JPH055621A - Distance measuring method - Google Patents

Abstract

PURPOSE:To enable highly accurate measurement of a distance by delaying a distance measuring sequence by more than a fixed time after a gain change of a gain control amplifier to operate the gain control amplifier in the optimum state. CONSTITUTION:Distance measuring light which is projected from 2 projection section 2 to be reflected on an object S1 is received with a photodetecting section 7. A distance-distance signal is calculated with a microcomputer 14 from an output of a gain control amplifier of an automatically focusing IC for amplifying the resulting output. At this point the level of an output signal of the gain control amplifier is monitored and adjusted when it is outside a specified range. In the changing of the gain, a time is set as required for the stabilization of the gain of the gain control amplifier and after the period set passes, the lighting of a projection section 2 is allowed so that an output of the gain control amplifier is sampled in stable state. Thus, higher distance accuracy is achieved without picking up an improper measuring data with unstable gain.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an active type distance measuring method used in a photographic camera, a still video camera or the like.

[0002]

2. Description of the Related Art In recent compact cameras, an active type distance measuring device for performing distance measurement by optical triangulation is often used. This active type distance measuring device arranges a light emitting unit and a light receiving unit while maintaining an appropriate base line length,
This projector emits near-infrared light toward the shooting scene,
The light reflected by the subject is received by the light receiving section. Since the incident position of the reflected light on the light receiving unit changes according to the subject distance, the subject distance can be measured by electrically checking the incident position of the reflected light.

The light projecting section comprises a light source and a light projecting lens for converting the distance measuring light from the light source into spot light, and the light receiving section comprises a light receiving element and a light receiving lens. An LED (light emitting diode) that emits near-infrared light is used as a light source, and a semiconductor position detector (PSD: Position Sensitive Detector) is used as a light receiving element. The PSD has two output terminals and outputs two channels of current depending on the intensity and position of incident light. Current of these 2 channels, or 2 corresponding to these
By determining the ratio of the two voltages, it is possible to obtain a signal that depends only on the incident position of the reflected light.

Since the intensity of the reflected light largely changes depending on the distance and the reflectance of the object, the PSD output signal may be too large or too small. In such a case, even if the ratio of the output signals of two channels is used, the measurement accuracy cannot be kept stable. Therefore, a pair of gain control amplifiers are used to keep the levels of the output signals of the two channels from the PSD within an appropriate range. Depending on the magnitude of the output signal of each PSD channel, if the gain control amplifiers are set to a common optimum gain and then distance measurement is performed, the PSD signal is increased or decreased to an appropriate level. It becomes possible to maintain the ranging accuracy in a stable manner regardless of changes in the reflectance and the reflectance.

In order to further improve the distance measurement accuracy,
A method is also known in which distance measuring light is projected onto a subject a plurality of times, the output signal of the gain control amplifier is taken into a microcomputer for each projection, and the subject distance is calculated from the average value of these. Furthermore, there is a multi-beam type that measures the subject distance at each distance measuring point by projecting the distance measuring light not only at the center of the shooting scene (shooting screen) but also at the peripheral area of the shooting area. is there. In this multi-beam type, since the subject distance is obtained for each distance measuring point, one subject distance is selected according to a predetermined priority order, and the taking lens is set according to this subject distance.

[0006]

The gain control amplifier requires about 1.5 to 2 msec for the output signal to stabilize when the gain is changed. Since the gain is in an unstable state before the stabilization time elapses, if the output signal of the gain control amplifier in this state is used, the distance measuring accuracy will be deteriorated. Further, when the near infrared light from the fluorescent lamp is incident on the PSD when the gain of the gain control amplifier is determined, the gain is determined in consideration of this noise light. In this case, since the gain is set to a value lower than the optimum value, good distance measurement cannot be performed.

Further, in the distance measuring device of the type that projects the distance measuring light a plurality of times, the output signal of the gain control amplifier may saturate (overflow) beyond a predetermined range during the light projection. It occurs when the illumination condition for the subject changes during the distance. In order to perform distance measurement with high accuracy, it is desirable to readjust the gain of the gain control amplifier and then perform distance measurement again, but the following problems occur in distance measurement under the illumination of a fluorescent lamp.

[0008] As is well known, fluorescent lamps are lighted intermittently at twice the frequency of the commercial frequency, the commercial frequency 60 H _Z
Then, the lighting interval F of the fluorescent lamp shown in FIG.
1 is 8.3 msec. When the projection distance F2 of the distance measuring light shown in FIG. 9B is set to 1 msec and the distance measurement is performed by projecting 18 times, two of them always match the peak of the light emission of the fluorescent lamp. The light is emitted at the timing, and the level of the output signal from the gain control amplifier exceeds the predetermined range. For this reason,
Under the illumination of a fluorescent lamp, the readjustment of gain and the re-ranging are repeated many times, resulting in a state in which the ranging cannot be performed.

The present invention has been made in view of the above background, and provides a distance measuring method capable of performing highly accurate distance measurement by operating a gain control amplifier in an appropriate state. The purpose is.

[0010]

In order to achieve the above object, the invention described in claim 1 is such that when the gain of the gain control amplifier is changed, the distance measuring sequence to be performed thereafter is changed. It is configured such that the distance measuring sequence is continued after delaying for a certain time or more and waiting for the gain control amplifier to stabilize. According to the invention described in claim 2, by delaying the operation of the light projecting means,
The distance measurement sequence is interrupted, which is effective in preventing unnecessary light projection.

According to the third aspect of the invention, the distance measuring light is set to P
When emitting light N times and performing one-cycle distance measurement, the gain of the gain control amplifier remains unchanged until the number of times that the output level from the gain control amplifier is out of the predetermined range reaches CM times. Distance measurement is performed. When the number of times the output level is out of the predetermined range reaches CM times, the gain of the gain control amplifier is changed, and then the distance measurement is repeated. Here, the value of CM is CM ≧ A when the projection interval of the distance measuring light is AT seconds.
It can be obtained from T × PN × 120. This formula is for commercial frequency 6
Since the CM value is determined in consideration of the lighting cycle of the fluorescent lamp of 0 Hz, it is necessary to eliminate the adverse effect that the distance measurement is repeated in response to the intermittent lighting of the fluorescent lamp and the distance measurement is repeated. You can

According to a fourth aspect of the present invention, when the output signal of the gain control amplifier exceeds a predetermined level, the output signal is fetched again with the same gain value after a lapse of 1 msec to 5 msec, and at this time, the gain signal is gained again. When the output signal of the control amplifier exceeds a predetermined level, the calculation of the distance data based on the output signal of the gain control amplifier is stopped.

[0013]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS In FIG. 1 showing the structure of a multi-beam type distance measuring device, a light projecting section 2 is composed of a light source section 3 and a light projecting lens 4, and an optical axis 4a of the light projecting lens 4 is of a photographing lens 5. It is substantially parallel to the optical axis 5a. The light source unit 3 includes three LEDs (light emitting diodes) 3 each emitting near infrared light.
It consists of a, 3b and 3c. These LEDs 3a to 3c are arranged horizontally, the central LED 3a is located on the optical axis 4a, and the LEDs 3b and 3c are located on the left and right thereof, respectively, and spot-shaped distance measurement is performed toward three points in the shooting scene. Project light.

The light receiving section 7 comprises a light receiving lens 8 and a rectangular PSD 9, and the optical axis 8a of the light receiving lens 8 is substantially parallel to the optical axis 5a of the taking lens 5. PS
D9 has a filter 10 (see FIG. 2) that transmits only near-infrared light on its light-receiving surface, and outputs a signal according to the amount of incident light that has passed through this and the incident position in the baseline direction. Output from 9a and 9b. The reflected light from the subject enters a position closer to the output terminal 9b side as the subject distance is shorter.

Further, the shorter the subject distance, the higher the intensity of the reflected light and the larger the absolute value of the signals from the output terminals 9a and 9b. However, by calculating the ratio of the sum and difference of these signals, the reflection A signal corresponding to only the incident position can be obtained without depending on the intensity of light. Note that the PSD 9 does not have a discriminating action in the horizontal direction, and if the incident height in the vertical direction is the same, the ratio of the output signals from the output terminals 9a and 9b is equal even if the incident position in the horizontal direction is different. It will be

The LEDs 3a to 3c have an autofocus I
Light emission is controlled via the LED driver 11 by a signal from C12. The autofocus IC 12 executes a predetermined distance measuring sequence according to a command from the microcomputer 14. Further, the autofocus IC 12 amplifies the signals from the output terminals 9a and 9b of the PSD 9, respectively, performs sample hold processing, and sends them to the microcomputer 14.

A general-purpose type is used as the microcomputer 14, and the calculation is performed based on the signals of the first and second channels output from the output terminals 9a and 9b of the PSD 9, respectively, and the distances are calculated for each of the LEDs 3a to 3c. Calculate the data. The microcomputer 14 selects the optimum one from the three pieces of distance data, and sends the lens position control circuit 15 the number of drive pulses corresponding to the selected distance data. The lens position control circuit 15 drives the stepping motor 15a to set the taking lens 5 at a position according to the distance data.

As shown in FIG. 2, the microcomputer 14 comprises a CPU 20, a ROM 21, a RAM 22, a serial I / O port 23 and an A / D converter 24.
As is well known, the CPU 20 incorporates various registers and a logical operation circuit, and executes a shooting sequence according to a sequence program written in the ROM 21.
The RAM 22 temporarily stores various data, flags and the like obtained during execution of the shooting sequence.

The autofocus IC 12 is a one-chip IC and is a preamplifier for converting the currents of the first and second channels output from the logic circuit 26, the gain controller 27, and the output terminals 9a and 9b of the PSD 9 into voltage. 28a, 28b and gain control amplifier 2
9a and 29b and sample and hold circuits 30a and 30b
And buffer amplifiers 31a and 31b.

Gain control amplifiers 29a, 29b
Is provided in consideration of the fact that the light amount of the reflected light incident on the PSD 9 decreases and the absolute value of the current from the output terminals 9a and 9b decreases when the subject is at a long distance. Optimal gain is given by the gain control process. Sample hold circuits 30a, 30
b receives the sampling pulse from the logic circuit 26 and samples and holds the signals amplified by the gain control amplifiers 29a and 29b, and these signals are buffered by the A / D converter 24 of the microcomputer 14 via the buffer amplifiers 31a and 31b. Send to. The logic circuit 26 is basically composed of a serial-in / parallel-out shift register, and the gain controller 27 reads the gain control data set at a predetermined bit position of the shift register, and based on this, the gain control amplifier 29a, The optimum gain common to 29b is set.

The serial I / O port 23 converts the parallel command data from the CPU 20 into a serial data pulse (AFSD) as described above, and the logic circuit 2
6, a transfer pulse (AFSCK) for transferring this serial data to the shift register, latching parallel data from the shift register, and the LED 3
a control pulse (A
FLCK) is output. Also, autofocus IC1
The analog signal output from 2 is the A / D converter 2
4 converts it into a 7-bit digital signal (representing 0 to 127 in decimal) corresponding to the signal voltage level.

FIG. 3 conceptually shows an 8-bit shift register 33 used in the logic circuit 26. Gain control data (GAIN) is in 5 bits of "D0 to D4", LED light emission data (LED) is in 2 bits of "D5 to D6", and light emission / reset of LED is in the bit of "D7". Switching data (SE
T) is assigned. The 5-bit gain control data can represent gain levels from "0" level to "31" level. Also, 2 bits of L
The ED emission data are decimal numbers “0”, “1”, “2”,
It represents four states of "3", and when it is "0", the LEDs 3a to
All of 3c are turned off, when it is "1", only the LED 3a emits light, and when it is "2", "3", the LED 3 respectively
b, the LED 3c is caused to emit light.

FIG. 4 showing a procedure of a basic distance measuring sequence.
In the process of detecting the offset value of the amplifier system of the autofocus IC 12, the LEDs 3a to 3c are sequentially emitted, and the gain is set to the optimum value according to the output signals of the gain control amplifiers 29a and 29b obtained at that time. A gain determining process for determining, a distance measuring process for taking measurement data under this gain and calculating distance data for each LED 3a to 3c, and a process for selecting an optimum one from the three distance data are sequentially performed. To be executed.

Next, the operation of the above embodiment will be described with reference to FIGS. As shown in FIG. 5, when the power switch of the camera is turned on, the data storage area in the RAM 22 is reset to the initial state. Next, various kinds of external data representing the loaded state of the film and the opened state of the lens barrier are read into the RAM 22. CPU
When the camera 20 determines that the camera is in the shooting standby state based on these external data, the shutter 20 unlocks the shutter button.

When the shutter button is pressed halfway, the CPU2
0 outputs a distance measurement command to the autofocus IC 12. The autofocus IC 12 drives the light projecting unit 2 to periodically generate spot-like near-infrared light, and projects the near-infrared light toward a subject. The object distance is measured by detecting the near-infrared light reflected by the object with the light receiving unit 7. After this distance measurement, the CPU 20 outputs a photometric command to start measurement of subject brightness. The exposure calculation is performed based on the obtained subject brightness and the film sensitivity signal, and the exposure time for obtaining the proper exposure is calculated.
It stores in 22. After the exposure calculation, the CPU 20 has already RA
The number of clock pulses corresponding to the distance data stored in M22 is sent to the lens position control circuit 15. As a result, the stepping motor 15a is driven, and the taking lens 5 is extended to the lens position according to the distance data.

When the shutter button is fully pressed, the RAM
A program shutter (not shown) is operated in accordance with the exposure time data read from 22. When the shooting of one frame is completed, the CPU 20 generates a film transfer command and operates a film transfer mechanism (not shown) to feed the film one frame. When the film transfer for one frame is completed, the CPU 20 sends a reset signal to the lens position control circuit 15, returns the taking lens 5 to the initial position, and one taking sequence is completed.

Next, the distance measuring sequence will be described in detail. When the distance measuring sequence starts, first, as shown in FIG. 6, offset value detection processing is executed. In this offset value detection processing, “D0 of the shift register 33 is
"01000000" is set to each bit of "~ D7". As a result, "GAIN = 8" and "LED =
Each command data of "0" and "SET = 0" is set. This command data is set at the time point P1 shown in the timing chart of FIG. 7, that is, the transfer pulse AFSC.
After K has sent out 8 pulses, the control pulse AFLCK
Is performed at the timing when the level becomes low. The serial data pulse is sequentially transferred to the shift register 33 at the rising edge of the transfer pulse AFSCK. Furthermore, after transferring the above command data, the CPU
“COUNT = 8” is set in a predetermined register in 20.

Of the command data set in the shift register 33, the data of "GAIN = 8" is sent to the gain control amplifier 2 via the gain controller 27.
9a and 29b, and these gains are set to the initial value (minimum value) "8". In the distance measuring device of this embodiment, "GAIN" is set so that the linearity of the amplification characteristic is maintained.
Is used within the range of "8" to "31", but this range is appropriately determined according to the specifications of the camera, the light emission amount of the LED, and the like.

After the transfer of the serial data, first, second
Digital signals AFD1 and AFD2 obtained by sample-holding the output signal from the channel and A / D converting the sampled signal are read into the microcomputer 14. In the procedure of this reading process, as shown in the timing chart of FIG. 7 and the flowchart of FIG. 8, the control pulse AFLCK is first set to the low level during the period ΔT1. This control pulse A
At time P2 when the period ΔT7 has elapsed since FLCK goes low, any one of the LEDs determined by the command data emits light. Then, after the period ΔT2, the sample hold circuits 30a and 30b sample and hold the outputs of the gain control amplifiers 29a and 29b at the timing when the AFSD becomes high level. Further, the light emission of the LED is stopped at the time point P3 when it becomes the low level after the period ΔT3. Further, when AFSD becomes high level after the period ΔT4, the sample hold processing ends.

In FIG. 7, the control pulse AFLCK is at the low level during the period ΔT1, which is LE.
When D is turned on, AFLCK is used in the offset value detection process performed by the command "LED = 0".
Is kept at a high level. Therefore, in the offset value detection process, the LED is turned off, and the gain control amplifiers 29a and 29b are set to "GAIN = 8". In this state, the output signals of the gain control amplifiers 29a and 29b are the sample and hold circuit 3
The samples are held at 0a and 30b.

When time ΔT6 elapses from the time P3 when AFSD becomes low level, the sample hold circuit 30
The output signals of a and 30b are buffer amplifiers 31a and 31b.
It is converted into a 7-bit digital signal by the A / D converter 24 via b, and these are sequentially fetched into the CPU 20 as the measurement data AFD1 and AFD2 of the first and second channels, and stored at a predetermined address in the RAM 22. ..

On the other hand, in the gain determining process and the distance measuring process,
The designated LED is lit once during the period ΔT1 in which AFLCK is maintained at the low level, and the output signals of the gain control amplifiers 29a and 29b are sample-held by the sample-hold circuits 30a and 30b during this lit. It Then, regardless of whether the LED is turned on or off, at the time when the high level period ΔT5 of AFLCK elapses, the acquisition cycle of one measurement data ends.

In the offset value detection process, the measurement data is fetched eight times, and the measurement data AFD1 and AFD2 are acquired.
You can get 8 each. These eight measurement data are averaged for each channel, and their average value is determined as the offset values OF1 and OF2. As described above, since the LED is turned off in the offset value detection processing, the offset values OF1 and OF2 are caused by noise light from the shooting scene incident on the PSD 9, noise of the signal processing system, and the like.

After the offset value detection processing, gain determination processing and distance measurement processing are performed. These processing is LED3
It is executed individually for a to 3c. The gain determination process is performed according to the flowchart of FIG. When the gain determination process is started, “GFLAG”, “OVFLAG”, and “GCOUNT” set in the RAM 22.
Is set to the initial value "0". Then, serial data representing a new command is transferred from the serial port 23 to the logic circuit 26, and at a time point P1 when AFLCK becomes low level, binary data “01000011” is set in each bit of the shift register 33. It As a result, “GAIN = 8”, “LED = 1”, “SE
The command “T = 1” is set, and only the LED 3a is in the light-emission enabled state. After this command is set, the period “ΔT2
The LED 3a is turned on once during "+ ΔT3", and the sampling process is performed during this turning on. Then, the reading process of the measurement data AFD1 and AFD2 is performed. Note that L
The lighting cycle of the EDs 3a to 3c is, for example, 1 ms.

Here, when the initial value of the gain is set or the gain is changed, the gain control amplifier 29
After stable operation of a and 29b, measurement data AFD
1, in order to perform the reading process of AFD2, the start of the distance measuring sequence is delayed by a predetermined time. That is, at the time point P1, command data is set in the shift register 33 and the gains of the gain control amplifiers 29a and 29b are set.
It takes about 1.5 to 2 mse until the gain of 29b stabilizes.
c takes time. Therefore, the period ΔT0 is set to 2 mse
c is set, and the LED 3a is allowed to light after this period has elapsed, and the gain control amplifier 29 is in a stable state.
The outputs of a and 29b are sampled. When the measurement data AFD1 and AFD2 are loaded with the gain kept constant, ΔT0 is “0” as shown in FIG.
Is.

As shown in FIG. 1, when the distance measuring light from the LED 3a is applied to the main subject S1, the reflected light passes through the light receiving lens 8 and is incident on the PSD 9. The signals output from both terminals 9a and 9b of the PSD 9 at this time include information on the intensity of light incident on the PSD 9 and the incident position. These signals are digitally converted by the A / D converter 24 and then measured data AFD1, AF
Captured as D2.

Next, the CPU 20 causes the measurement data AFD
It is determined whether any one of AFD1 and AFD2 has reached "127" in decimal. The values of AFD1 and AFD2 are "1.
When it reaches 27 ", it is determined that one of them is in a saturated state (overflow). Since this measurement data is extremely unreliable, after confirming that "GAIN = 8", "AFRESULTa ← 127"
(“AFRESULTa” will be described later), and the distance measurement by the LED 3a is completed. In this state, the subject in the center of the screen is very close,
Even when "GAIN = 8", the reflected light is too strong.

When the measurement data AFD1 and AFD2 are not saturated, the addition measurement data AFAD is calculated from the respective measurement data AFD1 and AFD2 by the procedure shown in FIG.
Calculation processing of the D and subtraction measurement data AFDIF is performed.
In this process, the measurement data AF is used to remove the noise component.
The offset values OF1 and OF2 are subtracted from D1 and AFD2. Further, when the result of the subtraction processing of the offset value becomes a negative value, or when the value of the subtraction measurement data AFDIF becomes a negative value, these values are set to "0" to simplify the subsequent calculation. ..

Next, it is determined whether or not the value of the addition measurement data AFADD is "136" or more. When this value is equal to or greater than "136", it is determined that the absolute value of each of the measurement data AFD1 and AFD2 is within a range suitable for executing the subsequent distance measurement calculation, and the gains of the gain control amplifiers 29a and 29b are It is determined as the value of "GAIN" at that time. And this "G
As it is set to "AIN", as shown in FIG.
Distance measurement processing by the ED 3a is started.

On the other hand, when the value of the addition measurement data AFADD is less than "136", "GAI" at that time is calculated.
After confirming that "GFLAG" indicating that the value of "N" was inappropriate is not "1", the gain is changed to increase the absolute value of the measurement data AFD1 and AFD2. Be seen. The changed gain value is the original “GAIN” value plus the correction value “N” and is written in the RAM 22. The value of the correction value “N” is set as shown in Table 1 below according to the size of the addition measurement data AFADD.

[0041]

[Table 1]

The value of "GAIN" is changed by setting command data including a new gain in the shift register 33. At this time, only the data of bits "D0" to "D4" of the shift register 33 is changed. Changes, and the data assigned to other bits remains unchanged. Then, the gain controller 26 uses the new “GAI
Gain control amplifier 29 according to the value of "N"
The gains of a and 29b are adjusted, and the same processing is repeated. Also in this repetitive process, after the command setting for changing the gain is delayed by ΔT0, the operation of the gain control amplifiers 29a and 29b is stabilized, and then the acquisition of the measurement data is continued again. During this iterative process, "GAIN>8" may overflow, but in this case, "GFLAG" is set to "1".
After being set to, the value of "GAIN" is set lower by "1".

The value of "GAIN" is "3" which is the maximum gain.
Addition measurement data A that has an appropriate absolute value even if it reaches 1 ”
When the FADD is not obtained, the reflected light from the subject is extremely weak or is not returned to the PSD 9. In this case, since the object distance corresponds to an extremely long distance, there is no practical benefit even if the distance measurement described below is performed. Therefore, in this case, the processing of "AFRESULTa ← 0" is performed without performing the distance measurement processing.

The number of times of resetting of the gain RN in this gain determination processing is obtained from the initial gain value IG, the maximum gain value GM, and the maximum correction value NM. RN = (GM-IG) ÷ NM In this embodiment, GM = 31 and IG = NM = 8, so RN = 2.9≈3. Therefore, the number of times of resetting the gain RN is limited to 3 times, and “GCOUNT =
If the value reaches “3”, “GAIN” at that time is selected, and the gain determination process ends. Due to this limitation, the intensity of the reflected light from the subject changes, and the measured data A
Even if the values of FD1 and AFD2 change, the gain determination process does not take time, and the gain can be determined quickly. Moreover, the gain adjustment range "31-
8 "is divided by the maximum correction value" 8 "to determine the upper limit of the number of gain adjustments. Therefore, even when the value of AFADD is extremely small, the gain value should be adjusted to the maximum value" 31 ". Is possible.

When the above gain determination processing is completed, LE
Distance measurement processing by D3a is started. 11 and 12
Represents the procedure of this distance measurement processing. In this distance measurement processing, after setting various initial data in the RAM 22, the LED
3a is turned on 18 times, and AFD1 and AFD2 each time
Read. Then, when reading AFD1 and AFD2 each time, it is determined whether either of them is saturated, or whether the value is less than “15” and the absolute value of the measured data is insufficient, and these are applicable. Occasionally, "OVCOUNT" indicating improper distance measurement is incremented by "1". In this case, the distance measuring light is projected again and the measurement data is read again.

Here, the upper limit of "OVCOUNT" will be described. If the upper limit value of "OVCOUNT" is too small, "OVFLAG" is likely to be set even if the illumination condition of the subject changes accidentally and momentarily. There is a need for data, and it tends to lack accuracy. Most of the cases where the illumination conditions change are fluorescent lamps, which are driven by a commercial power source, so the upper limit value CM of “OVCOUNT” is determined from the following equation in consideration of 120 Hz, which is twice that. Where AT is the projection distance of the distance measuring light,
PN is the number of times of light projection. CM ≧ AT × PN × 120

In this embodiment, the projection distance AT of the distance measuring light is 1 msec and the number of projections PN is 18. Therefore, when these are substituted into the above equation, the following is obtained. CM ≧ 1 × 10 ⁻³ × 18 × 120 = 2.16 Since “OVCOUNT” is an integer, the upper limit CM is rounded up to “3”. When the upper limit value CM is set from this equation, as shown in FIG. 23, even if the projection timing of the distance measuring light coincides with the emission peak of the fluorescent lamp up to twice, the distance measuring is performed without being affected by this. Can continue. In this case, to obtain one distance data, 2
The light will be emitted 0 times. Of course, this "OVC
The two pieces of measurement data that have become "OUNT" are not used in the distance measurement calculation.

When "OVCOUNT" is counted up and reaches "3" as a result, it is determined that the gain value determined by the gain determination process is inappropriate, and "OVCOUNT" is determined.
Set "VFLAG" to "1" and then "GAIN"
Is set low by "1" and the ranging process is restarted from the beginning. The gain change during the distance measurement is performed by resetting the command as described above. In this case as well, the delay process of time ΔT0 is performed before the LED 3a is turned on. When the time for setting a plurality of initial data in the RAM 22 is ΔT0 or more, the delay by the ΔT0 timer can be omitted.

In order to prevent the distance measuring process from being repeated many times, when "OVFLAG" is once set to "1" and then "OVCOUNT" reaches "3" again, that is, " When the total number of times of "OVCOUNT" is "6" or more, the value of "AFRESULTa" is determined as "M" corresponding to the value of "GAIN" at that time. At the time of this determination, the table shown in Table 2 is referred to.

[0050]

[Table 2]

As described above, when the value of "OVCOUNT" is less than "3", the distance measurement is continued with the same gain. However, in the later processing, the maximum value among the measurement data AFD1 and AFD2,
Since the minimum value is removed and a plurality of measurement data are averaged, even if there are two or more abnormal data in the measurement data, the influence thereof hardly appears in the distance measurement calculation.

In the process of fetching the measurement data AFD1 and AFD2, the minimum value and the maximum value among them are discriminated, and the AFD1 and ADF2 are respectively "SUM1" and "SUM".
2 ”will be added up. Then, in order to increase the reliability of the average value, the maximum value and the minimum value of each of AFD1 and AFD2 are subtracted from the total values "SUM1" and "SUM2" obtained by projecting 18 times, and then the average value is calculated. Calculation is performed. Although the distance measurement accuracy is reduced, the average value may be calculated without removing the maximum value and the minimum value. In this case, the divisor is "18".
Becomes

The average value thus obtained for each channel is corrected for the offset value according to the processing of FIG. 10 and then the addition measurement data AFADD and the subtraction measurement data AFD are added.
Converted to IF. Then, the value of these ratios is set to "12
7 ”is multiplied to calculate the distance data“ AFRESULTa ”, and the distance measurement processing by the LED 3a ends.

Subsequently, the gain determining process and the distance measuring process are performed while the LED 3b is turned on. Finally, LED3c
Gain determination process-distance measurement process. This allows
For the LEDs 3b and 3c, "AFRESULT
The values of "b" and "AFRESULTc" are obtained respectively. Note that division is performed in calculating the average value and distance data, but the results are rounded down, rounded up,
It is converted into an integer by rounding off.

As shown in FIG. 13, the LED 3a is used for the focus detection point SP1 set at the center of the shooting scene (shooting screen) 40.
The distance measuring light is projected onto the LED 3b, 3c, and the LEDs 3b and 3c
Distance measuring light is projected to the distance measuring points SP2 and SP3 set in the peripheral portion of. In this shooting scene 40, the focus detection point SP1 is located on the main subject S1, and the focus detection point SP2 is the main subject S2.
Located on top. Therefore, “AFRESULTa” indicates a value corresponding to the distance to the main subject S1, “AFRESULTb” indicates a value corresponding to the distance to the background subject S2, and “AFRESULTc” indicates “0” (because the distance measuring light is not reflected). (See FIG. 9).

The microcomputer 14 has a ROM 21.
According to the program stored in "AFRESULT
Confirm that "a" is a value corresponding to the subject distance from the closest distance to 5 m, and if it corresponds to this, "AFRE
"SULTa" is determined as the optimum distance data, and if not applicable, "AFRESULTa", "AFRES"
The one corresponding to the shortest object distance among “ULTb” and “AFRESULTc” is determined as the optimum distance data, and the distance measuring sequence is completed.

After the optimum distance data is determined in this way, the stepping motor 15a is driven through the lens position control circuit 15, and the taking lens 5 is set at the lens position corresponding to the optimum distance data. In addition, "AFRE
SULTa ”,“ AFRESULTb ”,“ AFRES
Various other algorithms can be selected as the algorithm for determining the optimum distance data from the "ULTc".

In the above embodiment, during the gain determination process,
When the outputs of the gain control amplifiers 29a and 29b are almost saturated, the gain is decreased by "1" and the measurement is performed again. In indoor photography, the subject is often illuminated by a fluorescent lamp, but in such a scene,
Even if the gain is appropriate, due to the noise light of the fluorescent lamp,
The outputs of the gain control amplifiers 29a and 29b may be saturated temporarily. Therefore, when the outputs of the gain control amplifiers 29a and 29b are almost saturated during the gain determination process, the measurement is performed again with the same value of "GAIN", and at the time of the re-execution, for example, for a certain period of time. Measurement data A for only 1 to 5 msec
It is preferable to delay the uptake of FD1 and AFD2. 14 to 17 show this embodiment. In addition,
In this embodiment, 8 bits (decimal numbers "0" to "255") are assigned to the measurement data.

As shown in FIG. 14, when the gain determination process is started, the "GSNG flag" is reset and "GCOUNT" is set to the initial value "0".
Further, similarly to the embodiment shown in FIG. 9, the gains of the gain control amplifiers 29a and 29b are set to the initial value "8", and the commands "LED = 1" and "SET = 1" are set. , Only the LED 3a can be turned on.

After setting the above command, as described above,
After the period ΔT0, the LED 3a is measured for a period of “ΔT2 + ΔT3” while being lit, and measurement data AFD1 and AFD2 are read. It is determined whether or not any of these has reached a saturation value, for example, "F0" in hexadecimal ("240" in decimal). If either one is the saturation value "F0"
In the above case, when it is confirmed that the gain value at this time is the initial value “8”, “AFRESULT
a ← 127 ”is performed.

When the gain value is "9" or more,
When the measured data AFD1 and AFD2 reach the saturation value, confirm that the saturation state has never been reached before
After setting the SNG flag, the measurement data AFD1, again with the same gain value after a time delay of 2 msec.
The reading of AFD2 is executed. If the measured data also reaches the saturation value, "1" is subtracted from the gain value at that time to determine the gain value, the gain determination process is stopped, and the distance measurement process is directly performed. To do. As a result, it is possible to prevent the gain determination process from taking a long time due to the repetition of the saturated state of the gain control amplifiers 29a and 29b. It should be noted that when the gain value is "9" or more, one of the measurement data ADF1 and AFD2 becomes a saturation value not in the beginning of the gain determination process but in the repetitive routine.

Measured data AFD again with the same gain
1, when importing AFD2, 2mse as described above
A time delay of c is performed. The reason is that the fluorescent lamp noise superimposed on the gain control amplifiers 29a and 29b has a waveform as shown in FIG.
This is because if the reading timing of 1, AFD2 and the lighting of the fluorescent lamp match at time t1, the noise component due to the fluorescent lamp remains for a period indicated by Δx (about 1.5 to 2 msec). Therefore, at least 1 mse
Due to the delay of c, the next measurement data AFD1, which is shifted from the timing t2 when the noise component becomes maximum on the minus side
By taking in AFD2, it is possible to obtain measurement data AFD1 and AFD2 that are less affected by noise.
This 2 msec may be secured by the ΔT0 timer.
In addition, in order to completely remove the influence of the noise component, 5m
A delay time of about sec is desirable, so 1 to 5 is practically used.
It is set appropriately within the range of msec.

When the measurement data AFD1 and AFD2 have not reached the saturation value with the gain being the initial value "8",
Based on these measurement data, the calculation process of the addition measurement data AFADD and the subtraction measurement data AFDIF is performed as described above. With respect to this addition measurement data AFADD, it is determined whether the value is a decimal number “136” or more. When this value is equal to or greater than "136", it is determined that the values of the addition measurement data AFADD1 and AFADD2 are at appropriate levels for executing the distance measurement calculation thereafter. Then, the gains of the gain control amplifiers 29a and 29b are determined as the value of "GAIN" at that time (initial value "8"), and the distance measuring process by the LED 3a is subsequently executed.

On the other hand, when the value of the addition measurement data AFADD is less than "136", as described above, "G"
"N" is added to the value of "AIN", "GCOUNT" representing the frequency of the gain determination processing is increased by "1", and the same processing is repeated. "GAIN>"
The saturation value may be reached at "8", but when the "GSNG flag" is in the set state, "GAIN ← GAIN-1"
After the gain is determined by, the distance measuring process is executed.

According to the above, when the measurement data AFD1 and AFD2 reach saturation values when the gain control amplifiers 29a and 29b have the initial gain of "8", the distance data is immediately determined to a value corresponding to the shortest distance, Even when the saturation value is not reached, when the addition measurement data AFADD is at an appropriate level, the gains of the gain control amplifiers 29a and 29b are set to the initial gain "8" and the distance measuring process is executed. Thus, while the value of “GCOUNT” is up to “1”, that is, the gain control 29
Since the influence of noise is small when the gains of a and 29b are small, it is advantageous to determine the gain early and perform the distance measurement processing.

On the other hand, when “GCOUNT ≧ 2” is set during the repetition routine of the gain determination and the gain value is set higher each time, “GCOUNT = 6” as shown in FIG. Data AF until reaching
D1, AFD2 are taken in, these are "SUM1",
"SUM2" is added up sequentially. And, "GCO
When “UNT = 6” is reached, the accumulated “SUM1”,
Measurement data AFD of each channel from "SUM2"
1, the average value is calculated after subtracting the maximum value of AFD2, and replaced with "AFD1" and "AFD2", respectively.

The measurement data AFD1, A thus obtained
For the FD2, as shown in FIG. 10, after the offset correction processing, the addition measurement data AFADD and the subtraction measurement data AFDIF are calculated. Then, “AFADD ≧ 1
When “36” is satisfied, the gain when “GCOUNT = 2” is determined as the optimum gain, and the distance measurement processing is directly performed. If "AFADD ≧ 136" is not satisfied and the gains of the gain control amplifiers 29a and 29b exceed the maximum value "31", the subject is located at such a long distance that almost no reflected light is incident on the PSD 9. Therefore, the distance data “AFRESULT = 0” for setting the taking lens 5 at infinity is obtained,
At this point, the distance measurement by the LED 3a ends. On the other hand, when the gain does not exceed the maximum value “31”, the “GSNG flag” is reset to “GCOUNT = 0” in order to perform the measurement for gain setting again.

In this way, the gain control amplifier 2
When the gains of 9a and 29b are set to be high, a plurality of measurement data AFD1 and AFD2 are taken in and the gain is determined in consideration of the average value of these, so that the gain can be set with little influence of noise. it can. In addition, the integrated value SU of the measurement data AFD1, AFD2
When calculating the average value from M1 and SUM2, AFD
1, the maximum value of AFD2 is subtracted. Therefore, the measurement data AFD1 and AFD2 are captured under the illumination of a fluorescent lamp that intermittently emits light at a frequency twice the commercial frequency, and the periodic noise components as shown in FIG. 17 cause the measurement data AFD1 and AFD2. Measurement data AFD captured at time t = t1
For 1 and AFD2, this can be eliminated when calculating the average value, which is very effective in setting an appropriate gain. Of course, the measurement data AFD1 and AFD2 taken at the time of undershoot such as t = t2 are also taken into consideration, and the minimum value of the measurement data is also subtracted from the integrated values SUM1 and SUM2 before calculating the average value. Good.

When the value of "GAIN" is determined by the above processing, the distance measuring processing by the LED 3a is executed as in the above embodiment. In this distance measuring process, as shown in FIG. 16, when “OVCOUNT” is less than “3” and the measured data is out of the proper range, the gain value is maintained as it is for confirmation. After a delay time of 2 msec, the measurement data AFD1 and AFD2 are read again. As a result, the measurement data A in the ranging sequence
Even if the reading timing of FD1 and AFD2 matches the lighting timing of the fluorescent lamp and the measurement data reaches the saturation value,
The next reading will give the correct measurement data.
In this way, 18 pieces of measurement data AFD1, AFD2
Are captured, and AFADD and AFDI are processed by the procedure shown in FIG.
F is calculated, and the distance data "AFRESULT" is calculated by the procedure shown in FIG.

In each of the above embodiments, the three LEDs 3a to 3c are used.
Is a multi-beam type in which distances are sequentially measured at three points in a shooting scene, but the present invention can also be applied to a single-beam type having one distance measuring point. For example, one LE
D may be used to project a spot or an elongated slit light onto the shooting scene. Further, the distance may be measured by emitting light from this one LED only once.

Further, the posture of the light projecting portion may be changed according to the posture of the camera, and the light may be projected onto a plurality of distance measuring points along the direction always parallel to the ground. CCD as the light receiving element
An image sensor or the like can be used, and a flash discharge tube, a lamp or the like can be used as a light source.
Furthermore, various controls can be performed by a logic circuit other than the microcomputer.

Furthermore, it is possible to omit the offset value detection processing or partially change the distance measurement processing procedure. If the power consumption due to the blinking of the LED is negligible, the LED is turned on for a time ΔT0 after the gain is set.
Instead of just delaying, unnecessary measurement data within the delay period may be discarded.

In the above embodiment, the maximum value is subtracted and the average value of the measured data is calculated only when the gain is set to a high value. However, at the initial stage of gain determination, that is, at the initial gain. Even when reading measurement data, multiple measurement data are obtained, and the maximum and minimum values among these are
Alternatively, one of them may be removed and then the average value of the measurement data may be calculated.

[0074]

As described above, according to the first aspect of the invention, when the gain of the gain control amplifier is changed, the measured data is fetched after delaying the gain for a stable period. Therefore, incorrect measurement data output under an unstable gain is not taken in, and therefore the ranging accuracy can be improved. Further, if the measurement is delayed by delaying the operation of the light projecting unit, there is an advantage that the light projecting unit can be prevented from being uselessly operated and the waste of the power source can be prevented.

According to the third aspect of the invention, the number of measurement data having an improper value is constant when the distance measurement operation is performed using a plurality of measurement data obtained by a plurality of light projections. When the distance is less than, the distance measurement is continued as it is, and when the fixed number of times is reached, the gain of the gain control amplifier is changed to perform the distance measurement again. Since the setting is made in consideration of the number of times the light is projected and the lighting cycle of the fluorescent lamp, it becomes possible to change the gain in response to changes in the illumination conditions for the subject.
Moreover, distance measurement does not become impossible even under the illumination of a fluorescent lamp.

Further, in the invention described in claim 4, when the output signal from the gain control amplifier is in a saturated state at the time of determining the gain of the gain control amplifier or changing the gain during the distance measurement sequence, After a time delay of 1 to 5 msec in which the influence of the noise component superimposed on the control amplifier is reduced, measurement is performed again under the same gain value to determine whether the saturation state is caused by instantaneous noise. Determine. Even in this measurement,
When saturation occurs, it is determined that the output signal is constantly at a high level, not due to the influence of noise, and normal distance measurement is performed based on the output signal from the gain control amplifier. Apart from the ranging sequence,
At that time, it is immediately set to the distance data at infinity or infinity, so there is no need to lengthen the distance measurement time for determining or changing the gain, which is very effective for efficient distance measurement processing. Is.

[Brief description of drawings]

FIG. 1 is an explanatory diagram showing an outline of a distance measuring device for carrying out the method of the present invention.

FIG. 2 is a block diagram showing a configuration of an autofocus IC.

FIG. 3 is a conceptual diagram of a shift register used in a logic circuit.

FIG. 4 is a flowchart showing an outline of a distance measuring sequence.

FIG. 5 is a flowchart showing a shooting sequence.

FIG. 6 is a flowchart showing offset value detection processing.

FIG. 7 is a time chart illustrating the timing of reading measurement data.

FIG. 8 is a flowchart showing a procedure for reading measurement data.

FIG. 9 is a flowchart showing a gain determination process.

FIG. 10 is a flowchart showing a calculation processing procedure of addition measurement data and subtraction measurement data.

FIG. 11 is a flowchart showing the first half of distance measurement processing.

FIG. 12 is a flowchart showing the latter half of the distance measuring process.

FIG. 13 is an explanatory diagram showing an example of a distance measurement scene.

FIG. 14 is a flowchart showing the first half of another example of the gain determination process.

FIG. 15 is a flowchart showing the latter half of gain determination processing.

FIG. 16 is a flowchart showing another embodiment of distance measuring processing.

FIG. 17 is a waveform diagram showing a noise component superimposed on the measurement data as the fluorescent lamp blinks.

FIG. 18 is a timing chart showing the correlation between the lighting period of a fluorescent lamp and the projection interval of distance measuring light.

[Explanation of symbols]

 2 Light projecting parts 3a to 3c LED 5 Photographing lens 7 Light receiving part 9 PSD 14 Microcomputer 28a, 28b Preamplifier 29a, 29b Gain control amplifier 30a, 30b Sample hold circuit 31a, 31b Buffer amplifier SP1 to SP3 Distance measuring point 40 Shooting scene

Claims (1)

Claim: What is claimed is: 1. A light projecting means for projecting distance measuring light to a subject, a light receiving means for receiving distance measuring light reflected from the subject, and an output signal of the light receiving means. In a range finder having a gain control amplifier for amplification and a calculation means for calculating a distance signal from the output signal of the gain control amplifier, the level of the output signal of the gain control amplifier is monitored and the level is out of a predetermined range. In this case, the gain of the gain control amplifier is adjusted, and when the gain is changed, the start of the distance measurement sequence is delayed by a predetermined time or more. 2. A light projecting means for projecting distance measuring light toward a subject, a light receiving means for receiving distance measuring light reflected from the subject, and a gain control amplifier for amplifying an output signal of the light receiving means. In a range finder having a calculation means for calculating a distance signal from the output signal of the gain control amplifier, the level of the output signal of the gain control amplifier is monitored, and if the level is out of a predetermined range, gain control is performed. A distance measuring method characterized by adjusting the gain of an amplifier and delaying the generation of a signal for allowing the operation of the light projecting means when the gain is changed. 3. A light projecting means for projecting distance measuring light toward a subject, a light receiving means for receiving distance measuring light reflected from the subject, and a gain control amplifier for amplifying an output signal of the light receiving means. In a distance measuring device having a calculating means for calculating a distance signal from the output signal of the gain control amplifier, the light emitting means is set to P at the light emitting interval AT seconds during distance measurement.
The light is emitted N times, the output signal of the gain control amplifier is monitored during the distance measurement, and when the number of times that this output signal is out of the predetermined level range reaches CM times, the gain of the gain control amplifier is adjusted and then measured. A distance measuring method characterized in that the distance is redone and the value of the CM is determined from the following equation. CM ≧ AT × PN × 120 5. A light projecting unit for projecting distance measuring light toward a subject, a light receiving unit for receiving the distance measuring light reflected from the subject, and an output signal of the light receiving unit. In a range finder having a gain control amplifier for amplifying the signal and a calculation means for calculating a distance signal from the output signal of the gain control amplifier, when the output signal of the gain control amplifier exceeds a predetermined level, 1 msec to 5 ms.
After elapse of time ec, the output signal is captured again with the same gain value. At this time, if the output signal of the gain control amplifier exceeds the predetermined level again, the distance signal is calculated based on the output signal of the gain control amplifier. Distance measurement method characterized by stopping.