JPH01261968A - Picture reader - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

[Detailed Description of the Invention] [Technical Field] The present invention relates to an image reading device that reads a document image by image;
In particular, the present invention relates to an image reading device having a light source that illuminates a document.

[Prior art]

In facsimile machines, digital copying machines, etc., the document is exposed to light from a light source such as a fluorescent lamp, and the strength of the reflected light from the document is detected by a CCD image sensor, etc.
2. Description of the Related Art An image reading device that photoelectrically reads a document image and obtains an electrical image signal is known.

In such an apparatus, the amount of light emitted from the original exposure light source affects the image signal, so it is necessary to adjust the light so that the light amount is optimal.

To control the light of this type of image reading device, a standard white plate is placed outside the document reading area as a reference for the white level, and the light source is energized so that the data read from this white plate becomes the standard white signal. It is proposed to control the amount.

In addition, due to uneven illuminance distribution of the light source, transmission characteristics of the lens, and variations in the sensitivity of the C'CD image sensor,
The output of an image of uniform density read by a qCD image sensor may not be uniform.

This state is called shading, and in order to correct this, the reading signal of the original image is electrically corrected based on the output of reading a standard white plate of uniform density using a CCD image sensor.

In devices having such a dimming function and a shading correction function, a common standard white plate is often used for both functions.

However, there is a problem in that the white level of this standard white plate, especially the white level with a reflection density of 0.1 or less, is difficult to control in manufacturing, resulting in high costs.

Furthermore, even if you manage the white level of this standard white board and assemble the device, the white level may deteriorate due to discoloration or darkening caused by temperature, humidity, impurities contained in the device cooling air, infrared rays from the lighting system, ultraviolet rays, etc. over time. This causes a problem in that the white level of the output image cannot be guaranteed, and optimal light control and shading correction cannot be performed.

〔the purpose〕

The present invention has been made in view of the above points, and an object of the present invention is to provide an image reading device that can satisfactorily perform dimming or shading correction for image reading.

Another object of the present invention is to provide an image reading device that can sufficiently cope with changes in the density of a reference member that serves as a reference for dimming or shading correction, and can perform satisfactory dimming or shading correction.

Another object of the present invention is to provide an image reading device that can always output uniform image signals.

The second and other objects and effects of the present invention will become clear from the following description.

〔Example〕

The present invention will be explained below using preferred embodiments.

[First Embodiment] FIG. 1 is a block diagram of a document reading device to which the present invention is applied. It is pressed by the document cover 100 and the document platen glass 1
A CC equipped with a plurality of light-receiving elements arranged in the main scanning direction to read the image information of the original 102 placed on the original 01.
An image sensor 103 such as D is used. Illumination light from illumination 104 is reflected on the surface of original 102 and is reflected by mirror 105 .
106. An image is formed on the image sensor 103 by a lens 108 via a lens 107 . An optical unit 1.13 consisting of an illumination 104 and a mirror 105 and an optical unit 114 consisting of mirrors 106 and 107 move at a relative speed of 2.1.

This optical unit is operated by a DC servo motor 109 □
The robot moves from left to right at a constant speed while applying PLL control. This movement speed is 22.5mm on the outward journey depending on the magnification.
It is variable from /sec to 360mm/see, and is always 800mm/sec on the return trip.

While reading the main scanning direction (hereinafter referred to as the Y direction) perpendicular to the sub-scanning direction (hereinafter referred to as the X direction) in which this optical unit moves with a resolution of 400 dots/inch using the image sensor 103, After moving from the home position to a predetermined position to the right, the robot moves back to the home position again to complete one scan.

As a result, the entire area of the document is read and scanned line by line.

Reference numeral 111 denotes a light-shielding plate, and when this light-shielding plate 111 crosses a home position sensor 110 consisting of a photo ink converter, it is detected that the optical unit is located at the home position. Reference numeral 112 denotes a standard density plate. The standard density plate 112 is used for shading correction and light intensity control of the illumination 104, and the detection position of the home position sensor 110 is the reading position of the standard density plate 112.

FIG. 2 shows a configuration example of an image processing block in the document reading device shown in FIG.

In this embodiment, the lighting 104 uses a fluorescent lamp, and is dimmed by a light source light amount control unit 201 using a high frequency lighting time control method using Cyrisk or the like.

The reading surface 202 is a surface that reflects light from the illumination 104 and has original density information read by the CCD line sensor 103. By moving the optical unit, the standard density plate 112 and the original 102 in FIG. 1 become the reading surface. .

The line-by-line analog image signal read by the CCD 103 is amplified by an amplifier 203, and then converted into a multivalued (8 bits in this embodiment) digital image signal 216 by an A/D converter 204.

In this embodiment, the black level of the document is read as 'o''-, and the white level is read as '255''.The black reference level and white reference level of the A/D converter 204 are fixed values. is given, the lighting 104 is completely turned off,
The offset of the amplifier 203 is adjusted so that the A/D conversion output in a state where no light is incident on the CCD 103 is 0 level, which is completely black.

A CCD drive signal generation circuit 205 generates CDD drive signals 206 such as a reset signal, a clock signal, and a horizontal synchronization signal necessary for driving the CCD 103, and also generates an A/
Generation of clock signal 207 to D converter 204, CCD
C which is an address signal for identifying each bit of 103
A CD address 208 is generated.

In this embodiment, since the CCD 103 is a 5000 pixel line sensor, the CCD address counts up from 0 to 4999 in response to the 5000 pixels read out in synchronization with the horizontal synchronization signal.

A shading data RAM 209 stores digital image signals for two lines of ccD from the A/D converter 2o4 for all pixels, and one of the two lines of data is selected by a bank switching signal 226 from the CPU 212.

Address selector 210 is shading data RAM
It is a selector for the address signal 211 given to the address signal 209,
An address switching signal 213 from the CPU circuit section 212 switches between the CCD address 208 and the CPU address from the address bus 214 of the CPU circuit section.

The shading correction table RAM 215 includes the illumination 10
This is a table for correcting non-uniformity of the read image signal 216 caused by uneven light distribution included in the 4 and lens 108, uneven sensitivity of each pixel of the CCD 103, amplification degree setting error of the amplifier 203, etc. Prior to the original reading operation, the CP
The correction table data is written in the shading correction table RAM 215 by U212, and when reading the original,
By addressing the correction table RAM 215 using the shading data 217 from the shading data RAM 209 and the read image signal 216, an image signal 218 with the non-uniformity of the read image signal 216 corrected is obtained.

The address selector 219 is a selector for an address 220 given to the shading correction table RAM 215.
The address switching signal 221 from the PU212 causes the CP
Either the U address 214 or the address containing the read image signal 216 and shading data 217 is selected.

The motor driver 223 controls the forward movement, backward movement, and speed of the optical system drive motor 109, and controls the speed of the optical system drive motor 109.
The motor 109 is driven by an encoder pulse signal 224 synchronized with the rotation of the DC servo motor 109 using PLL control. This encoder pulse signal 224 is also given to the CPU 212, and the CPU 2
12 detects the scanning position of the optical system by counting this encoder pulse signal 224.

The home position sensor 110 is a sensor made of a photo ink converter, and is fixed to the reading area of the standard density plate 112, and the light shielding plate 11 fixed to the illumination system 104
1 is detected to be at the location of the home position sensor 110, the reading surface 104 by the CCD 103 becomes a standard density plate.

The CPU 212 is a ROM that stores a control procedure program.
.

It consists of a microcomputer with built-in working RAM, calculation section, etc., and performs sequence control in this embodiment.

Performs image data correction processing, illumination system light amount control, and operation unit control.

The detailed configuration of the operation section 225 is shown in FIG. This operation section performs key input and display operations under the control of the CPU 212. 301 is a key for starting the dimming value measurement mode, and 302 is an LED indicating that the dimming value measurement mode is being executed.

Reference numeral 303 is a key for starting a reading operation mode for the original 102 placed on the original platen 101, and 304 is an LED that indicates that the original reading mode is being executed.

Reference numeral 305 is a key for setting the measured dimming value even brighter, and by pressing this key, the display 307
The dimming value adjustment display increases. Reference numeral 306 is a key for setting the measured dimming value to a darker value, and by pressing this key, the dimming value adjustment display on the display 307 decreases.

The display unit of the dimming value adjustment display on the display 307 is the ILSB of the A/D converter 204, and the value displayed here is saved in a battery backup memory (not shown) included in the CPU 212 even when the power of the device is turned off. will be displayed again when the power is turned on again.

With the above configuration, this embodiment has two operation modes. One is to adjust the illumination 104 so that the read image signal 216 of a document whose reflection density is controlled becomes a constant value, and measure the read image signal value 216 from the standard density plate 112 at that light intensity. This is a light value measurement mode. The other method is to dim the illumination 104 so that the read image signal value from the standard density plate 112 becomes the value measured in the dimming value measurement mode, and then read and scan the original 102 on the original platen 101 to perform shading. This is a document reading mode in which a corrected image signal 218 is output.

These two operating modes will be explained below.

(Dimmer value measurement mode) As a preliminary preparation for the light control value measurement mode, as shown in Figure 4,
A document 401 whose reflection density has been controlled is placed on the document table crow 101.
Place it on top. In this embodiment, as the original document 401 whose reflection density is controlled, a standard white paper of A3 size immediately after opening is used, for example, a stack of 10 sheets of A3 size white paper whose reflection density is controlled to be 0.07. Therefore, the reflection density from this original 401 is approximately 0.07, which can be considered as standard white.

After this setup, when the operator inputs the light control value measurement 5W301 shown in FIG. 3, the CPU 212 turns on the light control value measurement in progress display 302 and executes the light control value measurement mode shown in FIG.

In step 501, the optical unit 113 is advanced 10 cm toward the original from the position of the home position sensor 110 so that the original reading surface 202 is the original 401 placed on the original platen glass 101. At this time, the CPU 212 outputs 0, 05 per pulse from the optical system drive motor 109.
The stopping position of the optical system is determined by counting 2000 mm encoder pulses.

Next, in step 502, the illumination 104 is turned on and dimmed so that the A/D converter output 216 becomes [240j (reference dimming). Not setting the standard white to the A/D converter's full range output 255 is 0.0
This is for decomposing originals with a reflection density of 7 or less,
Furthermore, by providing 16 levels of dimming amplitude on the bright side during dimming, it is possible to avoid using the output saturation state of the A/D converter and perform accurate reference dimming. Further, in this embodiment, since the A/D conversion output is 8 bits, the dimming target value is set to r240j, but this target value is set as appropriate depending on the resolution of the A/D conversion.

In order for the CPU 212 to read this A/D converter output 216, the shading data RAM 209 is used.

First, the A input of address selector 210, that is, CC
Select the D address 208 and write the shading data R.
Let it be the address of AM209. Then, R
Write to AM209.

Next, the B input of address selector 210, that is, CP
The CPU 212 measures the light amount data by using the address bus 214 of U 21 '2 as the address of the RAM 209 and reading the data written in the RAM 209 via the data bus 222.

Here, the CPU 212 selects 8 pixels in the center, that is, 25 pixels from the 2497th pixel among the data of the 5000-pixel CCD line sensor 103 as pixels for capturing light amount data.
Eight consecutive pixels up to 04 pixels are used.

Then, the data of these 8 pixels is averaged to obtain light amount data. Then, data writing to and data reading from the RAM 209 are repeated while increasing/decreasing the illumination light emission amount data for the light amount control unit 201 so that this light amount data becomes 240J.

As a result, when the amount of illumination light emission that makes the light amount data r240J is determined, the optical unit 104 is moved to the home position sensor position in step 503 while maintaining the same illumination emission state, and the standard density plate 112 is moved to the home position sensor position in step 504. The light amount data from the RAM 209 is read.

The series of operations from step 501 to step 504 described above is repeated six times in step 505, and the light amount data from the standard density plate 112 for six times is measured.

Next, in step 506, the dimming target for the standard density plate 112 is determined by calculating the average value of four pieces of light quantity data by removing the maximum and minimum values of the six pieces of light quantity data.

When actually reading a document, by setting the reflected light amount data from the standard density plate 112 to this dimming target value, a portion of the document with a reflection density of 0.07 will always have a value of "240" in the A/D conversion output 216. It will be read in

This dimming target is stored in a predetermined address of the battery-backed nonvolatile memory in step 507, and a code indicating that "storage has been completed" is written. Then, in step 508, the dimming value measurement in progress display 302 is turned off to end the dimming value measurement mode.

Note that this dimming target value is based on the standard density plate 112 and the reference white 40.
1, it is necessary to execute this dimming value measurement mode when the standard density plate is replaced or when the density of the standard density plate changes due to changes over time.

(Image reading mode) FIG. 6 shows the operation flow in the image reading mode.

The CPU 212 detects that there is an input from the image reading 5W 303 of the operation unit 216, and the CPU 212 performs the image reading process.
lights up.

Next, in step 601, it is checked whether or not the dimming target value in the dimming value measurement mode has been set, based on the code indicating that "memorization has been completed" written in step 507 in the dimming value measurement mode. .

As a result, if the dimming target value has not been set, the image reading mode 304 is blinked in step 602, and a fixed value programmed in advance is used as the dimming target value. This fixed value is a value corresponding to the most frequently occurring concentration during manufacture of the standard concentration plate.

In this way, after adopting the dimming target value or fixed value in the dimming value measurement mode as the dimming target value, step 6
03, the optical unit 113 is moved to the position of the home position sensor 110 in order to read the reflected light from the standard density plate 112.

Next, in steps 604 and 605, shading data is captured. The shading correction used in this embodiment is performed by assuming that the shading data from the standard density plate 112 is equivalent to the shading data from the reference white original 401. For this purpose, standard concentration plate 11
The maximum value of the read data 216 from 2 is the reference white original 40
The lighting 104 is dimmed so as to match the standardized reading value "240" of 1.

To this end, first, in step 604, the light amount data read from the standard density plate 112 irradiated with appropriate illumination light amount data is written into the shading data RAM2Q9 using the COD address 208.

Next, the CPU 212 uses the B input of the address selector 210 to transfer the address bus 214 to the shading data RA.
Give it to M209 and take in the data from RAM209,
Data at the address corresponding to all bits of COD is A/D
Check whether the saturation level r255j of converter 204 has been reached.

If there is a bit that has reached J255J,
By subtracting the illumination light emission amount data given to the light amount control unit 201,
The light amount data read from the standard density plate 112 is written into the shading data RAM 209 again, and it is checked whether the saturation level has been reached. After confirming that all the bits have not reached the saturation level in this way, the bit (light control point) showing the maximum level is searched for.

The above operation is called a dimming point search.

Next, by similar data sampling operations and illumination light emission amount data operations, light adjustment is performed so that the average value of eight consecutive pixels including the light adjustment point approximately in the center becomes the normalized light amount data value r240J of the standard white original.

Next, in step 605, the standard density plate 112 in the dimming state is read by the CCD 103, and the read light amount data from ccDlo3 is stored in the lower bank of the shading data RAM 209 (signal 226 is at Low level) as shading data from the standard white original. Write.

Next, in step 606, the bank switching signal 226 is set to High level and the shading data RAM is
The upper bank No. 209 is used to perform dimming so that the light amount data from the standard density plate becomes the dimming target value determined in steps 601 and 602. That is, dimming is performed so that the average value of eight consecutive pixels including the dimming point used in step 604 becomes the dimming target value.

Next, in step 607, the shading data R
The bank to be used in the AM 209 is returned to the lower bank, the shading correction for the original read image 216 is performed in the shading correction table RAM 215 using the shading data written in step 605, and the optical unit 113 is sub-scanned in the X direction at a predetermined speed. The original is read and scanned.

FIG. 7 shows an outline of shading correction in this embodiment. The horizontal axis corresponds to each pixel of the CCD 103, and the vertical axis represents the A/D converter output value corresponding to each pixel. Characteristic a
is a shading characteristic corresponding to the standard white document sampled in step 605, and the document image signal read with this shading characteristic is output as an image signal C corrected by shading correction. Here, since the output of CCD 103 is proportional to the amount of light,
It is corrected as

Correction data based on this correction formula is written into the shading correction table RAM 215 when the power of the apparatus is turned on. That is, as shown in FIG. 8, shading data RA'M 20
The shading characteristic a from 9 is input in synchronization with each bit of COD, and the lower address is the shading characteristic a from reading the original.
The shading correction table RAM 215 is configured such that the /D conversion output is input and a correction output C is output by a combination of the a input and the b input.

As explained above, in this embodiment, uniform document reading without shading is performed with density control from the reference point to the normalized white level. 306. Step 60 using display 307
By adjusting the dimming target value used in step 6, it is possible to adjust the level of the white side of the read image. In other words, by increasing the dimming level when scanning an original than the value measured in dimming value measurement mode, the standard original white level increases isometrically, so the original is read brighter, and the value is lower than the measured value. This makes the manuscript read dark.

The key human power to make it bigger (brighter) is 305,
The key power to make it smaller (darker) is 306. C
The PU212 displays the dimming value 30 using this key manually.
Display 7. The display unit is the minimum decomposition level of the A/D converter 204, and when it increases, it is displayed as a plus, and when it decreases, it is displayed as a minus.

Note that in step 602 in this embodiment, if the dimming target value has not been set, the reading operation is not executed;
It is also possible to alert the operator by flashing the display.

Furthermore, it is not limited to reading images using black and white signals, but can also be applied to reading using color signals if the standard density plate is made to reflect white light as white light with a certain density.
The material may have high reflection characteristics for light of a specific wavelength.

In addition, if the dimming value variable operation is performed based on the result of the background density detection means of the document by pre-scanning or the like, the effect of automatic background skipping can also be obtained. In this case, pre-scan with a predetermined light control value, create a histogram for each light amount data near white, obtain light amount data from the background, and then calculate the light amount data W.
Automatic background skipping is also possible by changing the dimming target value A to A' shown by the following formula so that the value becomes level 240, and then reading and scanning the document.

W As explained above, by using a white original with a relatively controlled density such as copy paper that is available around you, there is no need to strictly control the density of the standard density plate used for samples of dimming and shading data. (It has the effect of becoming.

Furthermore, by intentionally changing the dimming value, the density of the output signal can be adjusted without using an additional circuit.

Furthermore, since uniform density shading characteristics can be obtained using an unspecified density plate, there is an advantage that a shading correction circuit for each density and a correction calculation for each density are not required.

In the embodiment described above, in order to make the shading data from the standard density plate 112 match the shading data from the reference white original 401 when importing shading data, the maximum value of the read data on the standard density plate 112 is set to the reference white original. 401 normalized reading r
The lighting was to be dimmed to match 240 J. Therefore, at the home position, a light control operation for capturing shading data and a light control operation for reading an original image are required.

Next, a second embodiment configuration will be described which makes it possible to perform accurate shading correction without performing separate light control operations for capturing shading data and for reading an original image.

It should be noted that the configuration of the first embodiment described above and the configuration of the second embodiment described below differ only in the software in the image reading mode of the CPU 212, and the image reading device shown in the first diagram of the first embodiment differs. The configuration of the image signal processing block shown in the second figure, and the configuration of the operation section shown in the third figure are also common to the second embodiment. Therefore, the image reading mode in the second embodiment will be explained below without repeating those explanations.

[Second example] (Image reading mode) FIG. 9 shows the operation flow in the image reading mode.

The CPU 212 detects that there is an input from the image reading 5W 303 of the operation unit 216, and displays the image reading in progress display 3.
04 is lit.

Next, in step 701, it is checked whether or not the dimming target value in the dimming value measurement mode has been set, based on the code indicating that "memorization has been completed" written in step 507 in the dimming value measurement mode. .

As a result, if the dimming target value has not been set, the image reading mode 304 is blinked in step 702, and a fixed value programmed in advance is used as the dimming target value. This fixed value is set to a value corresponding to the frequently occurring concentration during manufacture of the standard concentration plate.

After adopting the dimming target value or fixed value in the dimming value measurement mode as the dimming target in this way, step 70
3, the shading correction table RAM 215 is written.

In this embodiment, the shading correction is performed by reading the standard density plate 112 of each pixel of the CCD line sensor based on the idea that the reference white of the original is guaranteed by using the reflected light amount data from the standard density plate 112 as the dimming target value. Data is corrected so that the output matches the dimming target value.

FIG. 10 shows a shading correction method.

In this figure, the reading output of the standard density plate 112 is S.
Shading correction is performed on a certain pixel 103 to obtain a dimming target value N as a correction output.

The horizontal axis in FIG. 10 is the A/D conversion output before correction, and the vertical axis is the value after shading correction.

When shading correction is not performed, a straight line with an inclination of 45° makes the input and output values equal.

That is, (N-3) correction amounts are added to the pixels of the read output S of the standard density plate 112.

Generally, the CCD line sensor 103 is a device that generates an output linear with respect to the amount of light, so the correction amount for the original image read with the value X in FIG. 10 is -(N-3).

S Therefore, the corrected output y is given by the following equation.

y = x + -(N-8) The shading correction table RAM 215 that outputs the result of this correction calculation is configured as shown in FIG. A document reading signal read in the shading state is input.

Therefore, the CPU 212 selects the S input of the address selector 219 in step 703. Then, using the dimming target value N determined in step 701 or 702, all the shading data S and the original reading signal
The correction output y for the combination is calculated, and the calculation result is written into the shading correction table RAM 215. Then, the A input of the address selector 219 is selected.

Next, in step 7o4, the CPU 212
At the same time, the light intensity of the CCD 103 is adjusted to the dimming target value.
The shading data S from the standard density plate 112 is
The optical unit 113 is moved to the position of the home position sensor 110 in order to read the .

Next, in step 705, the illumination 104 is turned on and dimming is performed so that the maximum value of the read signal of one line of the standard density plate becomes the dimming target value N.

To do this, first, the light amount data read from the standard density plate 112 irradiated with appropriate illumination light amount data is stored in the shading data RAM 20 using the CCD address 208.
Write in 9.

Next, the CPU 212 uses the S input of the address selector 2]0 to send the address bus 214 to the shading data RA.
M209 to fetch data from RAM209, C
It is checked whether the data at the address corresponding to all bits of the CD has reached the saturation level r255J of the A/D converter 204.

If there is a bit that has reached r255J,
By subtracting the illumination light emission amount data given to the light amount control unit 201,
The light amount data read from the standard density plate 112 is written into the shading data RAM 209 again, and it is checked whether the saturation level has been reached.

After confirming that all the bits have not reached the saturation level in this way, the bit (light control point) showing the maximum level is searched for.

The above operation is called a dimming point search.

Next, by similar data sampling operations and illumination light emission amount data operations, dimming is performed so that the average value of eight consecutive pixels including the dimming point approximately in the center becomes the dimming target value N.

Next, in step 706, the standard density plate 112 in the dimming state is read by the CCD 103, and the output shading data S is stored in the shading data RAM 209.
write to.

Next, in step 707, the shading data RAM2
The shading data 217 read from 09 in synchronization with each bit of the CCD and the original read image 216 are used as the S input and 113 is sub-scanned in the X direction at a predetermined speed to perform document reading scanning.

FIG. 11 shows the state of shading correction at this time.

That is, since the light is adjusted so that the maximum value of the shading data S from the standard density plate 112 becomes the dimming target value N as shown in FIG. The output is shown in Figure 11 (b
), the original reading signal X becomes a y signal that does not include shading.

When writing the shading data S to the shading data RAM 209, even if the dimming of the illumination system is insufficient as shown in (a) and (b) of FIG. Since this information is used as data and the document is read in the same dimming state, information of the same density as the standard density plate 112 on the document always becomes the dimming target value N.

This is because, as is clear from Figure 10, shading correction works to ensure the degree of density of the standard density plate for the standard white document measured in the light control value measurement mode using the digital signal value. By accurately performing light adjustment in the value measurement mode and accurately measuring the relationship between the density of the reference white original and the standard density plate as the light adjustment target value N, the light adjustment during document reading may be a little rough.

As explained above, in this embodiment, uniform document reading without shading is performed with density control from the reference point to the normalized white level. 306. Step 706 using display 307
By adjusting the dimming target value used in , it is possible to adjust the level of the white side of the read image. In other words, by increasing the dimming level when reading an original than the value measured in the dimming value measurement mode, the standard original white level will increase isometrically, so the original will be read brighter, and it will be lower than the measured value. As a result, the manuscript reads darkly.

The key human power to make it bigger (brighter) is 305,
The key power to make it smaller (darker) is 306. C
The PU212 displays the dimming value adjustment 3 using this key manually.
07 is displayed. The display unit is the minimum resolution level of the A/D converter 204, and when it increases, it is displayed as a plus.
When the value decreases, a negative value is displayed.

Note that in step 702 in this embodiment, if the dimming target value is not set, it is also possible to warn the operator by blinking the display without executing the reading operation.

Furthermore, it is not limited to reading images using black and white signals, but can also be applied to reading images using color signals if the standard density plate is made to reflect white light as white light with a certain density.
The material may have high reflection characteristics for light of a specific wavelength.

In addition, if the dimming value variable operation is performed based on the result of the background density detection means of the document by pre-scanning or the like, the effect of automatic background skipping can also be obtained. In this case, pre-scan with a predetermined light control value, create a histogram for each light amount data near white, obtain light amount data from the background, and then calculate the light amount data W.
Automatic background skipping is also possible by changing only the light control value using the light control target value N' expressed by the following formula so that the image becomes the reference white level 240, and scanning the document.

N' 24O W Furthermore, it is not necessary to write the correction data to the shading correction table RAM 215 each time the image reading 5W 303 is performed manually, and it is only necessary to write the correction data once when the power of the apparatus is turned on.

As explained above, the density ratio of the standard density plate to the reference white original is measured using a standard density plate that does not have strict density control, and shading correction is performed to ensure the density ratio when reading the original. In order to achieve this, it is only necessary to roughly adjust the brightness for each document reading in a relatively short period of one hour, which shortens the document reading time and ensures that the white density is always maintained even if the illumination state and the density of the standard density plate vary. It is possible to perform guaranteed uniform document reading, and there is no need for a light source light amount control unit, a standard density plate, etc. to have high precision, and the cost of the apparatus is reduced.

[Third Embodiment] When an illumination lamp such as a halogen lamp with a stable light amount is used as the illumination 104 in FIG. do not need. This is because, in the case of a halogen lamp, there is a very high correlation between the lighting voltage and the amount of light, so if the lighting voltage of the halogen lamp is determined, a stable amount of light can be obtained.

In order to correct changes in the density of the standard density plate 112 and variations in density from device to device when such illumination is used as the reading device shown in FIG. 1, the following steps are taken in this embodiment. First, the relative density ratio of the standard density plate 112 to the standard white placed on the document table is determined. Second, a value F is obtained by multiplying the reference white level E by this relative density ratio. Thirdly, when reading the manuscript,
Shading correction is performed so that the original reading value X becomes y at a rate necessary to correct the reading value H from the standard density plate 112 to F. For this reason, this embodiment has two operation modes. One mode is the aforementioned standard density plate 112.
This is a density ratio measurement mode that measures the density ratio relative to the reference white. The other is a document reading mode in which the document 102 is read and scanned using this density ratio, and an image signal 218 that has undergone shading correction is output.

These two operating modes will be explained below.

(Density Ratio Measurement Mode) As a preliminary preparation for the density ratio measurement mode, as shown in FIG. As the original document 401 whose reflection density is controlled, a standard white paper of A3 size immediately after opening is used, for example, a stack of 10 sheets of A3 size white paper whose reflection density is controlled to be 0.07. The reflection density from 40 is approximately 007 and is considered standard white.

After this setup, the operator inputs the concentration ratio measurement 5W1401 in FIG.
Step 2 turns on the concentration ratio measurement in progress display 1402 and executes the concentration ratio measurement mode shown in FIG.

FIG. 14 shows a detailed configuration of the operating section 225 in this embodiment. This operation section performs key input and display operations under the control of the CPU 212. 1401 is a key for starting the concentration ratio measurement mode, and 1402 is an LED that indicates that the concentration ratio measurement mode is being executed.

Reference numeral 1403 is a key for starting a reading operation mode for the original 102 placed on the original table 10142, and 1404 is an LED that indicates that the original reading mode is being executed.

FIG. 15 shows a control flowchart of the CPU 212 in the concentration ratio measurement mode.

In step 1501, the original reading surface 202 is slid onto the original 401 placed on the original platen glass 101, and the optical unit 113 is advanced 10 cm from the position of the home position sensor 110 toward the original. At this time, the CPU 212 outputs 0, 05 per pulse from the optical system drive motor 109.
The stopping position of the optical system is determined by counting 2000 mm encoder pulses.

Next, in step 1502, the light source light amount control unit 201
The lighting data is output to the lighting 10 at a predetermined lighting voltage.
Turn on 4. In this case, the lighting voltage may be set relatively roughly as long as the maximum value of the A/D converter output 216 when reading the standard blank paper is less than the full range output 255. Therefore, the light source light amount control unit 201 of this embodiment is controlled by the CPU 2.
The lighting device 104 is configured to constantly supply a constant lighting voltage to the lighting device 104 in accordance with the lighting data from the lighting device 12 .

In this lighting state, the CPU 212 reads light amount data from a standard white document on the document table as an A/D converter output (standard white data measurement).

In order to read this A/D converter output 216, a shading data RAM 209 is used.

First, the eight power of address selector 210, that is, COD
Select address 208 and write the shading data RA
This is the address of M209. Then, the light intensity data for the CCDI line from the reference white original at a certain light intensity is RA
Write to M209.

Next, the B input of the address selector 210, that is, the CPU
The address bus 214 of 212 is used as the address of RAM 209, and the data written in RAM 209 is transferred to data bus 2.
22, the CPU 212 measures the light amount data.

Here, the CPU 212 selects 8 pixels in the center, that is, 25 pixels from the 2497th pixel among the data of the 5000-pixel CCD line sensor 103 as pixels for capturing light amount data.
Standard white data P is obtained by averaging data of eight consecutive pixels up to pixel 04.

As a result, when the standard white data P is determined, the optical unit 104 is moved to the home position sensor position in step 1503 while the illumination remains on, and the light amount data from the standard density plate 112 is read in step 1504. It is read from the RAM 209 as standard concentration data Q.

Next, in step 1506, the density ratio is determined using the standard white data P and the standard density board data Q using the following equation.

D-Double P This concentration ratio is stored in a predetermined address of the battery-backed nonvolatile memory in step 1507, and a code indicating that storage is completed is written. Then, in step 1508, the concentration ratio measurement display 1
402 is turned off to end the concentration ratio measurement mode.

Note that this density ratio is based on the standard density plate 112 and the reference white 40.
Since it means the ratio of the amount of reflected light to 1, it is necessary to execute this density ratio measurement mode when the standard density plate is replaced or when the density of the standard density plate changes due to changes over time.

(Image reading mode) FIG. 16 shows the operation flow in the image reading mode.

The CPU 212 reads the image of the operation unit 216 5W1403
The image reading in progress display 1404 is turned on upon detecting that there is an input.

Next, in step 1601, it is determined in step 150 whether or not the density ratio is set according to the density ratio measurement mode.
Check based on the code written in step 7 indicating that storage has been completed. As a result, if the density ratio has not been set, the image reading nail width display 1404 is blinked in step 1602, and a fixed value programmed in advance is used as the density ratio. This fixed value is set to a value corresponding to the frequently occurring concentration during manufacture of the standard concentration plate.

In this way, after adopting the density ratio measurement value or fixed value in the density ratio measurement mode as the density ratio, in step 1603, the shading correction table RA
Write M215.

In this embodiment, the reference white level of the original is guaranteed by setting the reflected light amount data from the standard density plate 112 to a value F (normalized standard density plate level) obtained by multiplying the reference white level E by the density ratio. Based on this idea, in the shading correction, light amount data is corrected so that the standard density plate output of each pixel of the CCD line sensor becomes the normalized standard density plate level F.

In this embodiment, 240 is used as the reference white level E.

FIG. 17 shows a shading correction method.

In this figure, shading correction is performed on a pixel with a COD whose standard density plate output is S, and a normalization level F is obtained as a correction output. The horizontal axis represents the A/D conversion output before correction, and the vertical axis represents the value after shading correction.

If shading correction is not performed, a correction amount of (1-3) is added to the standard density plate output S because the input and output values are equal to each other and the straight line has an inclination of 45°.

Generally, the COD line sensor 103 is a device that produces an output linear with respect to the amount of light, so the correction amount for the original image read with the value X in FIG. 17 is -(F-3).

Therefore, the corrected output y is given by the following equation.

Therefore, it is corrected to the reference white level E as shown in the following equation.

The shading correction table RAM 215 that outputs the results of this correction calculation is configured as shown in FIG. is input.

Therefore, the CPU 212 selects the B input of the address selector 219 in step 1603, and uses the normalized standard density plate level F to perform shading correction on the correction output y for all combinations of the shading data S and the document reading signal X. Write to table RAM 215. Then, the A input of the address selector 219 is selected.

Next, the CPU 212 moves the optical unit 113 to the position of the home position sensor 110 in order to read the shading data S from the standard density plate 112.

Next, in step 1605, the illumination 104 is turned on.

Next, in step 1606, the shading data S of the standard density plate 112 in that illumination state is written into the shading data RAM 209.

Next, in step 1607, the shading data RAM
The shading data 217 and the original read image 216 read from the shading data 209 in synchronization with each CCD bit are used as the S and X inputs of the shading correction table RAM 215, and the optical unit 113 is moved in the A document reading operation is performed by performing sub-scanning at a predetermined speed.

The state of shading correction at this time is shown in FIG.

In other words, the shading data S from the standard density plate becomes the standard density plate data F which is uniformly normalized (uniformly normalized standard density plate data F) as shown in FIG. X becomes a y signal that does not include shading.

At this time, if a standard white document with a density of 0.07 is placed on the document table, the read data is corrected to the reference white level 240 by shading correction.

As explained above, in this embodiment, the density ratio of the standard density plate to the standard white is calculated, and the normalized standard density plate level F is obtained by multiplying that value by the reference white level E. . By performing shading correction using F obtained in this way, it is possible to uniformly read a document such that the reading value of a standard white document is output as the standard white level E even when using an illumination system that does not perform dimming. It becomes possible. Further, even if the density level of the standard density plate changes, uniform reading of the document becomes possible by measuring the density ratio again.

Although the present invention has been described above using some preferred embodiment configurations, the present invention is not limited to these embodiment configurations, and it is understood that various modifications and changes can be made within the scope of the claims. Needless to say.

[Brief explanation of the drawing]

FIG. 1 is a cross-sectional view of the image reading device, FIG. 2 is an electrical signal processing block diagram, FIG. 3 is a diagram showing the operating section controlled by the CPU 212, and FIG. 4 is a diagram showing the setup of reference dimming. FIG. 5 is a flowchart of the dimming value measurement mode, FIG. 6 is a flowchart of the document reading mode in the first embodiment, FIG. 7 is a schematic diagram of shading correction, and FIG. 8 is an input of the shading correction table RAM. It is an output diagram. FIG. 9 is a flowchart of the document reading mode in the second embodiment, FIG. 10 is a diagram showing the shading correction method, and FIG.
13 is a schematic diagram of shading correction, FIG. 12 is an input/output diagram of the shading correction table RAM, FIG. 14 is a diagram showing the operation section in the third embodiment, and FIG. 15 is a diagram of the density ratio measurement mode. Flowchart diagram; FIG. 16 is a flowchart diagram of the document reading mode in the third embodiment; FIG. 17 is a diagram showing the shading correction method;
The figure is a schematic diagram of shading correction, where 102 is an original, 103 is a CCD, 112 is a standard density plate, and 204 is an A/
D converter, 209 shading data RAM; 212 is a CPU.

Claims (1)

[Scope of Claims] A reading means for photoelectrically reading a document image and outputting an image signal, a reference member used for measuring variations in the output of the reading means, a reading output of the document image of a standard density by the reading means, and Setting means for setting a rate of read output of the reference member by the reading means; variation in image signals obtained by reading the document image by the reading means using output signals obtained by reading the reference member by the reading means; An image reading apparatus, characterized in that the correction means corrects variations in image signals in consideration of the ratio set by the setting means.