JPH0130341B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a photoelectric conversion device for facsimile, and more particularly to a photoelectric conversion device for facsimile having means for compensating for conversion distortion of the photoelectric conversion device appearing in an output signal of a photoelectric conversion section.

In recent years, facsimile transmitters are equipped with CCD (Charge
Coupled Device) and MOS (Metal Oxide
A photoelectric conversion device using an image sensor array developed by Semiconductor technology is used. This involves a light source that illuminates the document to be read, an optical system that images the optical information on the illuminated document onto the image sensor, and the image sensor converts light into electricity. It is something to do. A photoelectric conversion device for facsimile using an image sensor is
(1) The image sensor itself has a scanning function similar to an electric circuit, allowing high-speed scanning without inertia; (2) Integrated using semiconductor technology, allowing for miniaturization and high reliability; (3) Sensitivity (4) Control is possible using digital signals, making signal processing easy.

However, a facsimile photoelectric conversion device using this image sensor has a problem in that the conversion coefficients of the light source, optical system, and sensor change with respect to each scanning point on the document, causing distortion in the output signal. For this purpose, corrections are made using a reflector in the illumination section, corrections using optical masks, lenses or fibers, corrections according to the integral value of the read signal (floating slice), etc. However, in practice, this is quite difficult. It requires extensive skill and precise adjustment, and is difficult to produce good results.

SUMMARY OF THE INVENTION An object of the present invention is to provide a photoelectric conversion device for facsimile in which a correction means that can be easily corrected is added to the above-mentioned drawbacks of the prior art.

An object of the present invention is to provide a conventional facsimile photoelectric conversion device that further reads a uniformly white diffuse reflection plate that also serves as a document feeding guide when the document is not in the reading position, and performs a photoelectric conversion operation in advance to convert the reading. This can be achieved by measuring and storing the characteristics and adding a means for evaluating the photoelectric conversion signal according to the measurement results.

The present invention will be explained below with reference to Examples.

FIG. 1 is a block diagram of one embodiment of a photoelectric conversion device for facsimile according to the present invention. 1st
In the figure, reference numeral 10 denotes a document that is moved by a drive means (not shown) that feeds paper according to a predetermined rule;
0 is an illumination means consisting of a light source for illuminating the original 10, a condensing section, a light shielding section, etc.; 30 is an optical system consisting of a reflecting mirror (which can be omitted depending on the case), a lens, a light shielding section, etc.;
40 is an optical-to-electrical conversion circuit including an image sensor array such as a CCD or MOS and its driving circuit; 5
0 is an amplifier circuit that performs impedance conversion and amplification, 60 is a correction circuit for conversion distortion, and 70 is a photoelectric converter for facsimile that receives a control signal given from the outside to the photoelectric converter for facsimile according to a certain procedure. This is a control circuit that operates the conversion device.

In this example, the light source is a straight tube fluorescent lamp, and the optical system is
F4.0f = 30mm lens, image sensor, 2048
Uses BIT's CCD image sensor.

All parts of this embodiment except for the correction means 60 and the control means 70 are known, and these basic parts operate at the timing shown in FIG. 2 and exhibit the characteristics shown in FIG. 3.

In FIG. 2, the signal CPO is the reference timing for operating the entire facsimile photoelectric conversion device, and in this embodiment, a 100 kHz clock is used. signal
SCO is a photoelectric conversion command signal, which has a pulse width of approximately 10 μs. When this signal is input, the image sensor drive circuit generates a photoelectric conversion command signal SC1 synchronized with CPO, and issues a photoelectric conversion command to the image sensor array. give. CP1 is a timing signal with a frequency 1/2 that of the reference clock, which coincides with the transfer timing of the output data from the image sensor array, and is used in the correction means connected later or in the part that utilizes the photoelectric conversion signal. VO is a photoelectric conversion output signal before correction. As shown in FIG. 3, this signal includes non-uniformity of illumination, non-uniformity of the optical system, and non-uniformity of the image sensor amplification means.
That is, (a) shows the non-uniformity of illumination, with the horizontal axis representing the position on the scanning line and the vertical axis representing the illuminance on the document surface. In the case of this example, if the correction reflector is not attached, the edges and center will be approximately
There is 50% heterogeneity. (b) is the non-uniformity of the optical system, and the vertical axis corresponds to the transmittance.
There is a difference of ~30%. (c) is the conversion characteristic of the image sensor. Non-uniformity in sensitivity of each light-receiving element inside the image sensor array, transfer efficiency of CCD, etc. are added, and as shown in the figure, pattern error, non-uniformity of semiconductor, etc. 10% appear. Therefore, the output signals are (a)(b)(c)
The synergistic effect of (d) (corresponding to VO) is obtained, and it cannot be used as is.

The details of each part will be explained below.

FIG. 4 is an electrical circuit diagram excluding the correction circuit of this embodiment. In FIG. 4, 100 to 102 are resistors, 103 is a variable resistor, 104 to 107 are capacitors, 108 is a crystal oscillator, 109 to 110 are inverters, 111 to 113 are flip-flops, 114 is a NAND gate, and 115 to 118 are
MOS driver, 119 is a CCD image sensor array, 120 is an operational amplifier, 121 is a digital counter 122 is an inverter, and the names of block numbers and signal names are the same as in FIGS. 1 and 2.

Flip-flops 111 and 112, NAND gate 114 and MOS drivers 115 to 118
is a drive circuit for the CCD image sensor array 119, and a resistor 102, a capacitor 107, a variable resistor 103, and an operational amplifier 120 are an amplification circuit for the output signal of the CCD image sensor array 119. Resistor 100~101, capacitor 104~
The circuit consisting of 106 and inverter 109 is an oscillation circuit using a crystal resonator 108 and oscillates at a frequency 256 times that of CPO. The inverter 110 and flip-flop 113 shape the waveform of the output of the oscillation circuit and adjust the duty of the clock pulse to 50.
This is a circuit for setting the clock to 50. It would be good to have a stable operation for the circuit that uses this clock, but it is not essential. Counter 121 is used as a frequency divider
It generates a clock CPO with a frequency 1/128 of CP2. The control circuit 70 performs this oscillation, waveform shaping, frequency division, and signal distribution. In addition to the signals explained in Figure 2, there is also a signal indicating that the photoelectric conversion device is in the position to read the document.
There is an SOP, which is inverted as a signal by an inverter 122 and used as a command signal for the correction circuit to measure the reading conversion characteristics. That is, when the SOP is off (0 level of the binary signal), the photoelectric conversion device measures the conversion characteristics and stores the measured value according to the photoelectric conversion command SCO.

Next, this measurement operation and correction operation will be explained with reference to FIGS. 5 to 8.

FIG. 5 shows the correction circuit 60, in which 131 to 138 are inverters, 139 is a MOS driver, 140 is an AND circuit, 141 to 142 are negative logic OR circuits, 143 is a NAND circuit, and 144
is a resistor, 145 to 146 are capacitors, 147 is a variable resistor, 148 is an analog switch, 149 to 146 are capacitors, 147 is a variable resistor, 148 is an analog switch,
150 is a digital counter, 151 is RAM
(Random Access Memory), 152 is a tri-state bus buffer, 153 is an analog value data selector, 154 is a programmable voltage divider, so-called ladder resistor, 155 is an operational amplifier,
156 is a comparator.

This correction circuit uses clock inputs CP1, CP2,
It is driven by the scanning command SC1 and the reading distortion correction command TR, and when TR = "1", it is in the reading distortion measurement mode, and when TR = "0", it is in the distortion correction mode, and reads the read signal input VO and calculates the distortion. Correct the influence, binarize it, and output it as VO1. That is, when TR="1", the read distortion (in the read signal VO of the diffuse reflector) is measured by an AD converter consisting of a counter, a programmable voltage divider, and a comparator, and is stored in the RAM.

When TR="0", reading distortion data is read out from the original reading signal, and the evaluation reference signal obtained by DA conversion via a programmable voltage divider is used as a threshold value to be binarized by a comparator.

The capacitor 146 is for blocking the DC level drift of the amplifier circuit shown in FIG.
DC regeneration is performed by a clamp circuit using an analog switch 148. This block is a so-called DC regeneration circuit. The operational amplifier 155 is a voltage follower so-called impedance conversion circuit for supplying sufficient power to the comparator 156.
The counter 149 is a circuit for generating a memory address of the RAM 151, and is a circuit for generating a memory address of the RAM 151.
Reset every scan line. analog switch 15
3, when TR is on, V _St (reference value, value greater than the maximum value of VO), when off (while reading the document),
The voltage divided by the variable resistor 147 is applied to the programmable voltage divider 154. The output V _REF of programmable voltage divider 154 is input to comparator 156 and compared with input signal VO. The comparison result is VO
1, and when TR is on, it controls the counting operation of the counter 150. The counter 150 is
In addition to being reset for each scanning line by SC1,
As will be explained later, inverters 131, 13
2. The block consisting of a resistor 144, a capacitor 145, and an AND circuit 140 is a so-called differential circuit.
Reset using the reset pulse PC created by differentiating the falling point of CP1. The output of the counter 150 is provided to the programmable frequency divider 154 when the bus buffer is opened only when TR is on. Comparator 156 allows counter 150 to count only if V _REF is less than VO. Therefore, the counter 150 continues counting until the count value increases so that V _REF slightly exceeds VO. When TR is on, NAND gate 143 passes the differential pulse. Therefore, the count value when the counter stops can be stored in the RAM 151. Naturally, the counter 150 is reset by a reset pulse that is sufficiently delayed from the write pulse by a so-called delay circuit consisting of inverters 134 to 137, so that the counter 150 is reset during the storage operation.
50 outputs are guaranteed. When TR is off, the bus buffer 152 is prohibited from outputting and storage to RAM is also prohibited, but CE is always on and RAM output is permitted, and the memory contents are output according to the output of the address counter 149. . In addition, CP1 and
It is assumed that the operating timings of the clocks of CP2 match, and the frequency ratio is given by the following equation.

CP2/CP1=2 ^N , where N is the frequency division ratio N of the programmable frequency divider, that is, the maximum countable number of the counter 150.

The overall operation will be explained with reference to FIG. 6th
The figure shows whether the document is in the reading position or not.
This shows that there are two operating modes depending on whether the TR created from the SOP is on or off. (distinguish whether it is black). V1 required in facsimile and OCR requires judgment of black and white, not of large or small.
Therefore, the original is inserted, TR is turned off, and
When the SOP is turned on, V0 is able to make correct decisions based on stored data.

This operation will be explained below with reference to FIGS. 7 and 8. FIG. 7 shows the operation when TR is on.
If we extend the time axis, V0 for CP1 is
It changes as shown in the figure. V _REF shown by the dotted line increases linearly in response to clock pulse CP2, and V0
It becomes a constant value when it exceeds the value of . This value is
Store it in RAM with the differential pulse of the falling edge of CP1. This value is proportional to the conversion factor of the reading system for the following reason.

First, if there is no document and there is a uniformly white diffuse reflection plate (not explained, but it also serves as a document feed guide) at the reading position, the illuminance distribution on the reading scanning line of the reflection plate is given by the following formula: can be expressed.

E=e(x)...(corresponds to Figure 3a) Similarly, the transmittance of the optical system is F=f(x)...(corresponds to Figure 3b) The image sensor array is S=s(x)... (corresponding to Fig. 3c). However, x represents the position on the scanning line.

Therefore, if the reflectance of the reflector is R=r(x), the output V0 is V0=E×F×S×R = e(x)・f(x)・s(x)・γ( x), and γ(x) of the reflector = k (constant)
Then, V0=e(x)・f(x)・s(x)・k, and V
0 is proportional to the conversion coefficient of the reading system. Therefore, the measured value V _REF (i.e., the partial pressure ratio N if expressed by the output of the counter) is:
It shows the conversion coefficient, and this value is stored in the RAM in the form of a binary number of the voltage division ratio N.

Next, FIG. 8 shows a case where the TR is turned off and the document is actually at the reading position. If you extend the time axis, for clock pulse CP1,
When TR is on, the measured value V0 is as shown by the two-dot chain line, and when the analog switch turns TR off and the divided potential of the variable resistor is selected, the evaluation reference signal V _REF becomes as shown in the dotted line. , it is possible to provide a comparison value that compensates for the conversion characteristics of the signal V0' obtained by actually reading the original, and it is possible to accurately determine whether the output V1 is black or white based on the slice ratio set by the variable resistor with respect to _VST .

As explained above, according to the present invention, a uniformly white diffuse reflection plate is read in advance when the document is not at the reading position, the conversion characteristics of the reading system are measured and stored, and the actual reading is performed according to the stored values. Because it determines whether the photoelectric conversion signal is black or white after reading the original,
It is possible to compensate for distortion in the conversion characteristics of the reading system and to actually read the density of the document itself.

Furthermore, since a uniformly white diffuse reflector is read out, there is no need to provide a special compensation space on the document.

Note that the present invention can be modified without being limited to the embodiments.

That is, in this embodiment, a RAM is used as the storage means, but a ROM (Read Only Memory) may be used so that measurement and writing to the ROM can be performed by adding it to the photoelectric conversion device.
That is, "in advance" does not necessarily mean each time a document is inserted.

In addition, in this embodiment, the number of storage addresses of the storage means and the number of bits of one scanning line of photoelectric conversion are made to match, but if the conversion distortion of the reading system does not show a sudden change, 1/N It may be reduced to '.
In this case, in the circuit of the embodiment, the clock CP1 input to the correction circuit 60 is changed to the output CP1 of the optical-to-electrical conversion circuit.
It is sufficient to set it to 1/N'.

Furthermore, in this embodiment, regarding the SCO interval,
Although it is assumed that the intervals are equal, the present invention can also be applied to a case where one reading scan is performed by performing two operations of pre-scan and main scan. Even in this case, the condition is that the output V0 always has a certain maximum value when there is no document.

[Brief explanation of drawings]

The drawings show an embodiment of the present invention, and FIG. 1 is an overall block diagram, FIG. 2 is an operation timing chart, FIG. 3 is a distortion characteristic diagram, FIG. 4 is an electric circuit diagram excluding the correction circuit, and FIG. Figure 5 is a correction circuit diagram, 6th to 8th
The figure is an operation timing chart. DESCRIPTION OF SYMBOLS 10... Original document, 20... Light source, 30... Optical system, 40... Optical-electric conversion circuit, 50... Amplification circuit, 60... Correction circuit, 70... Control circuit.

Claims (1)

[Scope of Claims] 1. Illumination means for illuminating the original, a uniformly white diffuse reflection plate that serves as a guide for feeding the original, optical means for forming an image of optical information on the original, and converting the formed optical image into an electrical signal. A facsimile machine that includes an optical-to-electric conversion means including an image sensor array and its driving circuit, an original driving means including a function of moving an original according to a predetermined rule, and a control means for operating these means according to a predetermined procedure. In the photoelectric conversion device for use in the photoelectric conversion device, when the document is not at the reading position, the uniform white diffuse reflection plate is read by the image sensor array, and the conversion characteristics of the entire reading system are measured by performing a photoelectric conversion operation in advance. A means for storing measurement data; and a means for creating an evaluation reference signal for a photoelectric conversion signal according to the stored measurement data when actually reading a document, and generating a photoelectric conversion signal actually read from the document according to the reference signal. 1. A photoelectric conversion device for facsimile, comprising means for obtaining a signal for slicing and displaying a black and white binary image.