JPS63211866A - Image reader - Google Patents

Abstract

PURPOSE:To constitute an image reader of an excellent quality of synthesis by controlling the outputs of respective photoelectric conversion elements at the time of synthesing so that the outputs comes to be the data of the areas of an original-image projected in bilateral symmetry with the optical axis for the center. CONSTITUTION:The followings are provided: an illumination means 6 to illuminate the surface of an original 1, plural photoelectric conversion elements 10 such as charge coupled device that scan the original 1 whose read area is divided in plural sections line after line along the main scanning direction and convert optical signals from the original 1 to electrical signals, a memory that stores read data from the element 10 in prescribed addresses, a counter means to control the read start and the end address of the data in said memory, and a switching control means to synthesize data from the respective photoelectric conversion elements. At the time of said synthesizing, the outputs of the elements 10 are so controlled that they come to be the data of the areas of the original projected in bilateral symmetry with the optical axis for the center.

Description

DETAILED DESCRIPTION OF THE INVENTION (Technical Field) The present invention relates to an image reading device such as an image sensor, and in particular, the present invention relates to an image reading device such as an image sensor. Multiple photoelectric conversion elements are installed in parallel for data reading.

The present invention relates to an image reading device.

(Prior art) Conventional image reading devices have a low pixel density (about 8 pixels/■), and the document size to be read is, for example, A3 size (297W1).
Because of its width (width), it is abbreviated as a charge-coupled device (hereinafter referred to as CCI), which is used for reading documents.

), it was sufficient to use one with about 2600 pixels. Recently, there is also an image reading device that uses a 5,000 pixel CCD and has a pixel density of 16 lines/mm to read A3 size items, but this also uses one CCD.
D was able to read the manuscript.

However, if you try to read a larger A2 size or AO size document at a pixel density of 16 lines/mn, even if you use a 5000 pixel CCD, 2 to 3 CCDs will
You have to divide the scanning area with
In the GL reading device, when a document image is read by dividing the area between a plurality of CODs, a new problem arises at the image seam portion.

FIG. 11 is a schematic configuration diagram of an image reading device that reads a document using a plurality of CCDs. As shown in the figure, a plurality of optical lenses 52 and a plurality of CCDs 53 are arranged in pairs along the main scanning direction of the JM document 51. , the document surface is exposed in a slit shape by an illumination lamp (not shown), and the reflected light from the document surface is divided into nine optical lenses 52 and transmitted, and is received by the corresponding CCD 53, thereby forming a document image. Each line is optically read. The data read by the CCD 53 is stored in a memory of a control section (not shown).

Picture fg with this kind of composition! In the reading device,
For example, the image joint 54 may be displaced due to variations in the mounting positions of the optical lens 52 and the CCD 53, or due to thermal expansion or contraction of the CCD 53 itself or the optical unit containing it, which may prevent proper image reading. Normally, when the CCD 53 or the optical unit expands in a direction perpendicular to the optical axis, the joint 54 becomes double, and when it contracts, the joint 54 separates, resulting in data loss.

Furthermore, when creating one line of composite data based on the read data from each COD, if all the read data is taken in and combined, the quality of the data deteriorates, especially at the joints. This is thought to be because the M T F'' of the optical lens that projects an image onto the CCD decreases toward the periphery, and because the aberration of the lens increases, making magnification errors and focus errors more likely to occur.

(Objective) An object of the present invention is to solve the problems of the prior art and provide an image reading device with high composition quality.

(Structure) For this reason, the present invention includes an illumination means for illuminating the document surface, a reading area divided into a plurality of parts along the main scanning direction, scanning the document line by line, and converting optical signals from the document surface into electrical signals. a plurality of photoelectric conversion elements such as CCDs, a memory for storing read data from the photoelectric conversion elements at predetermined addresses, a counter means for controlling read start and end addresses of the memory data; and switching control means for synthesizing data from the photoelectric conversion elements.

At the time of data synthesis, the output of each photoelectric conversion element is controlled so as to become the data of the original image area projected symmetrically to the left about the optical axis.

(Example) Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 is a schematic configuration diagram of an image reading device according to an embodiment of the present invention, FIG. 2 is an explanatory diagram showing the arrangement of an optical lens and a CCD, and FIG. 3 is a front view showing a fixed state of the optical lens and CCD. , FIG. 4 is an overall control block diagram.

As shown in FIG. 1, when a document 1 to be read is inserted from the document table 2, the leading edge is detected by the document entrance sensor 3, and the IJKm conveyance motor (not shown) is driven by the detection signal, and the conveyance roller 4 .5 rotates and the original 1 is conveyed to the original reading section, and the illumination lamp 6 is turned on.

When the leading edge of the document is detected by the sensor 7 to the document lens 1, a shading start signal is sent from a control section (not shown), and the light reflected from the document pressure plate 8 at this time is converted to the A/D (
Analog/digital) converted signal is shading R
The signal is stored in the AM and used as a correction signal for unevenness in the amount of light in the main scanning direction, unevenness in sensitivity of each CCD, etc. when reading the JR document.

The document surface of the document 1 conveyed to the document reading section is exposed in a slit shape by the illumination lamp 6, and the optical information made up of the reflected light is reduced to about 1/9 by the optical lens 9 and is transmitted from, for example, a solid-state sensor. The image is read by the CCD 10. This CCD10 is one in which 5000 reading elements with a pitch of 7 μm are integrated (for example, TC106C manufactured by Toshiba Corporation).
The reading density of original 1 is 16 lines/m on the original surface.
The document 1, on which information has been optically read line by line in this manner, passes over the contact glass 11 and is discharged onto the tray 13 by the discharge rollers 12.

This image reading device has a CCD 1 so as to be able to read a relatively large original 1 such as AO size, for example.
A plurality of 0s (three in this example) are used. Then, as shown in Fig. 2, 918 mm in the main scanning direction is divided into three, and three CCDs 10a, 10b, 10c: H2, 5
The readings are divided into sections of 1 mm each, and several lines are written on one base member 14 (see FIG. 3) so that the overlap width between CCD I0 is 4 to 7 mII+.

FIG. 3 is a diagram showing how the optical lens 9 and CCD 10 are attached. A printed circuit board 14 on which a CCD, O, and other electronic components are mounted is fixed to a mounting plate 19 via two supports 16 and a bracket 17. On the other hand, the optical lens 9 is directly fixed to a mounting plate 19 by a lens holder 18 at a position facing the CCD 10. This mounting plate 19 is further attached to the base member 14 via a mounting bracket 20.

The magnification and focus are adjusted by adjusting the vertical movement of the optical lens 9 and the mounting plate 19 relative to the CCD10.

The base member 14, which ultimately fixes the CCD 10, is made of aluminum casting from the viewpoint of machining and weight. The coefficient of linear expansion of aluminum is 23X10-1'/℃, so when the ambient temperature changes from 5℃ to 35℃ (
The amount of deviation due to thermal expansion with a temperature difference of 30°C is CGD 10
When calculating the span of 305mm, from ΔQ=Q・α・Δt (Q: length, α: coefficient of linear expansion, Δt: temperature difference) ΔQ = 3
05X23X10-'/℃×30℃: 210 p r
It becomes n.

CC with 5000 readout elements integrated at 7μm pitch
When the projection is reduced to 1/9 using D10, if the above-mentioned shift amount of 210 μm is converted into pixels, it will be shifted by 3.4 pixels (pickles).

This is due to the expansion of only the base member 14, and considering the position of the CCD 10, the expansion of the mounting bracket 1-17, 20, the printed circuit board 15, and the CCD 10 itself are also included, so the actual measured value was 30°C. Approximately Q, 8mm due to temperature change, 13 pickles in pixel equivalent (P 1xel)
It's about to shift.

If there is a deviation of about 0.8 mm at the seam, the output image will be missing or, conversely, doubled, resulting in a low image quality.

In this embodiment, the ambient temperature of the CCD is detected by a temperature sensor such as a thermistor, the detection signal is amplified, and the A/D
The digital value is converted and stored as CCD data. The address point at which RAM readout starts is changed to compensate for deviations due to thermal effects when the ambient temperature changes.

FIG. 4 is an overall control block diagram. C0D1~3
is driven by a clock signal of 8M + - 1z, and the driving clock and 5YNC signal etc. are CLOCK & TIMIN.
Made by the G GENERATOR department. CCD 1~
3 scans each assigned area on the document at the same timing and outputs it as an image data string of 5000 P 1xel.

Since the output of the CCD is as small as approximately 0.1■ at the white level, direct A/D conversion would result in a large conversion error due to the effects of noise, etc., so it is amplified approximately 10 times using the VIDEO AMP. ing. Since the drive clock is superimposed as noise on the CCD, only the signal components are extracted by sampling. GA], N C0NTR0L is a variable amplifier to keep the white level of the signal input to the A/D converter constant. Prior to scanning the document, it converts the white color of the document pressure plate into a pixel signal, and the maximum level is set to the A/D converter. The degree of amplification is controlled so as to reach the maximum value of conversion (3FH for 6 bits).

The A/D converter converts an analog signal into a digital signal, and here, a 6-bit one is used to convert the signal level from white to black into data of 64 gradations.

RAM is for memory of shading correction data, and C is 1 to 5000 pixels when reading the original pressure plate.
The OD output is recorded as 6-bit 1-data, and is used as reference data for correcting a decrease in peripheral light intensity of the optical lens, uneven sensitivity of each CCD, uneven light intensity of the illumination lamp, etc. R.O.
Data is written in M so that it operates as a type of arithmetic unit, and the output of each CCD when an original is scanned is taken out as data after shading correction.

The 6-bit shading-corrected pixel data from the ROMI is input to a toggle buffer consisting of RAM4 and RAM5. 5ELL has RAM4 and RAM
5 to be selected alternately.

It operates with a switching signal from the TIMIGE GENERATOR.

The first scanned data is 1 to 5000 in RAM4.
Store as 6-bit data. The RAM address is ADDRESS C0UNTERI at 8 MHz.
It is controlled by 5EL2 with the clock signal of . At the same time, the output of the CCD 2 is stored in the RAM 6 as 6-bit data from the ROM 2, and the output of the CCD 3 is similarly stored in the RAM 8.

In the next scan, the output of CCD1 goes to RAM5.

The output of CCD2 is sent to RAM 6, and the output of CCD3 is sent to R.
It is also stored on AM9. At this time, the data stored in the RAM 4 is read out by the ADDRESCOUNTER 2 which is address driven by a 24 MHz clock signal. ADDRESCOUNTER2 is composed of three blocks of counters, counters 1 to 3, as shown in the left block of FIG. Counter 1 controls the address of RAM4.5 that stores the data of CCD1, and counter 2 controls the address of RAM6.5 that stores the data of CCD2.
Counter 3 controls RAM 8.9. Counter 1 sets the reading start point as PIT) SW (PRE
SET DATA 5WITCI H), reading starts from address 52 and data up to address 4948 is read.

The reason for setting the data of each COD to be the 52nd instead of the 1st and ending at 4948 instead of 5000 is as follows. In other words, 1. The MTF of the optical lens that projects the image onto the CCD gradually decreases toward the periphery. This is to truncate the data and equalize the quality of the left and right ends. In this case, as shown in Figure 2, it is sufficient to provide six areas including overlap, so if calculated based on Figure 2, (312,5X3-9
18)/3=3.25mm, converted to pixel is 3.25
X 16 = 52 P 1xel.

When the value of counter 1 reaches 4948, it outputs a match signal and opens gate 1, allowing a 24MHz clock signal to be input to counter 2. At the same time, 5EL13 is switched so that data from ccD2 is output from 5ELII. Make it.

The counter 2 is inputted as preset data with an 8-bit A/D-converted value of the converted voltage value from the thermistor 21 attached to detect the temperature at the mounting bracket 2o of the CCD 2 placed in the center. There is. This value is set to be 52 at 20°C.

This setting is applied to the A2 amplifier as a DC offset by the VRI. By turning the VRI, the reading start point of the CCD 2 can be changed continuously by 12 times.

In this example, settings were changed analogously, but they can also be changed digitally using dip switches and an adder.

The data in RAM 6 is read by counter 2 from address 52 to address 4948 near 20"C. When the value of counter 2 reaches 4948, a match signal is output in the same way as counter 1, gate 2 is opened, and the 24MHz signal is sent to counter 3. At the same time, 5EL14 is switched so that the data of CCD3 is outputted from 5EI-12.

As with counter 2, the preset data for counter 3 is input by A/D converting the output voltage from the thermistor 21 attached to the mounting bracket 1-20 of CCD 3, and becomes 52 when the temperature is around 20℃. This setting method is the same as counter 2, but
Since CCD3 is twice as far from CCD1, the temperature correction value becomes larger, so the te degree of the amplifier of A4 is made larger than that of A2. Note that the expansion value at the CCD part is due to the expansion of the mounting member as mentioned above.
Since it cannot simply be doubled, it must be determined based on actual measurements.

The data in RAM8 is also 5 at around 20℃ with counter 3.
Read from address 2 to address 4948. 49 on counter 3
When 48-52=4896 clock signals are input,
Total number of data is 918X16=4896X3=14688
, and one line of main scanning is completed. The second line is
Data in RAM5.7.9 is read out sequentially.

If the temperature rises above 20°C, assuming that CGD I is fixed as shown in Figure 8, CCD2 will expand and move away, so at the same point on the document, CCD2 will move downstream of the document reading section. Move to the side.

In the configuration shown in Fig. 5, when the temperature rises, the electrical resistance value of the thermistor 21 decreases, and the voltage of A2 increases.
/ The D conversion value increases, the RAM 6 continues to be read, and the start address value increases to compensate for the seam deviation.

The amplification degree of A2 in FIG. 5 is determined by the resistance value of the thermistor 21 and R
The A/D conversion value per 1° C. is determined based on the resistance value of 1, etc. so that it becomes the desired correction value.

If the seam is misaligned, the total number of data will be incorrect if the final read address of the CCD 3 is fixed, so another counter 4 is provided, and data is read out by the counter 3 until the total number of data is reached. When the counter 4 reaches the total number of data, the gate 2 is closed and the operation of the counter 3 is stopped.

In FIG. 5, the temperature is detected by the thermistor 21, but other means, such as a thermocouple, can also be used by increasing the amplification of A2 or A4.

In the example shown in FIG. 5, compensation is made assuming that the positions of other CCDs vary with CCD2 as a reference, but compensation can be made in the same way even if the positions of other CCDs vary left and right with CCD2 as a reference. An example of the configuration in this case is shown in FIG.

Counter 5 is a counter that counts the number of scan data in order to change the CGDI scan end point depending on the temperature.The counter 5 presets the value of A/D1, changes the count number, and changes the upper limit value of counter 1. When counter 5 finishes counting, gate 3 is closed, clock signal input to counter 1 is stopped, gate 1-1 is opened, and counter 2 is closed.
starts counting. The reading start and end of this counter 2 are fixed. Counter 3.4 is the same as in FIG.

In this way, it is better to use CCD2 as a reference, since CCD2 and CC
The temperature adjustment of D3 becomes the same, so A2 and A4
This makes it easier to set the amplification level of the amplifier.

When the temperature range to be compensated becomes large, the change in resistance value of a thermistor is not linear, so a deviation occurs between the temperature deviation compensation and the actual value. In this case, after A/D conversion, RO
M may be connected, a compensation value may be stored in the ROM data, and compensation may be performed using that value. In this case, the relationship between temperature and compensation value may be non-linear. In addition, the temperature to be detected is not only detected by the mounting member, but also by installing multiple temperature sensors such as the COD itself and other mounting members, and converting the detection output from each temperature sensor into A/D and storing it in the ROM. Compensation as a function of temperature is also possible.

An example of this is shown in FIG.

It is desirable to have temperature compensation in the middle chamber while scanning documents, but if this is done constantly, the A/D converter may malfunction due to external noise, etc., and the output may fluctuate due to quantization errors, causing seams to be Since there is a risk of partial loss, it is recommended to prohibit the conversion operation of the A/D converter while scanning the document.

FIG. 8 schematically shows the case where one line of composite data is created from the data read by CCDs 1 to 3.

In addition, the aforementioned 5YNC signal, the output signals of CCD 1 to 3,
If you show the timing of the output signal of RAM4.6.8, the scan start signal, and the A/D conversion hold signal, the 9th
As shown in the figure.

Furthermore, FIG. 10 is an explanatory diagram showing a case where the CCD 2 is shifted to the left in the drawing due to two thermal expansions.

In the above embodiment, a case has been described in which the original image is reduced through an optical lens and read by a CCD, but the present invention is not limited to this. The present invention can also be applied to a close-contact type image reading device of the same magnification.

(Effect) Since the present invention has the above-described configuration, each CCD
Because the read data in the regret area is always output during synthesis, high-quality data that is less affected by MTF reduction, illuminance reduction, and magnification changes can be obtained, and the equal reliability of reading can be improved. can.

[Brief explanation of the drawing]

The figures are all for explaining an image reading device according to an embodiment of the present invention, and FIG. 1 is a schematic configuration diagram of the image reading device;
Figure 2 shows the optical lens and CCD of the image reading device.
Fig. 3 is a front view showing the installation state of the optical lens and CCD, Fig. 4 is an overall control block diagram, and Figs. 5, 6, and 7 are blocks of main parts. Fig. 8 is an explanatory diagram when the read data of each CCD is combined into one line, Fig. 9 is a timing chart, Fig. 10 is an explanatory diagram showing the case where the CCD is shifted due to thermal expansion, and Fig. 11 is an explanatory diagram when the reading data of each CCD is combined into one line. FIG. 3 is a diagram for explaining shared reading areas on a document surface. 1... Original, 9... Optical lens, 10
...CCD, 21...Thermistor. Figure 1 Figure 2 Figure 3

Claims (1)

[Claims] (1) An illumination means for illuminating the document surface, and a plurality of illumination means for scanning the document line by line and converting optical signals from the document surface into electrical signals whose reading area is divided into a plurality of sections along the main scanning direction. a photoelectric conversion element, and a memory that stores read data from the photoelectric conversion element at predetermined addresses, respectively;
It is equipped with a counter means for controlling the read start and end addresses of the memory data, and a switching control means for synthesizing the data from each photoelectric conversion element. An image reading device characterized in that the image reading device is controlled so that data of a document image area is projected symmetrically.