JPS58217908A - Detector of focus information - Google Patents

Abstract

PURPOSE:To obtain high-precision focus information by forming a negative feedback loop so that the clock frequency of a scanning ring counter is controlled by a peak value voltage, and setting the operation point so that a photoelectric converting element is nearly saturated with charges. CONSTITUTION:The oscillation output of a voltage-controlled oscillator 21 is supplied as clock pulses to the ring counter 22. The ring counter 22 consists of, e.g. 100 stages corresponding to respective photoelectric converting element groups 6 and 7, and respective phase output pulses outputted from the respective stages are supplied as read scanning pulses to the photoelectric converting elements in the groups 6 and 7, which are read. Namely, the photoelectric converting elements provided in arrays are read successively from left to right in a figure and the read direction is inverted at the left end, thus repeating the scanning operation successively.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a focus information detection device for a TV camera, a film still camera, etc., and in particular, image sensor groups are provided at multiple locations with different optical path lengths from the lens, and focus information can be obtained by each image sensor group. The present invention relates to a focus information detection means that uses, as focus information, a modulation component included in a time-series photoelectric conversion output, that is, a difference in a modulation component caused by the contrast of an object.

A conventional focus information detection device of this type is generally constructed as shown in FIGS. 1i and 2.

FIG. 1 is an optical principle diagram for detecting focus information.

As shown in the figure, the reflected light from the subject 1 passes through the lens 2,
It passes through the half mirror 3 and is imaged on the imaging plane A of the imaging device (or film). On the other hand, the light partially reflected by the half mirror 3 is guided to the half mirror 5. Approximately 50% of the light is reflected by the No.-7 mirror 6 and becomes light K1. It is supplied to conversion element tI+6. The remaining 50 beams of light that have passed through the half mirror 6 are supplied to a photoelectric conversion element group 7.

The photoelectric conversion element groups 6 and 7 are installed at locations with different optical path lengths from the lens 2, and the optical path length from the lens 2 to an intermediate position between them is the optical path length from the lens 2 to the image sensor 4t. It is set to match t1.

Therefore, when the focal point of the lens 2 is on the imaging plane of the image sensor 4, the photoelectric conversion element group 6,
All of the light receiving surfaces 7 are out of focus. That is, the focus of the photoelectric conversion element group 6 is shifted toward the so-called front focus, and the focus of the photoelectric conversion element group 7 is shifted toward the rear focus, and the amount of shift is the same.

The photoelectric conversion element groups 6 and 7 are a plurality of photoelectric conversion elements, for example, 100 photoelectric conversion elements arranged in a row. Therefore, by sequentially switching the 100 charging conversion elements in one direction and reading out the photoelectric conversion output of each element, the time-series contrast is determined according to the contrast imaged on each photoelectric conversion element arranged in a row. I get a signal. The modulation component of this contrast signal will show a peak value at the time of focus.

FIG. 2 is a block diagram showing the configuration of an electrical circuit for obtaining focus information from the optical principle configuration shown in FIG. 1. As shown in the figure, the photoelectric outputs of the photoelectric conversion element groups 6 and 7 are supplied to HPFs 8 and 9, respectively, and only the modulated components are detected. The output of HPF8.9 is an envelope! Detector 1
0.11 and the envelope component is detected.

The output of the envelope detector 10.11 is sent to the difference detection circuit 12.
The difference between the two is detected. The output of the difference detection circuit 12 is sent out from a terminal 13 as focus information.

FIG. 3 is a diagram showing the relationship between the focus position and the level of modulation components 68.78 read out from the photoelectric conversion element groups 6 and 7. As shown in Figure 3, when the levels of the modulation components 68 and 78 of both photoelectric conversion element groups 6 and 7 are at the same level (point 6), the imaging plane of the image sensor 4 cannot focus. It turns out that there is. In FIG. 3, t2 indicates the optical path length between the photoelectric conversion element groups 6 and 7.

Thus, when the binding surface of the image sensor 4 is in focus, the real modulation components e8.
78 are equal, the output level of the focus information output from the terminal 13 becomes OV. Also, if the focus shifts, both
8 and 78 are distorted, the focus information is shifted in the direction of positive potential or in the direction of negative potential, and a voltage proportional to the amount of focus shift is generated.

By the way, when comparing the modulation components of the time-series photoelectric conversion outputs obtained from a plurality of photoelectric conversion element groups, for example, two photoelectric conversion element groups, when the illuminance of the subject is low, the level of the modulation components decreases and there is a relative noise. component increases. Therefore, the focus 1 that is obtained is selected to be one in which a large amount of noise is included.As a result, the accuracy of detecting focus information is reduced. Conventionally, the integration time of the photoelectric conversion element was controlled according to the illuminance of the subject, and when the illuminance was low, the integration time was lengthened to improve the S/'N ratio. In other words, by lengthening the time during which the electric charge once charged in a photoelectric conversion element such as a photodiode is discharged to photoelectrons,
The S/N ratio of the modulation component is improved, and the detection accuracy of focus information is improved. However, such means has the disadvantage that the configuration of the control system becomes complicated. Another method for improving the ratio is to control the integration time so that the average value of the photoelectric conversion output matches the vicinity of the operating level, which is lower by a certain level than the saturation level of the photoelectric conversion element. However, since this method only sets the average value of the photoelectric conversion output to a predetermined value, for example, if a pulsed photoelectric conversion output occurs due to a local highlight object in the photoelectric conversion output obtained in time series, In this case, the above-mentioned Norse-shaped modulation signal is completely saturated. Such saturation of the modulation component generates harmonic components, which become noise components, thereby reducing detection accuracy.

That is, in FIG. 2, when each modulation component 68.78 of the photoelectric conversion element 86.7 is saturated in the previous stage of the IPF B, 9, the HPF 8, 9 will not detect the harmonic component due to this saturation. This results in an error in focus information.

The present invention has been made in consideration of the above circumstances, and its purpose is to set the peak value of the modulation component of the time-series photoelectric conversion output read out from the photoelectric conversion element group immediately before the saturation level. By controlling the integration time so that the
An object of the present invention is to provide a focus information detection device that can obtain focus information.

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

FIG. 4 is a block diagram showing the overall configuration of an embodiment of the present invention. However, the same parts as in FIG. 2 are given the same reference numerals and detailed explanations will be omitted.

In FIG. 4, a voltage controlled oscillator 21 shown at the lower left in the figure has its center oscillation frequency controlled by being given a feedback signal, which will be described later, as a control voltage.

FIG. 5 is a diagram showing the characteristics of the voltage controlled oscillator 2. As is clear from FIG. 5, as the human control voltage increases, the center oscillation frequency also increases.

In FIG. 4, the oscillation output of the voltage controlled oscillator 21 is given to the ring counter 22 as a clock pulse. The ring counter 22 has a 100-stage configuration, for example, corresponding to the number of elements in the photoelectric conversion element groups 6 and 7, and converts each phase output pulse outputted from each stage into each photoelectric conversion element group 6 and 7. A readout scanning pulse is applied to the element to perform reading. In other words, each light-transformer element arranged in a row is successively advanced from the left end to the right end in the figure, and when it ends, the action of continuing from the left end, i.e., scanning, is repeated in sequence. .

FIG. 6 is a circuit diagram showing the internal structure of the photoelectric conversion element groups 6 and 7. In FIG. 6, 4 is a photodiode, and 42 is a switching FET-t'6. This FmT
42 is each phase/? from the ring counter 22 applied to ff-). Rusnyozion and p1 photodiode 41
This is for not only reading the amount of charge but also charging the charge.

Further, 43 is a common line for signal readout, and in this embodiment, 100 photodiodes 41 and switching FETs are connected to this common line.

Note that a DC voltage +E is applied to the common line 43 via a resistor 44. In this way, each phase output from the ring counter 22 is given to the switching FET 42 as a scanning pulse for reading, and when the switching FIT 42 is turned on, the voltage +E is applied to the resistor 4.
Apply reverse bias to diode 4 through 4. As a result, the capacitor of the PN junction photodiode 41 is charged to a saturated state. Switching FET after charging
4J is off and the diode 41 is left with a reverse bias. If the diode 41 is not irradiated with light in this state, no charging current will flow even if the switching FLAT 41 is turned on. On the other hand, when light is irradiated during this time, electrons and oars are excited by the light, and a discharge current flows within the diode 4. As a result, the charge accumulated in the diode 41 decreases in proportion to the amount of incident light.

This lost charge amount is compensated for when the switching FIT 55 is turned on, and the charging current becomes the photoelectric conversion output.

A in FIG. 7 is an example of the read photoelectric conversion output, that is, the output waveform of the photoelectric conversion element groups 6 and 7.

In this way, the photodiode 4 shown in FIG. 6 integrates the incident light over one scanning period T to produce a photoelectric conversion output. Therefore, as the time width of the scanning period T becomes longer, the integration cap of the photodiode 4 for the same incident intensity as #1 increases, and the apparent sensitivity can be increased. This is called a charge accumulation mode, and is an effective means for increasing sensitivity. In the present invention, as described above, the integration time is controlled by controlling the oscillation frequency of the oscillator 2. That is, when the oscillation frequency of the oscillator 21 is lowered, one scanning period T by the ring collar/receiver 22 becomes longer, and the integration time also becomes longer. Conversely, if the oscillation frequency of oscillator 2 is increased, the integration time will be shortened. From the above, if the oscillation frequency of the oscillator 21 is lowered when the subject illuminance is low, the SA ratio of the photoelectric conversion output for which the integration time is long will be improved. Therefore, the feedper system for controlling the oscillation frequency of the oscillator 21 will be explained below.

In Fig. 8144, each output from the photoelectric conversion element groups 6 and 7 is supplied to I(PF 8 and 9) as well as to the flang circuits 23 and 24. The flang circuits 23 and 24 perform photoelectric conversion of the darkest contrast part. It is designed to double-amplify the output to ground potential and is constructed as shown in FIG.

In other words, in FIG. 8, when the outputs of the photoelectric conversion element groups 6 and 7 are supplied to the input terminal 51, the capacitor 62 is overloaded, the current is amplified by the current amplifier 63, and the flag output is not output. Sent out. capacitor 5
The output side of 2 is grounded via a resistor 5'5.

Therefore, when there is no signal at the input terminal 5), that is, when there is no contrast component in the signal, the output side of the capacitor 52 is at ground potential, and the output potential of the current amplifier 63 is also at ground potential.

On the other hand, when a signal having a contrast component as shown in FIG.
6 becomes conductive. As a result, the potential of Q automatically becomes the ground potential.

In FIG. 4, the respective outputs of the 2-amp circuits 23 and 24 are supplied to a mixer 25, where they are added and mixed. The output of the mixer 26 has a waveform as shown in FIG. 7B. That is, at the time of focusing, the charging conversion output shown in A in the figure and #11 have the same waveform.

The output of this mixer 25 is supplied to a peak hold circuit 26 to detect the peak value. As a result, Flang circuit 2
At 3.24, the photoelectric conversion output of the darkest portion is grounded to the ground potential, and the peak hold circuit 26 conversely detects the photoelectric conversion output of the brightest portion. The peak voltage held by the peak hold circuit 26 is transferred to the sampling hold circuit 2.
1.

FIG. 9 shows an example of the peak hold circuit 26 and the sampling hold circuit 27. As shown in FIG. 9, the output of the mixer 25, which is input to the input terminal 61 of the peak hold circuit 26, is connected to the diode 62. resistor 63
The capacitor 64 is charged through the voltage, and the peak value is held. The held voltage waveform is as shown in No. 7-C. PL is the peak level. In a region where the photoelectric conversion output B exceeds the peak value point B1 and gradually decreases, the diode 62 is cut off, and therefore has no effect on the peak hold voltage. The peak voltage charged in the capacitor 64 is amplified by a current amplifier 65 and then sent to the sampling and hold circuit 27 side.

C of FET 66 connected in parallel to capacitor 64
A reset pulse as shown in FIG. 7 is applied through a resistor 67. As shown in FIG. 4, this reset pulse is frequently generated by the reset pulse generator 28 which is triggered by the pulse of the final stage of the ring counter 22. The seventh diagram shows the waveform.

When the FICT 66 is given the reset pulse, it turns on and discharges the peak voltage held in the capacitor 64, making it possible to detect the peak value in the next scanning period. In this way, the peak voltage can always be tracked and maintained even for a moving subject image. In addition,
In the second amplifier circuit 23 and 24, the photoelectric conversion element groups 6 and 7
Since the DC component of is removed, even if the above element groups 6 and 7
Even if the DC component of the sensor lifts (due to temperature or changes over time), the detected peak value will not change.

On the other hand, as shown in FIG. 9, the sampling hold circuit 27 charges a capacitor 72 with the peak voltage signal sent from the peak hold circuit 26 via the Templing FET 71, and the charged voltage signal is sent out from the terminal 73. The phase 9 pulse of the ring counter 22 one step before the final stage shown in FIG. 4 is applied to the dart of the sampling FET 71 as a sampling flat pulse through a resistor 74. FIG. 7E shows the waveform.

When the FIT 7 J is given the above-mentioned pulse, it turns on and charges the capacitor 72 with the voltage signal from the peak hold circuit 26 . The peak voltage held in the capacitor 64 of the peak hold circuit 26 is transferred to the capacitor 72 of the sampling hold circuit 27. As a result, the peak value voltage of the photoelectric conversion output during one scanning period is stored in the capacitor 12, and this becomes the output of the sampling hold circuit 27. Note that the above output is the peak value voltage of the weighted average of both peak voltages of the photoelectric conversion element groups 6 and 7.

As shown in FIG. 4, the output of the sampling hold circuit 27 is supplied to the positive polarity side of the differential amplifier 29. A voltage adjusted by zolium 30 is supplied to the negative polarity side of the amplifier 29.

The above voltage obtained at Zrium 30 is the voltage that determines the operating voltage of the negative feedback loop, and this is the voltage that determines the operating voltage of the negative feedback loop.
7 is set near just before it becomes saturated. Differential amplifier 29
The output of is supplied to the voltage amplifier 21 mentioned above via the resistor 3.

Therefore, the oscillation frequency of the voltage control amplifier 2 is automatically controlled so that the amount of charge accumulated in the photoelectric conversion element 6.7 becomes close to saturation regardless of the illuminance of the object. Note that the diode J2 shown in FIG. d# Zolium 33 is used to limit the control voltage input to the voltage controlled oscillator 21 so that the oscillation frequency does not become extremely low, and the limiting voltage is determined by Zolium 33. is adjusted.

As explained above, according to the present invention, the photoelectric conversion output corresponding to the dark part of the subject illuminance of a plurality of photoelectric conversion/outputs is
A scanning ring counter that clamps to the same potential, obtains a peak value voltage by weighted averaging of the photoelectric conversion outputs corresponding to the brightest part of the subject illuminance at that time, and obtains the photoelectric conversion output with this peak value voltage. A negative feedback loop is formed to control the clock frequency of the photoelectric conversion element, and the operating point of this loop is set so that the accumulated charge of the photoelectric conversion element is near saturation. As a result, it is possible to provide a focus information detection device that can obtain a photoelectric conversion output with a high S/N ratio even when the subject illuminance is low, and can obtain highly accurate focus information.

[Brief explanation of drawings]

Figures 1 to 3 are diagrams for explaining the conventional device;
9 to 9 are diagrams showing one embodiment of the present invention, FIG. 4 is a block diagram showing the configuration, FIG. 5 is a characteristic diagram of a voltage controlled oscillator, and FIG. 6 is a diagram showing the configuration of a photoelectric conversion element group. FIG. 7 is a circuit diagram showing the operation waveforms of each part, FIG. 8 is a circuit diagram showing the configuration of the flange circuit, and FIG. 9 is a circuit diagram showing the configuration of the peak hold circuit and the rising hold circuit. Applicant's representative Patent attorney Takehiko Suzue Figure 5 Figure 6 Figure 8

Claims (1)

[Claims] (1) The focal point is determined based on the difference in modulation components caused by the contrast of the subject included in the photoelectric conversion output obtained over time (by a group of photoelectric conversion elements installed at multiple locations with different optical path lengths from the lens). In the apparatus for detecting information, means for flagging the photoelectric conversion outputs corresponding to the darkest part of the subject illuminance of the time-series photoelectric conversion outputs obtained from the plurality of photoelectric conversion element groups to the same potential; , this flange means that Shifran! A weighted average of the peak values is performed for the photoelectric conversion outputs obtained, and the above-mentioned -
The photoelectric conversion element includes means for retaining information only during the scanning period, information retained by the means, and means for controlling the time width of the scanning period and the charge integration time due to light incident on the photoelectric conversion element group. A focus information detection device characterized in that the charge of a photoelectric conversion element is automatically set to near saturation regardless of subject illuminance.