composition unit 1110 uses 2
Multivalued image information is selectively taken out from the binarizing unit 1107 and the pseudo halftone binarizing unit 1108 and combined as a one-page image. The image output unit 1111 displays or records an image on recording paper.

FIG. 18 shows a block diagram of the MTF correction unit 1106. In the figure, the pixel extraction circuit 11
Reference numeral 06a is for extracting image information of 9 pixels centered on one pixel of interest. The sum operation circuit 1106b calculates the sum of the respective density levels of the four input pixels, and the right shift circuit 1106c shifts the input binary bit data by 1 bit to the lower side to reduce the numerical value by half.
It is something to do. The left shift circuit 1106d doubles the numerical value by shifting the input binary bit data by 1 bit to the upper side. Sum operation circuit 1
106e is for calculating the sum of two density signals, and the difference calculation circuit 1106f is for calculating the difference between two density signals.

With the above arrangement, if the image processing apparatus of this embodiment starts operation, the line sensor 1101
Outputs the image signals read from the original image one line at a time. The A / D converter 1102 converts the image signal into a digital signal having a fixed number of bits, that is, multi-valued image information.
The gamma correction unit 1103 gamma-corrects the multi-valued image information according to the input / output characteristics stored in the gamma table 1104.

Now, for example, the A / D converter 1102
The multi-valued image information is output in 256 gradations with white density of "0" and black density of "255". The gamma correction unit 1103 sets the white density of "0" and black density of "63" in 64 gradations. Shall be output. In this case, the gamma correction unit 1103
According to the characteristic curve a shown in FIG. 19, the A / D conversion unit 1
The density of the multivalued image information input from 102 is corrected. That is, the characteristic curve b in the figure is a conventional correction characteristic, and in the present embodiment, in the low density region where the image information density is close to white and the high density region where it is close to black, the density is non-linear as in the conventional case. On the other hand, in the intermediate density range, the multi-valued image information is output with a linear characteristic in which the output is proportional to the input.

The line buffer 1105 temporarily stores the density-corrected multi-valued image information for a fixed number of lines such as two lines. MTF correction unit 1
The pixel extraction circuit 1106a of 106 extracts 9 pixels from pixels A to I according to the template (pixel matrix) shown in FIG. 10 from the latest one line read from the original image and the image information of the two lines before that. Extract sequentially. The central pixel E of the extracted pixels is the target pixel to be processed.

The sum calculation circuit 1106b calculates the sum of the pixel densities of the four extracted pixels B, D, F and H in binary. The right shift circuit 1106c shifts the calculated binary data to the lower bit by 1 bit to reduce the data value to ½. The left shift circuit 1106d is
By shifting the density data of the pixel of interest E to the upper bit by 1 bit, the data value is doubled. Sum operation circuit 110
6e calculates the sum of the doubled data value and the original data value, that is, the tripled value of the original data value. Then, the difference calculation circuit 1106f calculates a value obtained by subtracting the data value output by the right shift circuit 1106c from the data value output by the sum calculation circuit 1106e, as a correction pixel E ′.

As a result, the pixel densities of B, D, and
If E, F, and H are set, the corrected pixel E ′ that has been MTF corrected according to the known calculation shown in the following equation is obtained.

E '= 3.E- (B + D + F + H) / 2

The MTF correction unit 1106 thus sets the MT
The F-corrected multivalued image information is output. Binarization unit 1107
Simply binarizes the multi-valued image information by comparing it with a fixed threshold value.

On the other hand, the pseudo halftone binarization unit 1108 binarizes the image information stored in the line buffer 1105 by a predetermined dither process or error diffusion process.

Now, it is assumed that one page of an original image contains a binary image such as characters and a grayscale image such as a photograph. The image area determination unit 1109 determines whether each pixel of the image information is a binary image area or a grayscale image area by a known technique.

Based on the discrimination result, the image synthesizing unit 1110 inputs the image information of the binary image area from the binarizing unit 1107, and the image information of the grayscale image area from the pseudo halftone binarizing unit 1108. Input and combine into one page of image information. The image output unit 111 displays or records the combined image information of one page.

As described above, in the present embodiment, the multi-valued image information obtained by reading the original image is subjected to the non-linear density correction in the light density area and the dark area in the same manner as the conventional method, while the intermediate density Then, gamma correction is performed with linear characteristics. Then, the corrected multi-valued image information is temporarily stored in the buffer memory, and the multi-valued image information is subjected to MTF correction and simple 2
Binarization and binarization by pseudo halftone are executed.

Since the gamma correction is executed with a linear characteristic in the intermediate density area of the multivalued image information, the multivalued image information holding the intermediate density information can be obtained. In the binary image area, MTF correction and simple binarization are performed on the multi-valued image information, so that the effect of MTF correction is not impaired.
It is possible to obtain the same binary image information as that of the related art. Further, in the grayscale image area, the gamma-corrected multivalued image information is binarized by the pseudo halftone, so that the grayscale image information equivalent to the conventional one can be obtained.

In this case, the buffer memory 1005
Since only one (line buffer 1105) is required,
The device cost can be reduced. Further, since the gamma correction unit 1103 and the gamma table 1104 for executing the gamma correction are circuits independent of the A / D conversion unit 1102, the conventional A / D conversion unit 1102 can be used as it is.

Next, another embodiment of the present invention will be described.

In this embodiment, as shown in FIG. 20, the output of the line sensor 1101 is input to the A / D conversion unit 1112, and the output of the A / D conversion unit 1112 is directly input to the line buffer 1105. I am trying.

FIG. 21 is a circuit configuration diagram of the A / D converter 1112. In the figure, input signals are input to the + side input terminals of the plurality of comparators C1 to Cn, respectively. The number n of the comparators C1 to Cn
Is equal to the number of gradations of the extracted multivalued image information.
Each of the adjacent comparators "Ci" and "Ci + 1" has a resistor R1 between the negative input terminal of the comparator C1 and the negative input terminal of the comparator C2, and a resistor R2 between the same terminals of the comparators C2 and C3. A resistor Ri is connected between the negative side input terminals of each. Then, a constant reference voltage Vr is applied to the other end of the resistor Rn connected to the-side input terminal of the comparator Cn.

The outputs of the comparators C1 to Cn are input to the encoder circuit ENC, and the encoder circuit ENC outputs multivalued image information in 6 bits of bits D0 to D5.

If the image processing apparatus of the present embodiment having the above-described structure starts operation, the line sensor 1101
Outputs the image signals read from the original image one line at a time. This image signal is input to the + side input terminals of the comparators C1 to Cn in the A / D converter 1112, respectively.

The input image signal has a high voltage level in "white" and a low voltage level in "black".
The voltage level of the image "white" in this case is set as the reference voltage Vr. The reference voltage Vr is divided by the resistors R1 to Rn and applied to the-side input terminals of the comparators C1 to Cn. In this case, the voltage of the negative side input terminal is high on the side of the comparator Cn,
1 side becomes low.

As a result, when an image signal is input, the comparator C to a specific comparator Ci are turned on according to the voltage level of the image signal, and the comparator "Ci
From +1 "to the comparator Cn are turned off. The encoder circuit ENC outputs a binary digital signal corresponding to the number of comparators that are turned on.

In this embodiment, in this case, the digital signal is set to be output according to the characteristic curve c shown in FIG. 22 with respect to the input level of the image signal. That is,
The characteristic curve d in the figure is a characteristic corresponding to the conventional gamma correction, and in the present embodiment, in the low density region and the high density region of the image signal, it is a non-linear characteristic equivalent to the conventional gamma correction.
Output a digital signal. Further, in those intermediate density regions, a digital signal is output with a linear characteristic in which the output is proportional to the input. Such a characteristic is that each resistor R1.
It can be set by setting each resistance ratio of Rn to a predetermined condition.

The characteristic curves c and d in the figure are opposite in the bending direction to the characteristic curves a and b in FIG. 19, but the maximum value of the voltage level of the image signal is "white". This is because In this case, the "white" is output as the signal value "0" of the digital signal, so that the multivalued image information similar to the case of FIG. 19 is obtained.

The multi-valued image information thus obtained is temporarily stored in the line buffer 1105, and the same processing as that of the above-mentioned embodiment is executed.

As described above, in this embodiment, the A / D converter 1112 sets the input / output characteristics when converting an image signal into a digital signal to a predetermined gamma correction.
The same action and effect as those of the above-described embodiment can be obtained. In this case, the gamma correction unit 1103 and the gamma table 1104, which are provided in the above-described embodiment, are unnecessary, so that the device cost can be further reduced.

In each of the above embodiments, the number of gradations of the multi-valued image information output from the A / D converter 1102 is set to 256, and the number of gradations of the gamma correction unit 1103 is set to 64. As a matter of course, these gradation numbers can be set to arbitrary values.

[0134]

As described above, according to the present invention,
Input multi-valued image information obtained by scanning the original image, and perform gamma correction using the input / output characteristic correction characteristics that are non-linear between low density and high density of the multi-valued image information and linear in the middle density range. Then, the gamma-corrected multi-valued image information is temporarily stored in the buffer memory, and MTF correction and simple binarization are performed on the binary image portion of the stored multi-valued image information. Since the grayscale image portion is binarized by pseudo halftone, the image information of the binary image and the image information of the grayscale image, which are equivalent to the conventional one, can be obtained, respectively, and the buffer memory can be obtained. Since only one is required, there is an effect that the device cost is reduced.

[Brief description of drawings]

FIG. 1 is a schematic diagram for explaining the principle of an error diffusion method.

FIG. 2 is a block diagram showing an image processing apparatus according to an embodiment of the present invention.

FIG. 3 is a circuit diagram showing an example of a bit number conversion unit.

FIG. 4 is a data value correspondence diagram for explaining the operation of the bit number conversion unit in FIG.

FIG. 5 is a circuit diagram showing another example of a bit number conversion unit.

FIG. 6 is a circuit diagram showing still another example of a bit number conversion unit.

7 is a data value correspondence diagram for explaining the operation of the bit number conversion unit of FIG. 6;

FIG. 8 is a circuit diagram showing still another example of the bit number conversion unit.

9 is a data value correspondence diagram for explaining the operation of the bit number conversion unit in FIG. 8;

FIG. 10 is a schematic diagram showing a pixel matrix applied during MTF correction calculation.

FIG. 11 is a block diagram showing an image processing apparatus according to still another embodiment of the present invention.

FIG. 12 is a flowchart showing an example of processing of an MTF mode determination unit.

FIG. 13 is a flowchart showing another example of processing of the MTF mode determination unit.

FIG. 14 is a flowchart showing still another example of processing of the MTF mode determination unit.

FIG. 15 is a block diagram showing an image processing apparatus according to another embodiment of the present invention.

FIG. 16 is a block diagram showing an image processing apparatus according to still another embodiment of the present invention.

FIG. 17 is a block diagram showing an image processing apparatus according to still another embodiment of the present invention.

FIG. 18 is a block diagram showing a configuration example of an MTF correction unit.

FIG. 19 is a graph showing an example of a correction characteristic of gamma correction.

FIG. 20 is a block diagram showing an image processing apparatus according to still another embodiment of the present invention.

21 is a block diagram showing a configuration example of an A / D conversion unit in FIG.

22 is a graph showing the input / output characteristics of the A / D conversion unit shown in FIG.

FIG. 23 is a block diagram showing a conventional example of an image processing apparatus.

[Explanation of symbols]

1 Image input section 2 analog / digital converter 3-bit number converter 3a random number generator 3b AND circuit 4 Error correction unit 5 Binarization part 6 Error calculator 7 Data output section 10 MTF correction unit 11 MTF mode determination unit 15 Halftone mode binarization unit 16 Image area determination unit 17 Mode setting section 18 Image synthesizer 1101 line sensor 1102, 1112 A / D converter 1103 Gamma correction unit 1104 Gamma table 1105 line buffer 1106 MTF correction unit 1106a Pixel extraction circuit 1106b, 1106e Sum operation circuit 1106c Right shift circuit 1106d Left shift circuit 1106f Difference calculation circuit 1107 Binarization unit 1108 Pseudo halftone binarization unit 1109 image area discrimination unit 1110 image composition section 1111 Image output unit C1-Cn comparator R1-Rn resistance ENC encoder circuit

Front page continuation (51) Int.Cl. ⁷ identification code FI H04N 1/409 H04N 1/40 101D (58) Fields investigated (Int.Cl. ⁷ , DB name) H04N 1/40-1/409 H04N 1 / 46 H04N 1/60

Claims (3)

(57) [Claims] 1. A document image in which a binary image and a grayscale image are mixed in one page is read, and gamma correction and MTF correction are performed on a binary image portion of the obtained multivalued image information. In the image processing apparatus, which performs simple binarization by a threshold value, while performing gamma correction and binarization by pseudo halftone for a grayscale image portion, a multivalued image obtained by reading an original image is obtained. Gamma correction means for inputting information and performing gamma correction by the input / output characteristic correction characteristics that are non-linear in the same Log curve in the low density area and high density area of the multi-valued image information and linear in the intermediate density area A buffer memory for temporarily storing the gamma-corrected multivalued image information; a binarized image processing means for performing MTF correction and simple binarization on the stored multivalued image information; Stored in memory The image processing apparatus characterized by relative gray image of the serial multivalue picture information and a gray-scale image processing means for executing binarization by pseudo-halftone. 2. The image processing according to claim 1, wherein the gamma correction means is means for inputting multivalued image information as a digital signal and outputting gamma-corrected multivalued image information as a digital signal. apparatus. 3. The gamma correction means is an analog / digital conversion means for inputting multi-valued image information as an analog signal and outputting the gamma-corrected multi-valued image information as a digital signal. The image processing device described.