JPS63294177A - Circuit for correcting sensor output - Google Patents

Abstract

PURPOSE:To miniaturize the size of the titled circuit and to simplify the constitution of said circuit, by providing a reproducing, subtracting, and dividing functions attained by arithmetic operation to one nonvolatile memory. CONSTITUTION:A black output Vd is inputted to an AD converter 1 through an analog signal line 5 before an original is read and the digitized output of the converter 1 is written in a black output RAM 3 by opening a gate 2. After these operations are completed, the original is read and the image output Vs+Vd is inputted to the AD converter 1 through the analog signal line 5. When the image output Vs+Vd are inputted, the digitized output of the converter 1 is inputted to the address of a ROM 4 through a digital signal line 6, since the gate 2 is closed. Synchronously with the digitized output, the black output Vd stored in the black output RAM 3 is inputted to the other address of the ROM 4 through a digital signal line 7. In the ROM 4, a sensitivity output Vp=alphaXVd is previously calculated from the black output Vd and the black output Vd is subtracted from the image output Vs+Vd, and then, the Vs/Vp which is obtained by dividing the result Vs of the subtraction Vs+Vd-Vd by the sensitivity output Vp is stored. Therefore, the sensor output correcting circuit having the same accuracy as that of conversional ones have can be reduced in size and simplified in constitution.

Description

Detailed Description of the Invention [Industrial Application Field] The present invention is directed to a fully contact type sensor in an image reading device such as a color or multilevel facsimile machine which requires a small size and a high S/N ratio (signal-to-noise ratio). The present invention relates to an output correction circuit.

[Conventional technology]

Regarding the conventional sensor output correction circuit, see Japanese Patent Application Laid-Open No. 61-1
It is described in No. 2175. FIG. 3 is a block diagram showing an example of this conventional sensor output correction circuit. Third
In the figure, 1 is an A/D (analog/digital) converter, 2 is a gate (digital switch) that opens when storing the output of a black original, and 3 is a volatile memory for storing the output when reading a black original (black output RAM). , 9 is a subtraction circuit, 10 is a division circuit, 11 is a gate that opens when storing the sensitivity output,
12 is a volatile memory for storing sensitivity output (sensitivity output RAM
).

FIGS. 4(al) to (e) are waveform diagrams illustrating waveforms of various parts of the circuit when a document containing a shading pattern is read using a fully contact type sensor.
7-1.17-2.17-3 is a shading pattern, ], 8
, 1.8-1.18-2.18-3 is the ideal light receiving element output (image output), 19 is the black output, 20.20-1.20
-2.

20-3 is image output, 21 is white output, 22, 22-1
, 22-2.22-3 is the difference between the image output and the evaporation power, 23 is the sensitivity output, and 24.24-1.24-2.24-3 is the output of the calculation result. The operation in Figure 3 is shown in Figures 4 (a) to (e).
This is explained by: FIG. 4(a) shows that shading patterns are drawn in the order of shading patterns 17-1, 17-2, and 17-3 on a white original. When the light-receiving element for reading this document is at the reading position 16 in the figure, the ideal light-receiving element output (image output) 18 is the output 18-1 of the same density portion, as shown in FIG. 4(b). 18-2.18-
3 must have the same output amplitude. However, as shown in FIG. 4(c), the actual image output 20, including the gray pattern output 20-1.20-2.20-3 between white and black, is unbalanced along with the daily output 21 and the evaporation output 19. It becomes uniform. This is due to non-uniformity in light intensity on the document surface, non-uniformity between light receiving elements, and non-uniformity in the reading circuit, and is permanently fixed non-uniformity.

Therefore, in the conventional example shown in FIG. 3, before reading the original, the evaporation power 19 is input to the A/D converter 1 through the analog signal line 5, the digitized output is passed through the digital signal line 6, and the gate 2 is opened. The data is written into the evaporation output RAM 3 through the digital signal line 7.

Next, connect the daily output 21 to analog signal line 5 in the same way as A
/ D converter 1, and its digitized output is input to the subtractive side input of the subtraction circuit 9 through the digital signal line 6. At this time, gate 2 is closed and the evaporation power RA
M3 enters the read state, and the stored evaporation power 19 is input to the subtraction side input of the subtraction circuit 9 through the digital signal line 7. The difference between these two inputs (output black output), that is, the sensitivity output n, is passed through the digital signal line 13 to the gate 11.
Open and pass the digital signal line 14 to the sensitivity output RA.
Write in M12. FIG. 4(d) shows the waveform of the sensitivity output 23.

After the above operation, when reading the document containing the images 7-1, 17-2, and 17-3 in FIG.
In Figure 4 (output of cl 20-1.20-2.20-3
After the image output 20 including the image output 20 passes through and is digitized by the A/D converter 1, the subtraction circuit 9 subtracts the evaporation output 19 from the image output 20. This output 22-1.22-2゜22
The waveform of 22 including -3 (surface black output) is the 4th and 3rd waveform.
- It is shown together with the above sensitivity output 23 in Figure (d). This output 2
2 passes through the digital signal line 13, but the gate 11 is closed and is input to the dividend side input of the division circuit 10.

One sensitivity output RAM 12 is in a read state, and the stored sensitivity output 23 passes through a digital signal line 14 and is input to the divisor side input of the divider circuit 10. With these two inputs, an output (calculation output) 24 of the calculation result is output to the digital signal line 15. By the above correction operation, as shown in Fig. 4(e)
As shown in FIG. 2, the waveform of the calculated output 24 of (surface car black output)/sensitivity output is changed, including the outputs 24-1, 24-2, and 24-3, which are proportional to one novel of density and light without non-uniformity.

Note that either one or both of the division function and the subtraction function shown in FIG. 3 are provided in respective nonvolatile memories or in the same nonvolatile memory, resulting in a small, high-speed, highly accurate correction circuit.

FIG. 5 is a cross-sectional view in the scanning direction illustrating the structure of a complete contact type sensor when reading an original. In Figure 5,
25 is a complete contact type sensor, 26 is a glass substrate, 27 is a light receiving element, 28 is a protective film, 29 is a document, and 4 and 30 are illumination lights. The complete contact type sensor 25 includes a glass substrate 26, a plurality of light receiving elements 27 arranged in a line in the scanning direction on the glass substrate 26, and a protective film 28 that protects the light receiving elements 27. Note that although the number of light receiving elements 27 in a typical A4 size sensor is 1728, three elements are shown enlarged in the figure. Fully contact type sensor 2 with this configuration
The document reading operation in step 5 will be explained.

The sensor 25 and the original 29 are in almost complete contact with each other, and the illumination light 30 is applied from the back side of the sensor 25, passes through the glass substrate 26, and illuminates the original 29. The reflected light from the original 29 is photoelectrically converted by the light receiving element 27 to obtain a signal output Vs.

The reflected light from the original 29 is greatest when the original is white, and the signal output Vs is also the largest value in proportion to the reflected light (Vs=Vp
). As the reflection density of the original 29 increases, the reflected light decreases, and in the case of an ideal black original, the signal output Vs becomes zero (Vs=O). However, when the illumination light is irradiated from the back side, it passes through the glass substrate 26 and directly hits the light receiving element 27.
is illuminated by light, and outputs an invalid component Vd that is completely unrelated to the information on the document 29. Therefore, when a black original is actually read, the signal output (black output) Vs is an invalid component Vd (V
s - Vd ), and the signal output (output) when reading a white original is the invalid component (black output) Vs and the sensitivity output V
The sum of p is (Vs+Vp).

[Problem that the invention seeks to solve]

The above conventional technology does not consider the correlation between the black output and the output output of the fully contact type sensor, and includes volatile memories (black output RAM 3 and sensitivity output RAM 4) that store the black output and sensitivity output respectively, and a plurality of gates ( Since the gate 2 and the gate 14) are required, there is a problem that the circuit scale becomes large and complicated together with the subtraction circuit 9 and the division circuit 10, etc.

An object of the present invention is to provide a sensor output correction circuit which has functions equivalent to those of the conventional example, and which is miniaturized and has a simplified configuration.

[Means for solving problems]

The above purpose focuses on the correlation between the black output and the output output of a full contact type sensor, stores only the output output or black output of the sensor as a reference output, and stores other black outputs and sensitivity outputs or other outputs necessary for correction. A means for reproducing the sensitivity output by calculation from the above-mentioned stored output output or reference output value for black output, subtracting the black output from the image output of the sensor, and dividing the subtraction result by the sensitivity output, and the above-mentioned calculation This is achieved by a sensor output correction circuit in which a reproduction function, a subtraction function, and a division function are provided in one nonvolatile memory.

[Effect]

In the sensor output correction circuit described above, when reading the original 29 using the illumination light 30 from the rear side with the fully contact type sensor 25 having a plurality of light receiving elements 27 as shown in FIG. Since light hits the light receiving element 27 and outputs an invalid component Vd unrelated to the information on the document,
The black output when reading a black original is Vd, and the output when reading a white original is sensitivity output Vp and black output V
d, where the black output Vd is a value proportional to the irradiation light 30 from the back, and the sensitivity output Vp is the value proportional to the irradiation light 30.
has a value proportional to the reflected light after hitting the original 29, so the black output Vd and the sensitivity output vp are in a proportional relationship of 1.7.

Vp-αX■d, α=proportionality coefficient (1), and when using this relationship to calculate the black output Vd from the output output (Vp+vd), (vp+vd)=(1+α)XVd.

va= (Vp +Vd)/(1+α) (2) Therefore, by dividing the output output (Vp+Vd) by (1+α), the black output Vd can be obtained, and from equation (1), the black output V
The sensitivity output Vp can be obtained by multiplying d by α, so by storing only the sensor output (vp+vd) or the black output Vd as the reference output, the other black output Vd and sensitivity output vp or sensitivity output necessary for correction can be obtained. If you calculate and reproduce the other sensitivity output Vp using formula (11, +21), subtract the black output Vd from the image output (Vs + Vd) necessary for correction, and divide the subtraction result Vs by the sensitivity output Vp, The sensor output is corrected as the division result Vs/Vp.

〔Example〕

An embodiment of the present invention will be described below with reference to FIGS. 1 and 2.

In FIG. 1, 1 is A/D (Analog/D, 8).

2 is a gate, 3 is a volatile memory (black output RAM) for storing an output when reading a black original, and 4 is a nonvolatile memory IJ (ROM). Figure 1 shows the subtraction circuit 9 in Figure 3.
, divider circuit 1o, sensitivity output RAM 12 and gate 11 are R
This is a configuration replaced with OM4.

With this configuration, first, before reading a document, the black output Vd is input to the A/D converter 1 through the analog signal line 5, the digitized output is passed through the digital signal line 6, and the gate 2 is opened to connect the digital signal line to the A/D converter 1. 7, black output R, AM
Write in 3. After this operation is completed, the original is read. The image output Vs+Vd is input to the A/D converter 1 through the analog signal line 5. At this time, gate 2 is closed and its digitized output is sent to digital signal line 6.
, and enter the address of ROM4. In synchronization with this, the black output Vd previously stored in the black output RAM 3 is inputted to another address in the ROM 4 through the digital signal line 7. In other words, the address of the ROM 4 has upper bits and lower bits, which correspond to the surface output Vs + Vd and the black output Vd, respectively. Evaporation power Vd is stored in ROM4 in advance.
Calculate the sensitivity output Vp=αXVd from , and further calculate the image output V
The result of subtracting the evaporation power Vd from s+Vd and dividing the subtracted value Vs by the sensitivity output Vp is stored as Vs/¥p.

FIG. 2 is a diagram showing an example of the memory contents of the nonvolatile memory IJ (ROM) 4 shown in FIG. FIG. 2 shows the contents of the ROM 4 when each address input is 3 bits (8 gradations) and the sensitivity output Vp and evaporation output VdO ratio α=1. For example, if the image output Vs+Vd=''6'' and the evaporation output Vd=''4'', then the following calculation formula is used to calculate ((Vs+Vd)-Vd)X
8/'Vp, Vp-aXVd((Vs+Vd)-
Vd ) x8/'Vd = (6-4) The value obtained by subtracting "3" is memorized.

When inputting data to the ROM 4 with 8-bit accuracy using this circuit, the capacity of the ROM 4 can be 216=64K bits, and the calculation result can be obtained with 7-bit accuracy (128 gradations).

According to this embodiment, the subtraction circuit 9 and the division circuit 10 of the conventional example
The sensitivity output RAM 12 and the gate 11 can be replaced with the ROM 4, and even compared to the case where the subtraction circuit 9 and the division circuit 10 of the conventional example are provided in the ROM, it is possible to achieve the same accuracy and circuit structure as the conventional example. It can be simplified on a small scale.

In this embodiment, the evaporation power ■d is used as the reference output of the sensor to reproduce the sensitivity output Vp, but the output power Vp+Vd is used as the reference output, and the evaporation power Vd and the sensitivity output V
p may be played back.

〔Effect of the invention〕

According to the present invention, it is possible to realize an output correction circuit for a complete contact type sensor with the same precision as the conventional one, on a small scale, and in a simplified manner.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing an embodiment of the sensor output correction circuit according to the present invention, and FIG. 2 is a block diagram showing an embodiment of the sensor output correction circuit according to the present invention.
ROM) memory contents, Figures 3 and 11 are block diagrams illustrating conventional sensor output correction circuits, and Figure 4 shows the various parts of the circuit when reading more originals including light and shade patterns using a full contact type sensor. FIG. 5 is a sectional side view of the structure when a document is read by a completely contact type sensor. 1... A/D (analog/digital) converter 2...
・Gate 3...RAM (volatile memory) 4...ROM (nonvolatile memory) 25...Complete contact type sensor 26...Glass substrate 2
7... Light receiving element 28... Protective film 29...
・Original 30...Illumination light Vd...Evaporation power Vs+Vd...Image output vp+Vd
... Output output Vp ... Sensitivity output α ... Proportional coefficient, 12°

Claims (1)

[Claims] 1. Means for A/D converting the output of a fully contact type sensor having a plurality of light receiving elements, means for storing a reference output of the sensor, and black output and sensitivity output from the reference output. A sensor output correction circuit comprising: means for reproducing a black output from the image output of the sensor; and means for dividing the subtraction result by a sensitivity output. 2. The sensor output correction circuit according to claim 1, wherein the reproduction function, subtraction function, and division function of the black output and sensitivity output are provided in the same nonvolatile memory.