JPS5829674B2 - Binary signal control circuit - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a binary signal control circuit in a photoelectric conversion device such as a facsimile machine or an optical reader (OCR).

A binarization circuit is provided for optically capturing an image of a document in a photoelectric conversion device, and converting the resulting analog signal into a binary value using a predetermined threshold.

Generally, the background density and contrast of documents used in such photoelectric conversion devices are not constant, so it is not possible to obtain a good image by binarizing the analog signal obtained by photoelectric conversion using a fixed threshold value. There are many cases.

Therefore, in the past, a background density compensation circuit was installed, and after controlling the midpoint value and DC level of the analog signal obtained by photoelectrically converting the original according to the background density and contrast of the original, the circuit was converted into a binary value based on a certain threshold value. The image quality has been improved by changing the image quality.

The conventional background density compensation circuit will be explained below with reference to FIGS. 1 to 4.

FIG. 1 shows the configuration of the background density compensation circuit.

An image of the original 1 is formed on a photoelectric conversion surface on a photoelectric conversion element 3 such as a photodiode array using a lens 2, and a photoelectric conversion signal is obtained.

After this signal is amplified by an amplifier 4, it is applied to an automatic gain control and DC potential control circuit 5, as well as a black peak value detection circuit 6 and a white peak value detection circuit 7.

FIG. 2a shows the optical conversion signal amplified by the amplifier 4, and the black level of the original is at a high potential.

The black peak value detection circuit 6 detects point A of the waveform in the figure,
The white peak value detection circuit 7 detects point B in the figure.

The control signals obtained by both detection circuits are applied to the automatic gain control and DC level control circuit 5, which makes the difference between the black peak value and the white peak value of the photoelectric conversion signal constant, and also reduces the white peak value to zero level. Control the DC potential so that

FIG. 2 shows the photoelectric conversion signal controlled by the automatic gain control and DC potential control circuit 5.

Finally, this signal is applied to the binarization circuit 8 and is reduced by two levels according to a certain threshold level.

The threshold level is usually set to an intermediate potential between the black peak value and the white peak value.

Line C in FIG. 2 shows the threshold level, and FIG. 2 C shows a binary signal of a photoelectric conversion signal that has been binarized by the threshold.

As explained above, according to the background density compensation circuit, the threshold value is always an intermediate potential between the black peak value and the white peak value.
Good results can be expected for originals with uniform background density and contrast.

However, in actual manuscripts, the background density and contrast are often not constant even on the surface of a single manuscript.

For example, a second document copied using a copying machine is often sent by facsimile, but such documents often have non-uniform background density or low contrast.

Particularly when the size of the original document is small compared to the size of the photosensitive paper of the copying machine, the density of the peripheral part (section D) that protrudes from the original document may be lower than the center, as shown in Figure 3. A document whose background density is extremely different from that of the portion (E portion) is obtained.

When photoelectrically converting the portion of line F in the figure for such an original, the photoelectric conversion signal becomes as shown in Figure 4a, and the signal obtained by controlling this signal using a conventional background density compensation circuit is as shown in Figure 4. It becomes like that.

It is clear from the figure that a good image cannot be obtained even if this is binarized using a fixed threshold value as shown by line G in the figure.

This is because the peripheral area (part D) in Figure 3 is determined to be the white peak value, and for such originals, it is necessary to detect the background density in the center of the original as the white peak value and convert it into a binary value. However, it is very difficult to determine this on the circuit.

The foregoing description has been given of copied manuscripts, but even in general manuscripts, especially catalogs, there are many cases in which the background density and contrast differ depending on the portion, and the same can be said for such manuscripts as described above.

The present invention provides a method that allows such originals to be binarized using an optimal threshold value.

The present invention will be explained in detail below with reference to FIGS. 5 to 8.

FIG. 5 is a diagram showing an embodiment of the present invention.

In the figures, parts indicated by the same numbers as in FIG. 1 indicate the same parts as in FIG. 1.

The analog signal photoelectrically converted by the original 1, lens 2, and photoelectric conversion element 3 is amplified by an amplifier 4, and controlled to a constant amplitude by an automatic gain control and DC potential control circuit 5.

Then, the signal is applied to a first binarization circuit 9, a second binarization circuit 10, and a third binarization circuit 11, each having a threshold value with a constant potential difference, and three types of binary converted into a signal.

In FIG. 6a, an analog signal controlled to a constant amplitude,
The threshold H1 of the first binarization circuit, the threshold H2 of the second binarization circuit, and the threshold H3 of the third binarization circuit are shown.

The threshold value H1 and the threshold value H3, and the threshold value H3 and the threshold value H2 are each a constant potential difference ■.

It is configured to have.

6, d and f are threshold values H1, H2 and H, respectively.
3 shows a binary signal converted into a binary signal.

Among these binary signals, the first binary signal and the second binary signal are connected to the first change point detection circuit 12 and the second binary signal, respectively. It is added to the change point detection circuit 13 of No. 2.

These change point detection circuits output a pulse of a certain width (referred to as a change point pulse) at the point where a binary signal changes from white to black (or from black to white), and detect the change point. The change point pulse is applied to a change point number comparison circuit 14.

C and e in FIG. 6 are the first change point detection circuits 12, respectively.
and shows the waveform of a changing point pulse produced by the second changing point detection circuit 13.

The changing point number comparison circuit 14 compares the numbers of the two changing point pulses, and if one of the changing point pulses is larger than the other,
the first, second and third pulses so that the number of pulses for both is equal.
A threshold control voltage is generated to control the threshold of the binarization circuit.

That is, the binarization circuits 9, 10 and 11, and the change point detection circuit 12
and 13, and the closed loop formed by the change point number comparison circuit 14, the thresholds H1 and H2 are moved to a position where the number of corresponding change point pulses becomes equal and stabilized.

The threshold values H1°H'2 and HI3 in a of FIG. 6 thus represent stable threshold values.

Figure 7 shows the relationship between the threshold value and the number of changing point pulses; the number of changing points is zero at the threshold corresponding to the white peak value and black peak value potential, and the number of changing points exists between the white peak value and the black peak value. The number of change point pulses should be at a maximum at some threshold value.

Now, in this figure, if the first threshold value H1 and the second threshold value H2 are at the positions shown in the figure, then the change point comparison circuit 1
The output of 4 moves both the thresholds H1 and H2 in the direction where the number of change point pulses is large, that is, in the direction of the threshold value H1, and positions H'1 and H/l! where the number of change point pulses is equal. It becomes stable.

In this way, if a binary signal is extracted using the threshold H5 that exists at an intermediate potential between the thresholds H'1 and HI3, the binary signal will be approximately equal to 2 at the level where the number of changing point pulses is the greatest. It will be valued.

FIG. 6g shows a binary signal binarized by this method.

An example of a circuit for embodying the above embodiment is shown in FIG.

In the figure, 1 indicates a photoelectric conversion signal controlled by the automatic gain control and DC potential control circuit 9 of FIG.

C1, D, C2, D2 and C3, D3 indicate clamp circuits for changing the threshold according to the control voltage generated by the change point comparison circuit 14, and El and R2 respectively indicate the threshold H.
1 and a threshold value H3, and a storage battery for keeping constant the potential difference between the threshold value H3 and the threshold value H2.

ST1. ST2 indicates first, second and third binarization circuits, respectively.

12 and 13 are first and second change point detection circuits, respectively.

Reference numeral 14 denotes a change point comparison circuit, which is composed of two analog switches connected in series.

One end of the first switch S1 of the analog switch is connected to a negative potential -■, and the switch is closed only while the first change check circuit 12 is outputting a change point pulse. One end of S2 is connected to a positive potential 10, and the switch is closed only during the period when the second change point detection circuit 13 is outputting a change point pulse.

Further, the common end of the switches S1 and S2 is connected to a variable resistor R1 through a resistor R2.

Variable resistor ■R1 sets the threshold value H when there is no photoelectric conversion signal.
This is to adjust so that 1, H2, and R3 are at potentials near the midpoint between the white peak value and the black peak value.

In the circuit of FIG. 8, first the thresholds H7 and R2 are set to the seventh
If the position is as shown in the figure, the number of changing point pulses corresponding to the threshold H1 is greater than the number of changing point pulses corresponding to the threshold H2, so the analog switch S1 has a larger number of changing point pulses than the analog switch S2. When the number of point pulses is large, the pulses close frequently, and the output of the change point comparison circuit 14 moves toward a lower potential, resulting in C1 and Dl.

A binarization circuit ST4. The DC potential of the signals entering ST2 and ST3 is lowered, and the threshold value is moved to the black peak value side.

This movement of the threshold value continues until the number of first and second change point pulses becomes equal, and is stabilized at the threshold value H'.

In this way, in a stable state, the third threshold value H'3 is set to 2.
The converted binary signal is taken out from the binary conversion circuit ST3.

Note that R7, C4, C5, and C6 constitute a time constant circuit,
The threshold value does not change much during the period in which one to several lines of the original are photoelectrically converted.

Above, we have explained the method of opening and closing the analog gate to control the threshold value using the change pulse generated at the point of change from white to black in a binary signal. It is also possible to provide the same function as above by counting with a frequency counter, comparing the number of change points corresponding to the threshold values H1 and R2 with a comparison circuit, and controlling the threshold value with the output signal of the comparison circuit.

In addition, in FIG. 5, the photoelectric conversion signal is amplified by an amplifier 4, and then a black peak value detection circuit 6 and a white peak value detection circuit 7.
The waveform is controlled by the automatic gain control and DC potential control circuit 5, and then added to the threshold control circuit shown in FIG. Since it is not restricted by the sheath DC potential,
It is also possible to directly apply the signal amplified by the amplifier 4 of FIG. 5 to the threshold control circuit of FIG. 8.

As explained above, according to the present invention, regardless of the background density and contrast of the document,
Or, regardless of whether the image is white on a black background, the threshold of the binarization circuit is automatically set to a level where there are many points of change between white and black, and good image reproduction can be expected, and the effect is extremely high. It's a dog.

[Brief explanation of drawings]

Figure 1 is a configuration diagram showing a conventional binarization circuit, Figure 2 a, b
, c is a signal waveform diagram for explaining the operation of the binarization circuit shown in Fig. 1, and Fig. 3 is an explanatory diagram showing a document that is the source of the photoelectric conversion signal applied to the binarization circuit shown in Fig. 1. , FIG. 4 a and b are waveform diagrams showing the photoelectric conversion signals of the original shown in FIG. 3, FIG. A signal waveform diagram for explaining the operation of the embodiment, FIG. 7 is a characteristic diagram showing the relationship between the threshold value of the binarization circuit based on the present invention and the number of black and white change points of the photoelectric conversion signal, and FIG. 5 is a circuit diagram showing a specific example of the embodiment of FIG. 5. In the figure, 9, 10, and 11 are first, second, and third binarization circuits; 12 is a first change point detection circuit; 13 is a second change point detection circuit, and 14 is a change point number comparison circuit.

Claims (1)

[Claims] 1. In a binarization signal control circuit that binarizes an analog signal obtained by photoelectrically converting a document pattern image using a certain threshold level, a first binarization circuit having a first threshold;
a second binarization circuit having a second threshold; and a third binarization circuit having a third threshold that is an intermediate potential between the first threshold and the second threshold.
a change point comparison circuit that compares the number of change points of the output signals of the digitization circuit and the first and second binarization circuits to control the threshold values of the first, second and eighth binarization circuits; 1. A binary signal control circuit comprising: a binary signal that is binarized at a threshold value at which the number of change points of a document is greatest;