JPH0785265A - Image reader - Google Patents

Abstract

PURPOSE:To make it possible to generate a picture signal with resolution higher than the arrangement density of photoelectric conversion elements faithfully to a read picture while preventing the rise of cost by suppressing the arrangement density of the photoelectric conversion elements to a low value. CONSTITUTION:An ACC processing part 1c executes edge emphasizing processing based upon the states of adjacent picture elements in a subscanning direction. An 8-16 conversion part 1e estimates two picture elements based upon signal levels corresponding to respective picture elements in a signal edge- emphasized by the ACC processing part 1c while referring to the states of adjacent picture elements in a main scanning direction and generates a false accurate signal with twice of picture element density.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image reading apparatus which is applied to a facsimile apparatus, an image scanner apparatus or the like and which generates an image signal corresponding to a document.

[0002]

2. Description of the Related Art Generally, an image reading apparatus of this type forms a reflected light image from a document when the light emitted from a light source is applied to the document on a CCD line sensor, and the CCD line sensor outputs a corresponding electric signal. Is generated.

A CCD line sensor has a large number of photoelectric conversion elements arranged in a line at equal intervals on a substrate and outputs an electric signal in which the outputs of the photoelectric conversion elements are serially connected. The output of this CCD line sensor is an analog signal,
Since a facsimile machine or the like generally handles a binary image, the output of each photoelectric conversion element is binarized into either white or black with a predetermined threshold value to generate an image signal.

Therefore, in the conventional image reading apparatus, the resolution is determined by the arrangement density of photoelectric conversion elements. That is, when the photoelectric conversion elements are arranged at a density of 8 pieces / mm, the maximum resolution is 8 pixels / mm.

Therefore, in order to obtain a higher resolution, it is necessary to use a CCD line sensor having a high arrangement density of photoelectric conversion elements. However, the CCD line sensor in which the photoelectric conversion elements are arranged at a high density as described above becomes expensive because the required number of elements increases and the manufacturing becomes difficult.

[0006]

As described above, the conventional image reading apparatus has a problem that the apparatus cost increases if the arrangement density of photoelectric conversion elements is increased to realize reading at high resolution. It was

The present invention has been made in view of the above circumstances, and an object thereof is to suppress the increase in cost by suppressing the arrangement density of photoelectric conversion elements to a low level and to increase the cost of photoelectric conversion elements. It is an object of the present invention to provide an image reading device capable of faithfully generating an image signal having a resolution higher than the arrangement density of the read image.

[0008]

In order to achieve the above object, a first invention identifies a boundary between white and black in a read image based on each pixel signal and corresponds to a pixel located at the boundary. The edge enhancement processing is performed to bring the level of the pixel signal to be closer to a predetermined white level or a predetermined black level, and the boundary is identified only in the pixels arranged in the predetermined first direction (for example, the sub-scanning direction). Based on an edge enhancement processing unit such as an edge enhancement processing unit, and each pixel signal after the edge enhancement processing is performed by the edge enhancement processing unit, the level thereof is set to at least j levels (for example, 3). For example, a pseudo high-definition signal generating circuit is provided with, for example, two comparing circuits and a determining unit including a pseudo high-definition signal generating circuit for determining which one of the classification ranges In any determination means, for each read pixel position, and the classification range determined by the level of a corresponding pixel signal belongs the determining means, said first
Based on at least a part of the k (for example, 2) pixels determined for the pixel signal corresponding to the read pixel position adjacent to the predetermined second direction (for example, the main scanning direction) different from the direction, The corresponding k pixels are determined.

According to the second aspect of the present invention, for each pixel signal corresponding to each of a large number of read pixel positions, it is determined which of at least j (for example, 3) classification ranges the level of which belongs to is preset. For example, a determining unit including two comparing circuits and a pseudo high-definition signal generating circuit is provided, and the determining unit such as the pseudo high-definition signal generating circuit determines the pixel signal corresponding to each of the read pixel positions. The classification range determined by the determination means to which the level belongs and the k pixels determined for the pixel signal corresponding to at least one read pixel adjacent in the predetermined first direction (for example, the main scanning direction). At least one read pixel position adjacent to at least a part thereof in a predetermined second direction (for example, the sub-scanning direction or the sub-scanning direction and the oblique direction) different from the first direction. Based on at least a portion of k pixels determined for the pixel signal corresponding to, and adapted to determine a corresponding k pixels.

[0010]

According to the first aspect of the present invention by taking such means, according to the level of the pixel signal output by one photoelectric conversion element, that is, the brightness level of one read pixel,
k pixels are determined, and the number of pixels is multiplied by k. At this time, the determining unit refers to the pixels adjacent to each other in the predetermined second direction, but the edge emphasis processing unit determines the edge based on the state of only the pixels lined up in the predetermined first direction different from the second direction. Since the enhancement processing is performed and the correlation between the pixels adjacent to each other in the second direction is maintained, the states of the k pixels corresponding to one photoelectric conversion element are accurately determined.

According to the second invention, k pixels are determined according to the level of the pixel signal output from one photoelectric conversion element, that is, the brightness level of one read pixel, and the number of pixels is k times. It is said that Then, at this time, with respect to the determined k pixels, pixels adjacent to the predetermined first direction and a predetermined second direction different from the first direction are referred to, and after the state of the image is recognized in detail, k pixels are determined.

[0012]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is a functional block diagram showing the configuration of a facsimile apparatus configured by applying the image reading apparatus of the present invention.

This facsimile apparatus comprises a reading section 1, a main control section 2, a copy processing section 3, an image memory 4, a communication processing section 5, a recording section 6 and an operation panel 7. The reading unit 1 reads an image formed on a read document and generates corresponding image data. This reading unit 1
The generated image data is given to the copy processing unit 3, the image memory 4, and the communication control unit 5, respectively.

The copy processing section 3 transfers the image information supplied from the reading section 1 to the recording section 6 as necessary (for example, in the copy mode) under the control of the main control section 2. The image memory 4 stores the image data supplied from the reading unit 1 or the communication processing unit 5 as needed under the control of the main control unit 2. Further, the image memory 4 gives the stored image data to the communication processing unit 5 or the recording unit 6 as necessary under the control of the main control unit 2.

The communication processing unit 5 sends the image data supplied from the reading unit 1 or the image memory 4 to the communication line 8 as needed under the control of the main control unit 2. Further, the communication processing unit 5 receives the image data coming through the communication line 8 and gives it to the image memory 4 or the recording unit 6.

The recording unit 6 records an image corresponding to the image data given from the copy processing unit 3, the image memory 4 or the communication control unit 5 on a predetermined recording paper. The operation panel 7 has a large number of input sections such as key switches and a display section such as an LCD, and is used for receiving instructions from the operator and displaying various information to the operator.

The main control unit 2 controls the respective units of the facsimile apparatus as a whole. Now, the reading unit 1
The photoelectric conversion unit 1a, the quantization unit 1b, the edge enhancement processing unit (A
CC processing unit) 1c, binarization processing unit 1d, 8-16 conversion unit 1
e, a pseudo halftone processing unit 1f, a noise removing unit 1g, a reading control unit 1h, and a reading condition information storage unit 1i.

The photoelectric conversion section 1a further includes a document table 20, a light source 21, an optical system 22 and a CCD line sensor 23 as shown in FIG. The document table 20 is made of a transparent glass plate on which a document 24 is placed. The original 24 is placed on the original table 2 as needed by a transport system (not shown).
It is transported over 0. The light source 21 is composed of an LED or the like,
The original 24 is irradiated with light from the lower side of the original table 20. The reflected light from the document 24 at this time is reflected by the CCD line sensor 2 by the optical system 22 including, for example, a rod lens array.
An image is formed on the image 3 and is converted into an electric signal by the CCD line sensor 23.

As shown in FIG. 3, the CCD line sensor 23 comprises a large number of photoelectric conversion elements 23b arranged in a row on a substrate 23a at a predetermined density and at equal intervals, and outputs the output of each photoelectric conversion element 23b. It outputs electrical signals serially connected. The arrangement density of the photoelectric conversion elements 23b is 8 in this embodiment.
It is set to pcs / mm.

The electric signal output from the CCD line sensor 23 is given to the quantizer 1b as an output signal of the photoelectric converter 1a. The quantizer 1b is a well-known one that includes, for example, an A / D converter, a shading correction circuit, and an AGC circuit, and converts the electric signal supplied from the photoelectric converter 1a into an 8-bit digital signal. The shading correction for compensating the variation of the light amount of the light source in the main scanning direction and the AGC process for compensating the temporal change of the light amount of the light source are performed, and then applied to the ACC processing unit 1c.

The ACC processing unit 1c performs edge emphasis processing for emphasizing the black-and-white change point of the image on the digital signal supplied from the quantization unit 1b. In this edge enhancement processing, a numerical value obtained by multiplying a second derivative amount by a level corresponding to a pixel to be processed (converted pixel) and a level corresponding to a predetermined reference pixel located around the processed pixel by an appropriate coefficient is used. It is the process of subtracting from. The ACC processing unit 1c processes the digital signal after the edge enhancement processing into a binarization processing unit 1d, an 8-16 conversion unit 1e, and a pseudo halftone processing unit 1f.
Give to each.

The binarization processing unit 1d simply binarizes the 8-bit digital signal provided from the ACC processing unit 1c as black if the value is smaller than a predetermined slice level and as white otherwise. The binarization processing unit 1d gives a binary signal (hereinafter, referred to as a simple binary signal) obtained by performing binarization to the noise removing unit 1g.

The 8-16 conversion unit 1e performs a process described in detail later on the signal corresponding to 8 dots / mm provided from the ACC processing unit 1c to perform a binary signal corresponding to 16 dots / mm (hereinafter referred to as a pseudo high definition signal). (Referred to as ") and supplied to the noise removing unit 1g.

The pseudo halftone processing unit 1f is the ACC processing unit 1
Based on the digital signal given from c, a pseudo-halftone signal according to a predetermined pseudo-halftone method is generated and given to the noise removing unit 1g.

The binarization processing unit 1d and the 8-16 conversion unit 1e
The pseudo halftone processing unit 1f operates alternatively under the control of the reading control unit 1h. The noise eliminator 1g includes the noise eliminator 1 in various binary signals provided from the binarization processor 1d, the 8-16 converter 1e or the pseudo halftone processor 1f.
The noise generated by the signal processing in the stage before g is removed. Specifically, since the ACC process emphasizes a slight level fluctuation of the signal and may generate a dirty image, referring to the signal before the ACC process, clear white becomes white and clear Force black to black. The noise removing unit 1g outputs the processed signal as image data.

The reading control unit 1h operates under the control of the main control unit 2 and totally controls each unit of the reading unit 1. The reading condition information storage unit 1i is for storing various information necessary for the reading control unit 1h to control each unit and realize the reading operation.

On the other hand, the recording section 6 includes a thermal print head 6a, a normal recording processing section 6b, and a pseudo high-definition recording processing section 6.
c, a recording control unit 6d and a recording condition information storage unit 6e. As shown in FIG. 4, the thermal print head 6a includes a head unit 40 and a drive circuit 41.

The head portion 40 has a large number (n) of heat generating resistors 40b having a substantially parallelogram shape on an insulating substrate 40a made of, for example, ceramics or alumina.
One-dimensional arrays are arranged in parallel at equal intervals with a density of mm. Further, on the insulating substrate 40a, one common lead electrode 40 is connected to one end of each heating resistor 40b.
c is provided. Also, n individual lead electrodes 40 are connected to the respective other ends of the heating resistors 40b.
d are provided respectively.

The drive circuit 41 energizes each heating resistor 40b in response to a control signal given from the normal recording processing section 6a or the pseudo high definition recording processing section 6b. The drive circuit 41 is integrally formed with the head unit 40.

The thermal print head 6a having the above construction
Are arranged as shown in FIG. That is, the thermal print head 6a is arranged such that the heating resistor 40b faces the platen roller 50 and the arrangement direction of the heating resistor 40b is along the longitudinal direction of the platen roller 50, and the spring 51 presses the platen roller 50. Has been done. An ink film 52 and a recording paper 53 are inserted between the thermal print head 6a and the platen roller 50 in a state of overlapping each other. The ink film 52 is wound around the sprocket 54 in a roll shape. Furthermore, the tip of the ink film 52 is fixed to the sprocket 55, and the ink film 52 is wound around the sprocket 55. The recording paper 53 is conveyed in the direction of the arrow in the figure as the platen roller 50 rotates. Reference numerals 56 and 57 denote guide shafts, which keep the ink film 52 with respect to the thermal print head 6a in a predetermined state (the state shown in the drawing).

The image data supplied from the copy processing unit 3, the image memory 4 or the communication processing unit 5 is input to the normal recording processing unit 6a and the pseudo high definition recording processing unit 6b, respectively.

The normal recording processing section 6b generates a control signal for recording an image having a density of 8 dots / mm in the main scanning direction by the thermal print head 6a based on the supplied image data and supplies it to the thermal print head 6a. .

The pseudo high-definition recording processing unit 6c generates a control signal for recording an image having a density in the main scanning direction of 16 dots / mm on the thermal print head 6a based on the supplied image data, and the thermal print head 6
give to a.

The normal recording processing unit 6a and the pseudo high-definition recording processing unit 6b operate alternatively under the control of the recording control unit 6d. The recording control unit 6d operates under the control of the main control unit 2 and controls each unit of the recording unit 6 as a whole.

The recording condition information storage unit 6e includes a recording control unit 6e.
d is for storing various kinds of information necessary for controlling each unit and realizing the recording operation. Next, the operation of the facsimile apparatus configured as above will be described.

(1) Reading Operation First, when the operation panel 7 is instructed to execute facsimile transmission or copy, the main control section 2 receives this instruction and instructs the reading control section 1h to read the original. In response to this, the reading control unit 1h operates each unit of the reading unit 1. However, the binarization processing unit 1d, the 8-16 conversion unit 1e, and the pseudo-halftone processing unit 1f conform to the mode set with the partner terminal according to the designation on the operation panel 7 during facsimile transmission, and At the time of copying, any one of them is operated according to the mode designated on the operation panel 7. Specifically, this facsimile machine has "standard mode", "fine mode", "high definition mode", "super high definition mode", "standard halftone mode", "fine halftone mode", and "high definition halftone mode".
And 8 reading modes of "ultra high definition halftone mode", and when any of "standard mode""finemode" and "high definition mode" is selected, the binarization processing unit 1d , 8-16 conversion unit 1e when "super high definition mode" is selected, and "standard half tone mode""fine half tone mode""high definition half tone mode" and "super high definition half tone mode" When is selected, the pseudo halftone processing unit 1f is operated.

The above reading modes are under the following conditions. "Standard mode": 8 dots / mm x 3.85 lines / mm, binary black and white. "Fine mode": 8dots / mm x 7.7lines / mm, black and white 2
value.

"High-definition mode": 8 dots / mm x 15.4
lines / mm, black and white binary. "Ultra high definition mode": 16 dots / mm equivalent x 15.4 line
s / mm, black and white binary. "Standard halftone mode": 8dots / mm x 3.85lines / m
m, pseudo halftone.

"Fine halftone mode": 8 dots / mm x7.
7 lines / mm, pseudo halftone. "High-definition halftone mode": 8 dots / mm x 15.4 lines /
mm, pseudo halftone. "Ultra high definition halftone mode": 16 dots / mm equivalent x15.
4 lines / mm, pseudo halftone.

The photoelectric conversion unit 1a turns on the light source 21 to illuminate the original document 24, and also conveys the original document 24 at a predetermined feed pitch by a conveyance system (not shown). The feed pitch is a pitch corresponding to the density in the sub-scanning direction according to the reading mode. Then, the reading by the CCD line sensor 23 is repeated at a predetermined timing synchronized with the feeding of the document 24. The electric signal generated by the CCD line sensor 23 in this way is quantized by the quantizer 1b.
Is converted to an 8-bit digital signal at
After undergoing edge enhancement processing in the ACC processing circuit 1c,
It is input to the binarization processing unit 1d, the 8-16 conversion unit 1e, and the pseudo halftone processing unit 1f.

(1-1) When the Binarization Processing Unit 1d is Operating In this case, the binarization processing unit 1d simply determines whether each pixel is white or black based on the digital signal. The simple binary signal shown is generated. Specifically, consider a case where the positional relationship between the CCD line sensor 23 and the black image to be read is in the state shown in FIG. 6A. In FIG. 6A, the hatched figure shows the position of the black image to be read.

In this state, the digital signal output from the ACC processing section 1c corresponds to the signal shown in FIG. 6 (b). Although FIG. 6B shows the analog state for convenience, the output of the ACC processing unit 1c is actually the level at the position indicated by “x” in FIG. 6B expressed by 8 bits. ing.

Then, in the binarization processing unit 1d, as shown in FIG.
By binarizing with a predetermined threshold value shown in FIG.
A simple binarized signal as shown in (c) is obtained. Here, focusing on the portion indicated by A in FIG. 6A, since a black image is applied to about half of the corresponding photoelectric conversion element 23b, its output level is an intermediate level between the black level and the white level. It has become. However, since this level is higher than the threshold value, it is set to the black level after binarization.

The simple binarization signal generated by the binarization processing section 1d in this manner is for the heating resistor 40 per line.
It contains the same number (n) of pixel data as b,
It corresponds to ts / mm.

(1-2) When the 8-16 conversion section 1e is operating In this case, the 8-16 conversion section 1e outputs a digital signal equivalent to 8 dots / mm from one pixel information to two pixel information. Is converted into a pseudo high definition signal equivalent to 16 dots / mm. This conversion process is performed as follows.

It is assumed that the positional relationship between the CCD line sensor 23 and the black image to be read is as shown in FIG. 7 (a). Note that, in FIG. 7A, a hatched figure shows a black image to be read.

In this state, the digital signal output from the ACC processing section 1c corresponds to the signal shown in FIG. 7 (b). Although FIG. 7B is shown in an analog state for convenience, the output of the ACC processing unit 1c is actually shown in FIG. 7B.
The level at the position indicated by "x" is represented by 8 bits.

First, in the 8-16 conversion section 1e, the signal level corresponding to each photoelectric conversion element 23b is set to "high brightness" based on the predetermined first threshold value and second threshold value (first threshold value> second threshold value) which are different from each other. Classify into either "medium brightness" or "low brightness". Here, "high brightness" is a range equal to or higher than the first threshold. The “medium brightness” is a range equal to or higher than the second threshold and lower than the first threshold. The "low brightness" is a range lower than the second threshold. Now, in the portion indicated by B in FIG. 7B, the signal level is equal to or higher than the first threshold value, and therefore, it is determined to be “high brightness”. In the portion indicated by C in FIG. 7B, since the signal level is lower than the second threshold value, "low brightness"
It is determined that The portion indicated by D and E in FIG.
In the portion indicated by, since the signal level is equal to or higher than the second threshold and lower than the first threshold, it is determined to be “medium luminance”.

Based on the "high brightness", "medium brightness" and "low brightness", two adjacent pixels (reference pixels) that have already been generated are referred to and two pixels are generated in the pattern shown in FIG. To do. That is, regarding “high brightness”, both of the two pixels are determined to be white pixels regardless of the state of the reference pixel. Regarding “low brightness”, both two pixels are determined to be black pixels regardless of the state of the reference pixel.
Regarding "medium brightness", the reference pixels are "white, white", "white,
When it is either “black” or “black, white”, the two pixels are sequentially “white” and “black”, and the reference pixel is “black,
When it is “black”, the two pixels are sequentially “black” and “white”.
Is decided.

Thus, the pixels in the portion indicated by B in FIG. 7B are each converted into two white pixels. Figure 7 (b)
Pixels indicated by C are converted into two black pixels. In the portion indicated by D in FIG. 7B, since the reference pixels are both “black”, the pixels in the portion indicated by D are converted into two pixels in the order of “black” and “white”. And
Since the reference pixel is “white, white” or “white, black” in the portion indicated by E in FIG. 7B, the pixel in the portion indicated by E is two pixels in the order of “white” and “black”. Is converted to.
As a result, a pseudo high definition signal corresponding to the pixel array as shown in FIG. 7C is generated.

(1-3) When the pseudo halftone processing section 1f is operating In this case, each pixel is either "white" or "black" based on the digital signal in the pseudo halftone processing section 1f. However, a pseudo-halftone signal corresponding to an image that looks like a halftone image when viewed as a whole image is generated according to a well-known pseudo-halftone method.

When the reading mode is the "ultra-high definition halftone mode", the pseudo halftone processing section 1f converts one pixel into two pixels by repeating the data of each pixel twice. Obtain a signal equivalent to 16 dots / mm. Then, based on this signal, a pseudo-halftone signal is generated according to a well-known pseudo-halftone method.

The binarization processing unit 1 is as described above.
Each signal generated by the d, 8-16 conversion unit 1e or the pseudo halftone processing unit 1f is converted into AC by the noise removal unit 1g.
By forcing the clear white into white and the clear black into black with reference to the signal before the C processing, the noise removing unit 1
After the noise generated by the signal processing in the stage preceding g is removed, it is output as image data.

(2) Facsimile transmission operation When the operator instructs the execution of facsimile transmission on the operation panel 7, the communication processing section 5 calls the designated party under the control of the main control section 2, After the above is established, preparatory processing such as mode setting and communication training is performed by exchanging predetermined control information with the partner terminal. When the preparation process is completed, the image data generated by the reading unit 1 as described above is received from the reading unit 1 directly or via the image memory 4 and subjected to redundancy suppression coding or modulation. It is sent to the communication line 8.

When the transmission of all the image information is completed, the communication control unit 5 performs confirmation processing with the partner terminal and then releases the communication line 8. (3) Facsimile reception operation On the other hand, when a call is made through the communication line 8, the communication processing unit 5 captures the line in response to this, and then the mode setting by exchanging predetermined control information with the partner terminal is performed. Perform preparation processing such as communication training. Then, after the preparation process is completed, the image data that arrives via the communication line 8 is received, and demodulation and decoding of the redundancy suppression code are performed. Then, the communication processing unit 5 gives the received image data to the recording unit 6 in the normal reception mode, and gives it to the image memory 5 in the memory reception mode.

When the reception of all the image information is completed, the communication control section 5 performs confirmation processing with the partner terminal and then releases the communication line 8. (4) Copy Operation When the operator instructs execution of copy processing on the operation panel 7, the main control unit 2 operates the copy processing unit 3 and then causes the reading unit 1 to read a document. When the reading unit 1 receives the image data generated as described above, the copy processing unit 3 transfers the image data to the recording unit 6, and causes the recording unit 6 to record the corresponding image.

(5) Recording operation Copy operation, facsimile reception or image memory 4
When the recording of the image corresponding to the image data stored in is designated by the operator on the operation panel 7,
The main controller 2 instructs the recording controller 6d to start the recording operation. In response to this, the recording control unit 6d causes the normal recording processing unit 6d to
b or the pseudo high-definition recording controller 6c,
The operation is performed according to the mode set with the partner terminal during the recording operation accompanying the facsimile reception, and according to the mode designated on the operation panel 7 during copying. Specifically, the present facsimile apparatus has a "normal recording mode" and a "pseudo high-definition recording mode". In the "recording mode", the pseudo high-definition recording processing section 6c is operated.

In the "normal recording mode", based on the image data whose main scanning direction density is 8 dots / mm, the thermal print head 6a in which the array density of the heating resistors 40b is 8 pieces / mm is set in the main scanning direction. In this mode, an image with a pixel density of 8 dots / mm is recorded. The density in the sub-scanning direction in this "normal recording mode" is 3.85 lines / mm,
Either 7.7 lines / mm or 15.4 lines / mm,
This is determined by the image data. The "pseudo high-definition recording mode" is based on image data whose main scanning direction density is 16 dots / mm, and the thermal printing head 6a in which the array density of the heating resistors 40b is 8 pieces / mm is applied to the pixels in the main scanning direction. In this mode, an image with a density of 16 dots / mm is recorded. The density in the sub-scanning direction in this "pseudo high-definition recording mode" is 15.4 lines / mm.

(5-1) When the normal recording processing unit 6b is operating In this case, the normal recording processing unit 6b includes the copy processing unit 3,
Image data having a density in the main scanning direction of 8 dots / mm 2 is supplied from either the image memory 4 or the communication processing unit 5. Based on this image data, the normal recording processing unit 6b
The heating resistor 40b corresponding to the pixel of "black" is energized, and the heating resistor 40b corresponding to the pixel of "white" is energized.
The print data and the control signal for controlling the drive circuit 41 of the thermal print head 6a so as to be energized are generated.

Driving circuit 4 for thermal print head 6a
In No. 1, the heating resistor 40b corresponding to the pixel of "black" is energized only in accordance with the print data and the control signal given from the print processing unit 6b, and a predetermined amount of energy is applied to the heating resistor 40b. . At this time, the amount of energy applied to the heating resistor 40b is determined by the heating resistor 40a.
Is the amount of energy that can generate heat above the temperature required to form "black" pixels over the entire surface.

At this time, the recording paper 53 has a sub-scanning direction density (3.85 lines / mm, 7.7 lines / mm) corresponding to the image data.
mm, 15.4 lines / mm).

(5-2) When the pseudo high definition recording processing section 6c is operating In this case, the pseudo high definition recording processing section 6c includes any one of the copy processing section 3, the image memory 4 and the communication processing section 5. Gives image data having a density in the main scanning direction of 16 dots / mm. The pseudo high-definition recording processing unit 6c uses the 1st and 2nd, 3rd and 4th ..., [n-
The 1] th and the nth are respectively paired, and based on the state of the pair of pixels (original pixels), the density of one recording pixel is determined to be one of three levels from level 0 to level 2 according to an agreement described later. To do. Each level is level 0
<Level 1 <Level 2

FIG. 9 is a diagram schematically showing an arrangement for determining the recording density. In FIG. 9, white parts of the original pixels indicate “white” pixels, and black parts indicate “black” pixels.

As shown in this figure: When both two original pixels are "white" (black ratio 0 /
Level 2 in 2). Level 1 when only one of the two original pixels is "black" (black ratio 1/2).

Level 2 when the two original pixels are both "black" (black ratio 2/2). It is arranged as follows. In this way, by converting the two original pixels into one recording pixel, the number of pixels becomes 1/2 of the original number, which corresponds to 8 dots / mm 2. Then, the pseudo high-definition recording processing unit 6c sets the heating resistor 40b corresponding to each recording pixel to a predetermined energy amount corresponding to the level of each recording pixel based on the recording data composed of the array of recording pixels thus determined. The recording data and the control signal for controlling the drive circuit 41 of the thermal print head 6a so as to be energized are generated.

Driving circuit 4 for the thermal print head 6a
In No. 1, the heating resistor 40b corresponding to the pixel of level 0 is not energized according to the recording data and the control signal supplied from the pseudo high definition recording processing unit 6c. Further, a predetermined energy amount E _S is applied to the heating resistor 40b corresponding to the pixel of level 1. Further, a predetermined energy amount E _L set to be larger than E _S is applied to the heating resistor 40b corresponding to the pixel of level 2.

At this time, the ink film 52 and the recording paper 53 are conveyed at a pitch corresponding to 15.4 lines / mm corresponding to the image data. Since the heating resistor 40b has a parallelogram shape, the current distribution when a voltage is applied is biased as shown in FIG. FIG. 10 shows the results of the current distribution in the heating resistor 40b obtained by numerical analysis using the boundary element method, where the black dots are the measurement points.
The direction of the line indicates the direction of the current at the measuring point, and the length of the line indicates the magnitude of the current at the measuring point.

As can be seen from this figure, the current increases toward the central portion of the heating resistor 40b. Here, the calorific value at a certain point in the heating resistor 40b is
It is proportional to the square of the amount of current at that position. Therefore, the amount of heat generated is large in the central portion of the heating resistor 40b.

Now, in order to form "black" pixels on the recording paper, a certain amount of heat or more is required. Therefore, when the voltage applied to the heating resistor 40b is small, the
A "black" pixel is formed by heat generation in a range indicated by 1a (hereinafter referred to as an effective heat generation range). As the applied voltage is increased, as shown by 61b and 61c in the figure,
The effective heat generation range expands. Therefore, the heating resistor 40
By changing the amount of energy applied to b,
The size of the “black” pixel can be changed.

In the pseudo high definition recording processing section 6c, the energy amounts E _S and E _L are set as follows. E _S is set to a value necessary to form an effective heat generation range 71 having a size comparable to the size of one pixel at 16 dots / mm 2 in a part of the heat generation resistor 40b as shown in FIG. . Further, E _L is set to a value required to form the effective heat generation range 72 in the entire area of the heating resistor 40b as shown in FIG. The amount of energy supplied to the heating resistor 40b can be changed by the voltage applied to the heating resistor 40b and the energization time.

Thus, in the case where the image (original image) represented by the image data given to the recording unit 1 is the one in which the "black" pixels 81a and 81b are discretely located as shown in FIG. 13 (a). , Only one of the two corresponding original pixels is a “black” pixel, and a part of the heating resistor 40b (center and right in FIG. 13) is displayed on the original image as shown in FIG. 13 (b). 1
An effective heat generation area 82a having a size comparable to the size of a pixel,
82b is formed. Therefore, the recorded "black" pixels are discretely formed like the original image.

On the other hand, when the original image is solidified as shown in FIG. 14A, the "black" pixels 83a, 83b, 83c,
When 83d is located, the heating resistor 40 whose corresponding two original pixel data are both "black" pixels
An effective heat generation area 84a is formed in the entire area of b (center in FIG. 14) as shown in FIG. 14 (b). Also, at both ends of the image, one of the two corresponding original pixel data
Heating resistor 40b (FIG. 14) in which only one is a “black” pixel
Since the same voltage as in the case of FIG. 13 is applied to the left and right), the effective heating area generated by the applied voltage is shown in FIG.
The areas indicated by 84b and 84c are similar to those in the case of 3. However, the effective heating area 84 of the adjacent heating resistor 40b is
Since "a" extends over the entire area of the central heating resistor 40b, it is affected by the heat of the heating resistor 40b and is adjacent to the heating resistor 40b.
Effective heating regions 84d and 84e are generated (on the left and right in FIG. 14). Thus, the effective heat generating areas of the heat generating resistors 40b are combined, and a continuous large black image is recorded. In addition, since the effective heat generating area of the adjacent heat generating resistors (not shown) does not extend to the entire area of the portion of the heat generating resistor 40b corresponding to both ends of the image corresponding to the outside of the image,
As shown in FIG. 14B, it does not become an effective heat generation area.
Therefore, an image faithful to the original image shown in FIG. 14A is recorded.

The above is the outline of the configuration and operation of the facsimile apparatus according to the present embodiment. Next, the main part of the present invention will be described in more detail. Now, the photoelectric conversion element 23b
The arrangement density is 8 pieces / mm, but the principle of being able to generate image data equivalent to 16 dots / mm is as described in detail above. Hereinafter, some embodiments will be described for enabling good image reading based on this principle.

(First Embodiment) When the 8-16 converter 1e generates a pseudo high definition signal, the ACC by the ACC processor 1d
The processing may adversely affect the image quality of the image corresponding to the pseudo high definition signal.

First, the details of the ACC process and the influence of the ACC process on the generation of the pseudo high definition signal will be described. Figure 15
Taking the case of reading the image shown in (a) as an example, the digital signal output from the quantizing unit 1b is represented by information as shown in FIG. 15 (b). FIG. 15 (a)
15a is a low-density black image in white, 15b is a high-density black image in white, and 15c is a low-density black image in black.
15d is a white image in black. The black image 15a
The black image 15c and the black image 15c have the same density. Figure 15
In (b), each square corresponds to the output of one photoelectric conversion element, and the number of its name corresponds to the brightness level. In addition,
The output of the quantizer 1b is 8 bits, and the brightness level is actually shown in 256 steps, but here, for convenience, it is shown in 6 steps of level 0 to level 5.

As shown in FIG. 15B, the black image 15a
Although the black image 15c and the black image 15c originally have the same density, the level after quantization is different from the level 3 in the black image 15a and the level 2 in the black image 15b. this is,
This is because the black image 15a is read whiter than the black image 15b due to the influence of the white surroundings. Therefore, in each pixel shown in FIG. 15B, levels 0 to 2 are black (level 0) and levels 3 to 5 are white (level 5).
Considering the case where binarization is performed by the binarization processing unit 1d, the black image 1
5a is changed to a white image, and black image 15b is changed to a high-density black image. As shown in FIG.
5a and the black image 15b are assimilated into the surroundings and disappear.

Therefore, in the ACC processing unit 1c, as shown in FIG. 17, a pixel to be processed (referred to as a conversion pixel) b
, And by using pixels (referred to as reference pixels) a, c, d, and e adjacent to the conversion pixel b, b ′ = b− {α (a + c−2b) + β (d + e−2b)} ... (1) (where α and β are predetermined coefficients) and the calculation result b ′ is used as the data of the conversion pixel b. Note that this processing is performed as ACC processing by setting all pixels as converted pixels and performing them.

By such processing, the pixels located at the edges of the black image and the white image are emphasized as shown in FIG. 18A, and the loss of the image is prevented as shown in FIG. 18B. . Note that FIG. 18A is an example when the counts α and β are 3/4.

The above is the ACC processing, which is extremely effective for improving the image quality when image data is generated by the simple binarization processing. However, when the pseudo high definition signal is generated, since the data in the portions indicated by D and E in FIG. 7 have changed, the original state of the pixel cannot be estimated. Specifically, for example, based on the data after the above-described ACC processing is performed by the ACC processing unit 1c on the output data of the quantization unit 1b as shown in FIG. When the pseudo high-definition signal is generated in, the pseudo high-definition signal corresponds to a rough image as shown in FIG.

Therefore, in the present embodiment, when the 8-16 conversion unit 1e is operating, the ACC processing unit 1c uses only the pixels adjacent to the conversion pixel in the sub-scanning direction (d and e in FIG. 17) as reference pixels. Then, the ACC process is performed by performing the following operation: b ′ = b−γ (d + e−2b) (2) That is, the coefficient α in the equation (1) is set to “0” to remove the information in the main scanning direction, and the coefficient β is changed to another appropriate coefficient γ. No processing is done.

As a result, the intermediate level information remains in the signal after the ACC processing, and it becomes possible to favorably generate the pseudo high definition signal in the 8-16 conversion section 1e. Specifically, based on the data after the above-mentioned ACC processing of the present embodiment is applied to the output data of the quantizing unit 1b as shown in FIG. When a high-definition signal is generated, the pseudo high-definition signal corresponds to a high-definition image as shown in FIG. 19 (b).

Thus, according to this embodiment, high-definition reading can be realized well. Second Embodiment On the other hand, in the basic operation relating to the generation of the pseudo high definition signal in the 8-16 conversion unit 1e described above, two already generated pixels that are adjacent in the main scanning direction are used as reference pixels. There is. Therefore, the state of the image in the main scanning direction can be estimated well, and a signal faithful to the read image can be generated, but it is difficult to judge the state of the line along the sub scanning direction, for example. However, depending on the image state of the read document, there is a possibility that the image may not be faithful in the sub-scanning direction.

Therefore, an embodiment capable of generating a pseudo high-definition signal faithfully showing in the sub-scanning direction will be described below. FIG. 21 is a block diagram showing a specific configuration of the 8-16 conversion unit 1e. 8-16 converter 1
e is composed of comparison circuits 91 and 92, a threshold value generator 93, a pseudo high definition signal generation circuit 94, a conversion table 95 and a reference pixel data memory 96.

The comparing circuit 91 compares the level of each pixel in the output signal of the ACC processing section 91 with the first threshold value generated by the threshold value generating section 93, and when the pixel level is higher, outputs WH. Set to "H" level.

The comparing circuit 92 compares the level of each pixel in the output signal of the ACC processing section 91 with the second threshold value generated by the threshold value generating section 93, and when the pixel level is higher, outputs BL. Set to "H" level.

The threshold value generating circuit 93 has a predetermined first threshold value indicating the upper limit of "medium brightness" as shown in FIG. 7 and a predetermined second threshold value indicating the lower limit of "medium brightness" as shown in FIG. Occurred and output.

The pseudo high-definition signal generation circuit 94 refers to the output signals of the comparison circuits 91 and 92, refers to the conversion table 95 and the reference pixel data memory 96 as necessary, and outputs 1 of the output signal of the ACC processing unit 91. The pixel data is converted into two recording pixels, and a pseudo high-definition signal indicating the presence or absence of an image point of each recording pixel at the “H” level and the “L” level is generated and output.

The conversion table 94 stores information on conversion patterns that the pseudo high-definition signal generating circuit 94 refers to when performing conversion processing. Reference pixel data memory 96
Of the pseudo high-definition signal output from the pseudo high-definition signal generation circuit 94 stores the data of the recording pixel to be used as a reference pixel when performing the subsequent conversion process in the pseudo high-definition signal generation circuit 94. It is for.

Next, the 8-16 conversion unit 1 configured as described above
The operation of e will be described. First, if the respective outputs of the comparison circuits 91 and 92 are WH = “H” and BL = “H”,
The level of the pixel currently input is the first threshold value and the second threshold value.
Since both threshold values are exceeded, it can be seen that the brightness is “high brightness”. If WH = "L" and BL = "H",
Since the level of the currently input pixel is equal to or higher than the second threshold value and lower than the first threshold value, it can be seen that the pixel has “medium brightness”. Furthermore, if WH = “L” and BL = “L”, the level of the currently input pixel is lower than both the first threshold value and the second threshold value, so it can be seen that the brightness is “low brightness”. In this way, the outputs WH of the comparison circuits 91 and 92,
Pixels currently input based on BL are "high brightness"
It can be classified into "medium brightness" and "low brightness".

Therefore, in the pseudo high definition signal generating circuit 94,
Based on the outputs WH and BL of the comparison circuits 91 and 92, WH =
When “H” and BL = “H”, two recording pixels are both set to black pixels, and when WH = “L” and BL = “L”, both recording pixels are set to white pixels.
At this time, the conversion table 95 and the reference pixel data memory 96 are not referred to.

When WH = “L” and BL = “H”, the pseudo high-definition signal generating circuit 94 converts the conversion table 9
5 and the reference pixel data memory 96 are referred to, and two recording pixels are determined.

Here, in the present embodiment, as shown in FIG. 22A, the recording pixels (referred to as conversion pixels) P1 and P2 to be determined are adjacent to the conversion pixel P1 in the sub-scanning direction. The recording pixel b adjacent to the recording pixel a and the conversion pixel P1 in the sub-scanning direction is used as a reference pixel, and the respective data of the recording pixel a and the recording pixel b are read from the reference pixel data memory 96. Then, the conversion table 95 is referred to based on the data of these two reference pixels.

The conversion table 95 stores conversion patterns as shown in FIG. 22B which are set in consideration of the statistical properties of the image. In the figure, “0” indicates a white pixel and “1” indicates a black pixel.

Thus, according to this embodiment, the print pixels are determined in consideration of the states of the print pixels adjacent in the sub-scanning direction. As a result, it is possible to generate a pseudo high-definition signal that is faithful to, for example, a line along the sub-scanning direction.

(Third Embodiment) In the second embodiment, since the pixel adjacent to the conversion pixel P2 in the sub-scanning direction is not referred to, the pseudo high definition signal more faithful to the read image is generated. Then, the determination of the image state in the sub-scanning direction is insufficient.

Therefore, in this embodiment, as shown in FIG. 23A, the recording pixel a and the conversion pixel P2 which are adjacent to the conversion pixels P1 and P2 in the sub-scanning direction of the conversion pixel P1 are in the sub-scanning direction. The recording pixel b adjacent to the recording pixel b and the recording pixel c adjacent to the conversion pixel P1 in the main scanning direction as reference pixels,
The respective data of the recording pixel a, the recording pixel b, and the recording pixel c are read from the reference pixel data memory 96. Then, the conversion table 95 is referred to based on the data of these three reference pixels.

The conversion table 95 stores conversion patterns set as shown in FIG. 23B in consideration of the statistical properties of the image. In the figure, “0” indicates a white pixel and “1” indicates a black pixel.

Thus, according to this embodiment, the conversion pixel P
The recording pixels are determined in consideration of the respective states of the recording pixels adjacent to each of P1 and P2 in the sub-scanning direction. This makes it possible to generate a pseudo high-definition signal that is more faithful to the read image than in the first embodiment.

(Fourth Embodiment) On the other hand, in the second and third embodiments, it is difficult to judge the state of the diagonal line, and the diagonal line may not be faithful depending on the image state of the read document. It may disappear.

Therefore, in the present embodiment, as shown in FIG. 24A, the recording pixel a and the conversion pixel P1 which are adjacent to the conversion pixel P1 and P2 diagonally to the left of the conversion pixel P1 are in the sub-scanning direction. A recording pixel b adjacent to the conversion pixel P2, a recording pixel c adjacent to the conversion pixel P2 in the sub-scanning direction, and a recording pixel d adjacent to the conversion pixel P1 in the main scanning direction as reference pixels. The respective data of the recording pixel b, the recording pixel c, and the recording pixel d are read from the reference pixel data memory 96. Then, the conversion table 95 is referred to based on the data of the four reference pixels.

The conversion table 95 stores conversion patterns as shown in FIG. 24 (b) set in consideration of the statistical properties of the image. In the figure, “0” indicates a white pixel and “1” indicates a black pixel.

Thus, according to this embodiment, the conversion pixel P1
On the other hand, the recording pixels are determined in consideration of the respective states of the recording pixels that are diagonally adjacent to each other. This makes it possible to generate a pseudo high-definition signal that is faithful to, for example, an oblique line and is more faithful to the read image than the second embodiment.

(Fifth Embodiment) By the way, in the fourth embodiment, since the pixel adjacent to the conversion pixel P2 in the diagonally upper right direction is not referred to, the pseudo high definition signal more faithful to the read image is generated. In this case, the judgment of the image state in the oblique direction is insufficient.

Therefore, in the present embodiment, as shown in FIG. 25A, the recording pixel a and the conversion pixel P1 which are adjacent to the conversion pixels P1 and P2 diagonally to the left of the conversion pixel P1 are in the sub-scanning direction. A recording pixel b adjacent to the conversion pixel P2, a recording pixel c adjacent to the conversion pixel P2 in the sub-scanning direction, a recording pixel d adjacent to the conversion pixel P2 diagonally to the upper right, and a conversion pixel P1 adjacent to the conversion pixel P1 in the main scanning direction. The recording pixel e being recorded is used as a reference pixel, and the respective data of the recording pixel a, the recording pixel b, the recording pixel c, the recording pixel d, and the recording pixel e are read from the reference pixel data memory 96. Then, the conversion table 95 is referred to based on the data of the five reference pixels.

The conversion table 95 stores conversion patterns as shown in FIG. 25 (b) set in consideration of the statistical properties of the image. In the figure, “0” indicates a white pixel and “1” indicates a black pixel.

Thus, according to this embodiment, the conversion pixel P
The recording pixels are determined in consideration of the respective states of the recording pixels that are diagonally adjacent to P1 and P2. As a result, it is possible to generate a pseudo high-definition signal that is more faithful to the read image than the above-described fourth embodiment.

The present invention is not limited to the above embodiments. For example, in each of the above embodiments, the photoelectric conversion element 2
CCD line sensor 2 with 3b array density of 8 / mm
Although the case where image data corresponding to an image corresponding to 16 dots / mm 3 is generated using 3 is illustrated, the invention is not limited to this.

In each of the above embodiments, the image reading device of the present invention is applied to a facsimile device, but it may be an independent image reading device or applied to a device other than the facsimile device. . In addition, various modifications can be made without departing from the scope of the present invention.

[0109]

According to the first aspect of the present invention, the boundary between white and black in the read image is identified based on each pixel signal, and the level of the pixel signal corresponding to the pixel located at the boundary is set to a predetermined white level or a predetermined level. The edge enhancement processing is performed based on the state of only the pixels lined up in a predetermined first direction (for example, the sub-scanning direction). Processing means,
For each pixel signal that has been subjected to edge enhancement processing by the edge enhancement processing means, it is determined which of at least j stages (for example, 3 stages) of classification ranges the level of which is preset, for example, 2 And a determining unit including a pseudo high-definition signal generating circuit. For example, in the determining unit such as the pseudo high-definition signal generating circuit, it is determined that the level of the corresponding pixel signal belongs to each read pixel position. The classification range determined by the determination means and k (for example, 2) determined for pixel signals corresponding to read pixel positions adjacent to each other in a predetermined second direction (for example, the main scanning direction) different from the first direction. The corresponding k pixels are determined based on at least a part of the pixels.

The second aspect of the present invention determines, for each pixel signal corresponding to a large number of read pixel positions, to which of at least j stages (for example, 3 stages) of classification ranges the levels are preset. For example, a determining unit including two comparing circuits and a pseudo high-definition signal generating circuit is provided, and the determining unit such as the pseudo high-definition signal generating circuit determines the pixel signal corresponding to each of the read pixel positions. The classification range determined by the determination means to which the level belongs and the k pixels determined for the pixel signal corresponding to at least one read pixel adjacent in the predetermined first direction (for example, the main scanning direction). At least one read pixel position adjacent to at least a part thereof in a predetermined second direction (for example, the sub-scanning direction or the sub-scanning direction and the oblique direction) different from the first direction. Based on at least a portion of k pixels determined for the pixel signal corresponding to, and adapted to determine a corresponding k pixels.

As a result, it is possible to generate an image signal having a higher resolution than the arrangement density of the photoelectric conversion elements faithfully to the read image while suppressing the increase in cost by keeping the arrangement density of the photoelectric conversion elements low. It becomes an image reading device.

[Brief description of drawings]

FIG. 1 is a functional block diagram showing a configuration of a facsimile apparatus configured by applying an image reading apparatus of the present invention.

FIG. 2 is a side view showing a specific configuration of a photoelectric conversion section 1a.

FIG. 3 is a plan view showing a specific configuration of a CCD line sensor 23.

FIG. 4 is a diagram showing a specific configuration of a thermal print head 6a.

FIG. 5 is a diagram showing an arrangement state of a thermal print head 6a.

6 is a positional relationship between the CCD line sensor 23 and a black image to be read, "standard mode" and "fine mode" of the output signal of the ACC processing unit 1c and the generated image signal.
FIG. 6 is a diagram showing a correspondence relationship in “high definition mode”.

FIG. 7 is a diagram showing a positional relationship between the CCD line sensor 23 and a black image to be read, and a correspondence relationship between an output signal of the ACC processing unit 1c and a generated image signal in an “ultra high definition mode”.

FIG. 8 is a diagram schematically showing a pattern for determining pixels in an 8-16 conversion unit.

FIG. 9 is a diagram schematically showing an arrangement relating to determination of recording density in the pseudo high-definition recording processing unit 6b.

FIG. 10 is a diagram showing a current distribution in the heating resistor 40b when the heating resistor 40b is energized.

FIG. 11 is a diagram showing an effective heat generation area corresponding to a level 1 pixel.

FIG. 12 is a diagram showing an effective heat generation area corresponding to a level 2 pixel.

FIG. 13 is a diagram showing a formation state of an effective heat generation area of a heat generation resistor 40b when an original image in which dots are discretely positioned is recorded.

FIG. 14 is a diagram showing a formation state of an effective heat generation area of a heat generation resistor 40b when recording an original image in which dots are positioned in a solid state.

FIG. 15 is a diagram illustrating ACC processing.

FIG. 16 is a diagram illustrating ACC processing.

FIG. 17 is a diagram showing a relationship between a conversion pixel and a reference pixel in ACC processing.

FIG. 18 is a diagram illustrating ACC processing.

FIG. 19 is a diagram for explaining a defect when general ACC processing and pseudo high-definition reading are used together.

FIG. 20 is a diagram showing an example of an image corresponding to image data generated when ACC processing according to the first embodiment of the present invention and pseudo high-definition reading are used together.

FIG. 21 is a block diagram showing a specific configuration of an 8-16 conversion unit 1e.

FIG. 22 is a diagram showing a relationship between a conversion pixel and a reference pixel and a conversion pattern in a pseudo high definition signal generation process according to the second example of the present invention.

FIG. 23 is a diagram showing a relationship between a conversion pixel and a reference pixel and a conversion pattern in the pseudo high definition signal generation processing according to the third example of the present invention.

FIG. 24 is a diagram showing a relationship between a conversion pixel and a reference pixel and a conversion pattern in a pseudo high definition signal generation process according to the fourth example of the present invention.

FIG. 25 is a diagram showing a relationship between a conversion pixel and a reference pixel and a conversion pattern in the pseudo high definition signal generation processing according to the fifth embodiment of the present invention.

[Explanation of Codes] 1 ... reading unit 1a ... photoelectric conversion unit 1b ... quantization unit 1c ... edge enhancement processing unit (ACC processing unit) 1d ... binarization processing unit 1e ... 8-16 conversion unit 1f ... pseudo halftone processing Part 1g ... Noise removal part 1h ... Read control part 1i ... Read condition information storage part 23 ... CCD line sensor 23b ... Photoelectric conversion element 91, 92 ... Comparison circuit 93 ... Threshold value generation part 94 ... Pseudo high-definition signal generation circuit 95 ... Conversion Table 96 ... Reference pixel data memory

Claims (3)

[Claims] 1. An electric pixel signal of a level corresponding to the quantity of reflected light from a large number of read pixel positions two-dimensionally set in a read image by a large number of photoelectric conversion elements arranged at predetermined intervals. In the image reading device for reading the read image by converting each of the two, the boundary between white and black in the read image is identified based on each pixel signal, and the level of the pixel signal corresponding to the pixel located at the boundary is identified. Is performed to bring the edge closer to a predetermined white level or a predetermined black level, and the boundary is identified based on the state of only the pixels lined up in the predetermined first direction. For each pixel signal after the edge enhancement processing has been performed by the enhancement processing unit, it is determined to which of at least j stages of classification ranges the level belongs in advance. And a classification range determined by the determination means that the level of the corresponding pixel signal belongs to each of the read pixel positions, and the read pixel position is adjacent to a predetermined second direction different from the first direction. And a determination unit that determines the corresponding k pixels based on at least a part of the k pixels determined for the pixel signal corresponding to the read pixel position. apparatus. 2. An electric pixel signal having a level corresponding to the quantity of reflected light from a large number of read pixel positions two-dimensionally set in a read image by a large number of photoelectric conversion elements arranged at a predetermined interval. In the image reading device for reading the read image by converting the read image into each of the above, the level of each pixel signal corresponding to a large number of read pixel positions belongs to any of the j-level classification ranges in which the level is preset. A determination range for determining whether or not the corresponding pixel signal level belongs to each of the read pixel positions, and at least one read adjacent in the first direction. At least a part of the k pixels determined for the pixel signal corresponding to the pixel and at least one adjacent to a predetermined second direction different from the first direction. An image reading apparatus comprising: a determining unit that determines the corresponding k pixels based on at least a part of the k pixels determined for the pixel signal corresponding to the pixel position. . 3. The determining means uses pixels adjacent to the left side, the upper left side, the upper side, and the upper right side of the k pixels to be determined for determining the k number of pixels. The image reading apparatus according to claim 2.