JP2710240B2 - Automatic focusing device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial applications]   The present invention relates to an automatic focusing device for a camera.
Focus adjustment, especially for moving subjects.
Automatic focusing device to be used. [Prior art and its problems]   Automatic focus adjustment (auto focus, hereinafter abbreviated as AF)
In the case of a camera equipped with a camera,
Calculate the defocus amount for the focal plane, and calculate the result
Drive the lens based on the
ing. If the subject is a moving object and the defocus amount
Changes, the defocus due to the movement of the subject
Determine the rate of change of the amount, and use this to
Performs correction and focuses on moving subjects
Cameras that match are already proposed.   The focus is on the output of the focus detection element for performing focus detection.
Is affected by noise such as electrical noise and dark output of the device.
There is a gap and moves based on such detection results
If you calculate the change speed of the defocus amount for the subject,
The error increases and stable correction cannot be performed.
There was a problem that accurate AF could not be performed. [Object of the invention]   The present invention has been made to eliminate the above problems.
It is fast and appropriate for the moving speed of the subject.
To provide an automatic focusing device from which focusing can be obtained
Aim. [Configuration of the Invention]   The automatic focus adjustment device according to the present invention is capable of receiving a plurality of pixels.
The output from the optical element is applied to the expected focal plane of the taking lens.
Based on the detection result of the focus detection means that performs focus detection.
In an automatic focusing device for adjusting the focus of a shadow lens,
The focus detection means is driven in time series to output a focus detection value.
Control means, and the control means performs time-series detection.
Based on the multiple focus detection values issued,
Calculates the amount of change in the focus detection value caused and outputs it in chronological order
A first calculating means, and a time series output from the first calculating means.
Previous data in chronological order according to the magnitude of the input change
Perform arithmetic processing with different gravities on subsequent data in chronological order
Selecting means for selecting, and time series from the first calculating means.
The output change amount is selected by the selection means.
Second arithmetic means for performing the calculated arithmetic processing, and the focus detection value
The processed change amount calculated by the second calculating means.
Driving means for driving the taking lens based on the
It is characterized by the following. [Example]   Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings.
Explain physically.   FIG. 1 shows a block configuration of the camera.
The left side of the line A-A 'shows the interchangeable lens LZ, and the right side
3 shows a main body BD of the camera. Both are clutch 101,
It is possible to mechanically couple by 102
The interchangeable lens LZ is connected to the camera body BD by the clutches 101 and 102.
The lens circuit 103 on the interchangeable lens LZ side when mounted on
And the reading circuit 104 on the camera body BD are connected to connection terminals JL1 to JL.
5, electrically connected by JB1 to JB5
You.   In this camera system, the lens system of the interchangeable lens LZ
The passed subject light is in the center of the reflection mirror 105 of the camera body BD.
And is reflected by the sub-mirror 106.
Light is received by the CCD image sensor 107 in the focus detection module.
The optical system is configured to operate. This CCD image
The sensor 107 measures the defocus amount of the interchangeable lens LZ
A plurality of photoelectric conversion elements.
Signals from each photoelectric conversion element
A well-known one that can be used can be used.   The interface circuit 108 is a CCD image sensor 107
Drive or subject data from the CCD image sensor 107.
And the captured data
To the controller 109. Controller 109 is a CCD
Focus position based on subject data from image sensor 107
Z-focus amount indicating the amount of deviation of the lens from the position | Δ |
And whether the lens position is shifted forward (front pin) or rear
To the direction (rear pin)
The signal with the scum direction is calculated.   The motor MO1 is driven based on these signals, and the
Rotation is performed by slip mechanism 110, drive mechanism 111 and clutches 102,1
01 is transmitted to the transmission mechanism 112 via the
Adjust the focus by moving the lens system of the lens LZ back and forth in the optical axis direction.
Do. At this time, monitor the amount of movement of the lens system.
Encoder 113 is connected to the drive mechanism 111 of the camera body BD.
Drive of the lens system from the encoder 113
The number of pulses corresponding to the amount is output.   Note that the slip mechanism 110 is fixed to the driven part of the interchangeable lens LZ.
When the above torque is applied, the power of the motor MO1 slips.
To the motor MO1.
The load is not applied.   Here, the reading circuit 104 on the camera body side
To the lens circuit 103 via the terminals JB1-JL1
Clock for synchronization during data transfer via pins JB2-JL2.
Pulse and then read data through terminals JB3-JL3
A reading start signal for starting the reading is transmitted.
Also, terminals JL4-JB4 are connected from the lens circuit 103 to the reading circuit 104.
Serial data is transmitted via the. In addition, terminal JB51
-JL5 is a common ground terminal.   First, a read start signal is sent to the lens circuit 103
And the data of the lens circuit 103 is synchronized with the clock pulse.
The reading circuit 104 is sent out. The reading circuit 104 is a terminal,
Clock pulse output to the
Input serial data in parallel based on the
The data is converted into data KL and sent to the controller 109.   The controller 109 is based on the data K from the reading circuit 104.
Of the subject image from the interface circuit 108.
The amount of defocus | Δ |
Calculates the number of pulses N to be detected by K · | Δ |
I do. Further, the controller 109 determines whether the data
Motor drive according to the signal in the defocus direction
Move motor MO1 clockwise or counterclockwise through valve 114
And it is equal to the calculated value N from the encode 113.
When the pulse is input to the controller 109, the focus
The node lens system moves by the moving amount Δd to the in-focus position.
Then, the rotation of the motor MO1 is stopped. like this
If the focus is adjusted by proper focus adjustment, the controller 109
A predetermined signal is sent to the display circuit 115 from the
And the distance to the subject are displayed.   The schematic operation of the camera has been described above. Next, FIG.
More detailed control operation in controller 109 using
Will be described. The same parts as those in FIG.
Signs are attached.   109 is a matrix that operates in the controller described above.
It is a microcomputer (hereinafter referred to as a microcomputer), and 12
1 opens and closes the shutter according to the start and end of exposure
In addition, the mirror 10
Exposure control circuit for mirror up and aperture control
You. Reference numeral 122 denotes a photometry circuit, and a signal corresponding to the brightness of the subject
The number is digitized and sent to the microcomputer 109. 123
Film sensitivity automatic reading circuit that reads the loaded film sensitivity
And the read film sensitivity is taken into the microcomputer 109.
It is. 124 is a motor MO based on a signal from the microcomputer 109.
Drive 1 to wind up the film one frame
125 is for setting the shutter speed and aperture value.
The exposure setting circuit is
I will.   107 is the CCD image sensor described above, and
The image information is transferred to the microcomputer 1 via the interface circuit 108.
Incorporated in 09. 114 controls the focus adjustment motor MO1
The motor driver 113 controls the driving amount of the motor MO1.
This is an encoder that outputs the number of pulses. 103 and 104
Denotes a lens circuit and a reading circuit, and 115-1 and 115-2
Indicates the focus detection status and the distance to the subject, respectively.
Display circuit.   S₁Is turned on at the first stage of pressing the release button (not shown).
This switch S₁By turning on
The microcomputer 109 is turned on and executes the AF flow described later.
I do. S_TwoIs on in the second stage of pressing the release button
This switch S_TwoBy turning on
The release flow described above is executed. S_ThreeIs the reflection mirror
Switch that is turned on when the
When the release member is charged, switch S_ThreeIs off
Becomes S_FourSwitches between continuous shooting mode and single-frame shooting mode.
Switch for changing the shooting mode._Five
Turns on when exposure is completed, and winds up one frame of film.
A switch that is turned off upon completion. Each of the above switches S₁
~ S_FivePrimary side is grounded and connected to the microcomputer 109
The next side is pulled up to voltage V via resistor R respectively.
ing.   Next, the control operation of the camera with the above configuration
Explanation will be given according to the   By pressing the release first stage, switch S₁But
When the microcomputer 109 is turned on,
Run   First, in step S1, various flags are reset,
Microcomputer to know the time in each operation in step S2
The free-run timer in 109 is started. And
In step S3, AF is performed from the lens circuit 103 through the reading circuit 104.
Lens focal length and aperture value required for
Indicates the amount of movement of the lens in the optical axis direction with respect to the number of rotations.
Feeding amount conversion coefficient K, lens aperture F value A₀₀, Lens
Data Dkl for conversion into
The data is taken into the microcomputer 109. Next step S4
The set aperture value and shutter speed etc.
Read from the road 125, and will be described later in the next step S5.
Focus detection calculation and focus on the lens based on this calculation result
An AF operation driven to the position is performed. In step S6
Receives the photometry result obtained by the photometry circuit 122, and
In step S7, the film sensitivity reading circuit 123
Film speed is captured. In step S8,
Exposure calculation for exposure was performed by input data
Thereafter, the process returns to step S3 to perform a loop operation.   FIG. 4 shows the AF operation routine in step S5.
Is shown.   First, in step S401, the interface circuit 108
The subject light is integrated by the CCD image sensor 107 through the
Be done. In the next step S402, the CCD image sensor 107
The integrated subject data is extracted for each pixel (this
Operation is called data dump), interface circuit
After being A / D converted in 108, it is taken into the microcomputer 109.
In step S403, focus detection is performed using the data of the subject.
The above operation is performed. In addition, the optics to which the subject light is input
A detailed description of the system and the focus detection calculation is not necessary here.
The details are described in JP-A-59-126517.
ing.   In the next step S404, tracking correction is performed as described later.
A preliminary operation is performed. In step S405, the CC
Defocus amount based on data from D image sensor 107
And its direction are calculated, and the defocus
A determination is made as to whether the amount of data can be detected. Subject image
If the image is greatly blurred or has low contrast,
Proceeding to step S406 is determined to be impossible. Step S406 smell
To find the focus detectable part when focus cannot be detected
Operation to move the lens to the
Is terminated. Lowcon ska
If not, the low-con
The scan starts. And this low contrast scan
Is repeated until the low contrast scan is completed.
If the point cannot be detected, the display circuit goes to step S408.
A blinking display indicating that focus detection is not possible is displayed on 115-1.
After that, return to step S6 in FIG.
I do.   On the other hand, in step S405, the defocus amount can be detected.
If it is determined that the function is
And the defocus amount DF obtained in step S3 in FIG.
Conversion of lens extension which is one of the lens data
Lens driving amount ERR from coefficient K (pulse count unit)
Is calculated.   ERR = DF × K   In step S410, it is determined whether the lens is stopped.
Is stopped, the lens is focused in step S411.
It is determined whether the lens is in the state and the lens is
Then, in Step S412, the in-focus display is displayed on the display circuit 115-1.
After that, the process returns to the flow of FIG.
Proceed to. On the other hand, if the lens is out of focus, go to step S413
Go to the defocus direction obtained by this routine execution
Is the defocus calculated in the previous routine execution.
Direction is determined to be the direction opposite to the
If the orientation is reversed, it may cause an error when the lens drive is reversed.
Backlash caused by backlash in the lens drive system
Is corrected, and the process proceeds to step S424. Lens drive direction is the same
If so, the process proceeds to step S414 to perform the following correction described below.
Judgment is necessary whether AF drive mode is required, and tracking correction
If the mode is determined, follow-up correction is performed in step S415.
Thus, the driving amount of the lens is corrected. In the next step S416
The focus determination for the tracking mode is performed, and the focus state is determined here.
If set, the focus is displayed in step S417, and then step S4
Proceed to 24.   On the other hand, if the lens is being driven in step S410,
Proceed to step S421 and proceed to step S415, which was determined this time.
The defocus direction including the correction
Is compared with the focus direction and it is determined that the direction is reversed.
Then, after the driving of the lens is stopped in step S423,
To return. The reason for stopping the lens here is that the lens
This is for minimizing the overshoot of the in-focus position. Differential
If the focus direction has not been reversed, proceed to step S422.
Whether the following correction is necessary as in step S414
If a determination is made and tracking correction is necessary, go to step S415.
The process proceeds to step S424 when unnecessary.   In step S424, according to the obtained defocus amount,
It is determined whether the lens is near the focus (ani zone),
If not in the near zone, the lens is high in step S425
Set to be driven at speed, in the near zone
If there is, the lens is driven at low speed in step S426
Is set to be Then, in step S427, the session
After the lens is driven at the
The focusing calculation is performed in step S6 and subsequent steps.   FIG. 5 is a diagram of FIG. 3 for explaining the tracking correction.
Steps S3 to S6 are described in more detail. Stay
At the start of CCD image sensor 107 integration in step S501
Time TM1 of the free-run timer
At 502, the lens drive amount at the start of this integration is controlled.
Event counter EVTCNT for trawling
Event value T1 is read. CCD image at step S503
The integration of the disensor 107 is performed. In step S504, the CCD
The integration end time TM2 of the image sensor 107 is a free-run timer.
From step S505.
The count value T2 is read by the counter EVTCNT. So
After that, in steps S506 and S507, as described above, the CCD image
The data dump of the image sensor 107 and the focus detection calculation are performed.
Be done. Most of the time required to execute this routine
Are spent for steps S501-S507 above. rear
As described above, the integral obtained by executing the AF routine
The value MI of the lens drive event counter EVTCNT at the center point is
This is stored in the register R. In step S508, this
The event count value MI stored in the register R is
Stored as the MIL in R 'in the register that stores the operation value
Is done. In step S509, similarly, the time at the integration central point
TMI writes TMIL in the register that stores the previous operation value.
Remembered. In step S510, the current routine execution
Event counts at the start and end of integration
From the EVTCNT count values T1 and T2
Value (T1 + T2) / 2 is calculated, and a new MI
Is stored as In step S511, similarly,
From the times TM1 and TM2 at the start and end of integration,
Time (TM1 + TM2) / 2 is calculated and stored in the register.
You.   In the next step, from the previous integration center to the current integration
The amount of lens drive between the hearts is calculated.
This will be described with reference to the drawings.   The horizontal axis is time, and the vertical axis is the value of the event counter EVTCNT.
I do. T1₁, T1_TwoIs the integral I₁, I_TwoAt the start of TM1₁, TM1_Two
The event counter EVTCNT count value at
₁, T2_TwoIs the same integration end time TM2₁, TM2_TwoEvent ca
Counter value, T3₁, T3_TwoAre the same integral
Operation C performed later₁, C_TwoEnd time TM3₁, TM3_TwoEvent council
This is the count value of EVTCNT. MIL and MI are T1
₁And T2₁, T1_TwoAnd T2_TwoFrom the count value of the central point of integration obtained from
is there. Where TM3₁= TM1_TwoIt has become. Register R,
The value of the event counter EVTCNT stored in R ′ is
Is rewritten when the computation of
C_TwoEnd time of TM3_TwoEvent counter EVTCNT
Und value T3_TwoThe event count value MI in the register R
Is stored and the previous operation C₁B obtained by
The vent count value MIL is stored in the register R '. This
The event count value of
Is the value obtained by converting the amount of focus into the amount of movement of the encoder.
Defocus from subject image plane at each integration center
The position indicating the amount of data is shown.   However, as shown in FIG.₁Is over and next
Times integration I_TwoIs disconnected from the event count value when
Continuation occurs. This is due to the characteristics of the CCD image sensor 107.
TM3₁Integral T₁Focus detection result
The focus detection system is used to rewrite
Due to the error of The value of MIL-MI is calculated in one cycle of integration.
This does not indicate the exact amount of lens movement. This
Assuming that the gap at the discontinuous part of DT is DT, this value DT
C₁At the end of the calculation, the result of the calculation is
The set value SERR and the lens position at this time are shown.
Event counter EVTCNT value T3 (= T3₁) And SERR
Given by −T3. Thus, one cycle of the focus detection time
The amount of movement ITI of the lens within
Obtained by subtracting the value (MI-DT) subtracted by the DT
It is. That is,   ITI = MIL− (MI−DT).   In steps S512 and S513, the above DT and ITI calculations are performed.
After that, in step S514, the event
The value T3 of the counter EVTCNT is read. And step S5
At time 15, the time TM3 at the end of the operation is read, and thereafter the third time
The process proceeds to step S7 in the figure.   Now, FIGS. 6 and 7 show step S4 in FIG.
Flow after judging whether the lens is stopped or driving in 10
Is described in detail.   FIG. 6 shows a flowchart when the lens is stopped.
This mode is used when the flow is first executed
Or at the time of in-focus confirmation or in-focus. S
At step S601, the lens is currently within the predetermined focusing zone.
It is determined whether or not there is, and if it is within the focusing zone, step S602
The lens drive amount ERR calculated this time is
As the previous lens drive amount LERR
The lens drive set value EV calculated this time
TCNT is similarly set to SERR. Now in step S603
The previous result is the same as the defocus direction obtained
And as described later in step S604
Along with the in-focus display, the distance to the subject is displayed on the display circuit 115-2.
The release is displayed (details will be described later), and then the third
Returning to step S6 in the figure, focus detection is repeated again.
It is.   On the other hand, if it is determined in step S601 that it is out of the focusing zone,
Proceed to step S605 to determine whether the execution of the flow is the first time
If it is the first execution, proceed to step S606 and follow
The flag is reset. This tracking flag is used for focus detection.
To enter the follow-up mode to make corrections to improve the follow-up
It is used to determine whether In step S606 ′,
Past data WR used for filtering
(1), WR (2), LWR (1) are reset. Soshi
In step S607, the event counter EVTCNT is counted.
The non-update flag, which prohibits value updates, is reset and the next
Calculation result obtained in the tracking mode in step S608
The tracking correction flag for sequentially correcting
You.   When the next flow is executed, step S605
Proceed to S610, S611, S612, and
The focus direction is compared with the current defocus direction.
You. If the direction is reversed, in step S613 the lens
Drive backlash is compensated and then
Proceed to S606, the follow-up flag is reset as the initial setting
You. If you follow this mode, you can leave this mode
It means. On the other hand, if the defocus direction is the same
For example, the process proceeds to step S614, where the movement amount WR of the subject is calculated.
This calculation method will be described with reference to FIG.   The vertical axis shows the focal position of the lens, and the horizontal axis shows time
I have. A in the figure₁₆Indicates the true change of the moving subject, B
₁₆Is a lens driven to follow the subject.
The distance between two points at the same time is
This shows the amount of scum. Currently, operation C_ThreeAt the end of TM3
And the previous integral I_TwoAnd this integration I_ThreeIntegration point TM with
The amount of movement of the subject detected in one cycle between IL and TMI
WR has already been found and the integral I_Two, I_ThreeCenter point TMIL, TM of
The defocus amounts DFb and DFc at I are also obtained. TMIL
The driving amount of the TMI lens is ITI.
Integral I_TwoThe center point TMIL and the integral I_Three1 between the center point TMI of
As a cycle, the life caused by the movement of the subject detected during this period
The amount of defocus WR is   WR = ITI + DFc-DFb Becomes Where DFc is the current defocus amount,
Set to ERR in pulse count units at step S409
DFb is the previous defocus amount, and
Set to LEER in pulse count units at 624
You. Therefore, the above equation is   WR = ERR + ITL-LERR Is required.   Returning to FIG. 6, the process then proceeds to step S614 'to read the value WR.
The calculation processing is performed as follows.
Is called filtering for convenience. Fig. 6
An example of the filtering process in step S614 'is
This is shown in FIG.   In focus detection, a CCD image sensor 107 is generally used.
Values may vary due to noise caused by noise or circuit noise.
A crack occurs. That is, ERR and LER in step S614
Since the value of R varies, the obtained defocus
The amount WR also includes variations. Therefore, this
If the correction is made with the defocus amount WR,
As a result, stable control could not be performed, causing problems in AF accuracy.
You. Therefore, in order to absorb this variation, the past detection
The average value of the data is taken into account. That is,
In step S1401, the unit is calculated by the calculation of WR / (TMIL-TMI).
The change amount WR (0) of the defocus amount per time is calculated.
It is. This change amount WR (0) is the defocus light of the subject.
The velocity in the axial direction is shown. Next step S1402
Then, the value of the change amount WR (0) is larger than the predetermined value HS.
Is determined, the moving speed of the subject is high, and the change amount WR
If (0) is larger than the predetermined value HS, step S1403
To the next calculation described in the next step S1404.
The changed amount LWR (1) and the current amount WR (0)
The averaged change LWR (0) is calculated by the following equation.
You.   Then, in step S1404, the process proceeds to the next cycle.
To use as the previous data,
LWR (0) is set as LWR (1). On the other hand,
The moving speed of the object is low and the change amount WR (0) is less than the predetermined value HS.
If it is smaller, the process proceeds to step S1405 to calculate the following equation
, The averaged change amount LWR (0) is calculated, and step S14 is performed.
Go to 04.   Therefore, the averaged change amount LWR obtained in step S1403
In (0), past data is stored in a relationship as shown in the following equation.
Included, as well as the averaging change obtained in step S1405
In the quantity LWR (0), the past data
Data will be included.   WR (i) is the defocus amount WR before i times
And this filtering is linked to the ranging cycle
Therefore, the value of i is finite. Where the above two equations are compared
By comparison, the first expression is the past data compared to the first expression.
The specific gravity of the container is small. In other words, adopt the formula
If so, we ’re focusing on the current data,
Excellent responsiveness.
Of the above data,
This is advantageous for   This will be described with reference to FIG. In FIG. 17, the vertical axis is the data
The rate of change of the focus amount, and the horizontal axis shows time.
You. And the speed of change of the focal position due to the true movement of the subject
A degree₁₇Indicated by. The past data shown in the equation here
If filtering with large specific gravity is performed, B₁₇Indicated by
As described above, it is detected that the subject is moving. That is,
In the area of D where the moving speed of the object is small, the accuracy is high and stable.
Area where the moving speed of the subject changes greatly
In this case, a response delay occurs. New data represented by formula
If filtering with large specific gravity is performed, C₁₇Shown in
There is no stability at all, but responsiveness is good. That is. D
Expression filtering is more accurate and stable in
In the region of E, the response is
Filtering is advantageous.   That is, if the moving speed of the subject is low,
Past data that has been adopted and average data obtained
The emphasis on
Reliable data can be obtained. Conversely, the subject moved rapidly
In such cases, the formula is adopted, and
By focusing on the data obtained at the moment,
Improved responsiveness and the ability to follow fast-moving subjects
Become.   The focal position of the lens is almost inversely proportional to the position of the subject.
When the subject moves at a constant speed in the optical axis direction,
Inversely proportional. Also, the changing speed of the focal position of the lens
It is inversely proportional to the square and changes rapidly. others
To switch the filtering method in this way
Is extremely effective.   In step S1402, the change amount WR (0) is fixed.
Compared to the value HS, but the previous averaged change LWR (1)
You may make it compare with. In addition,
As before, we should not take an exponential average of past data.
As shown in Fig. 14 (B), a simple weighted average is used.
May be used. That is, the change amount WR (0) of the subject movement is
When it is larger than the fixed value HS, the process proceeds to step S1406,
The average change LWR (0) is calculated by the following formula.
You.   WR (1) and WR (2) are the changes of the previous and previous times, respectively.
Is the amount of On the other hand, the change amount WR (0) of the subject movement is predetermined.
If the value is smaller than the value HS, the process proceeds to step S1407, and the next
The average change amount LWR (0) is calculated by the calculation of the equation.   That is, the formula is excellent in responsiveness similar to the above formula,
The formula is also excellent in stability in the same way as the formula.
You. Then, in step S1408, the previous change amount WR
(1) is set as the amount of change WR (2) two times before. S
In step S1409, the change amount WR (0) calculated this time is before
It is set as the amount of change WR (1).   FIG. 14 (C) and FIG. 14 (D) show the filters respectively.
14 shows another embodiment of the ring processing.   FIG. 14 (C) shows the processing performed in FIG. 14 (A).
Stable without determining the size of the moving amount WR (0) of the object
Of average moving distance LWR (0) with emphasis on performance
I have. FIG. 14D shows the magnitude of the moving amount WR (0) of the subject.
Without performing the judgment of the
By the weighted average of the data WR (0) to WR (2),
The leveling movement amount LWR (0) is calculated. In this case, the 14th
As in step S1407 of FIG.
May be added. That is, the calculation in step S1410
May be expressed by the following equation.   Also, here, the data of the past two times is adopted.
From the processing time and the capacity of the microcomputer 109, more past data
Data may be used.   Also, filter the defocus speed WR (0) of the subject.
Instead of moving the subject
WR may be used. Alternatively, the focus detection value itself
Filtering, then the defocus speed of the subject
May be required. In addition, filters that emphasize current data
If you select the filtering method, the subject moves at high speed.
In order to indicate that the lens is moving,
It is better to control at high speed.   Returning to FIG. 6, in the next step S615, the state of the following flag is set.
The state is determined. Describe the operation according to the flow of this flow
Then, as described above, the first execution of this flow
Since the follow-up flag has been reset,
The process proceeds from step S615 to step S616. Here, the flat
The magnitude of the leveling movement amount LWR (0) is determined. Here
The solid predetermined value AB is the variation in the calculation result and the focusing zone.
It is regarded as the in-focus zone even if the tracking correction is not performed in consideration of the width of the
Means the value set for the defocus speed.   Now, if the average moving amount LWR (0) is less than the predetermined value AB
For example, the process proceeds to step S607, but the average moving amount LWR (0) is
If the value is equal to or greater than the fixed value AB, the following flag is set to 1 in step S617.
After the setting, the process proceeds to step S607.   In this case, the next time the flow is executed, the following flag is set.
The tracking mode is set by the
The process proceeds to S618 through S601, S614, and S615. Here are the steps
Similarly to S161, the average moving amount LWR (0) is equal to the predetermined value AB.
Comparing, if LWR (0) is less than the specified value AB, follow-up correction
The process proceeds to step S607 without being performed, but LWR (0) is
If the value is equal to or larger than the fixed value AB, the process proceeds to step S619. Where
Another movement amount WR of the subject is considerably larger than the predetermined value AB.
It is compared with the predetermined value AX, and if WR is less than AX, step S621
The non-update flag is reset at
The optional correction flag is set.   On the other hand, when the movement amount WR is equal to or more than AX, the process proceeds to step S623.
Then, the non-update flag is set, and thereafter, step S608
And the tracking correction flag is reset. This flow
Is changed by changing the direction of the camera during follow-up mode.
This is when the amount of movement WR of the object has changed
In this case, the event counter EVTCNT is not updated and
Counter value is left as it is, tracking correction is also prohibited,
AF control is performed with the count value
Also, for this reason, follow-up corrections are also prohibited.
I have. Here, the value WR before filtering is used
Response may be delayed due to filtering
Instead, it can be detected at high speed.   When the next flow is executed, the value that has changed greatly
Since it has been replaced as the driving amount LERR,
If the direction of the camera remains unchanged, this time the subject moves
Since the amount WR does not change significantly, go to steps S621 and S622.
The non-update flag is reset, and the tracking correction flag is
Is set.   Thus, when the defocus amount changes greatly,
Indicates that the operation result at this time is ignored once and the previous data
As a result, AF control is performed, and after that
If the camera's orientation remains the same, then move the subject
When the amount WR becomes less than AX and proceeds to step S621,
AF with the greatly changed lens drive amount LERR
Control is exercised.   After step S608 and step S622, step S624
And the lens drive amount ERR of this time is
In step S625, the data
The focus direction is replaced as the previous direction. So
Then, in step S626, the tracking correction flag is determined,
If the correction flag is set to 1, in step S627
The tracking correction, which will be described in detail later, is calculated, and the lens driving amount ER is calculated.
The value of R is corrected. Therefore, in step S624,
Lens drive that does not follow the lens drive amount LERR
The quantity ERR is stored. Next, in step S628, follow
If one set has been set, the additional
It is determined whether the subject is within the focus zone and the focus zone
If it is within the focus range, the focus indication and distance
After that, the display in FIG. 4 is performed.
Proceed to S424. On the other hand, in step S628, the tracking flag is reset.
Is out of focus and in step S629
If there is, the process proceeds to step S424. With the following focus zone
Is not in the focusing zone at the end of the current calculation
It may be, but the effect of tracking correction in the future timing
Appears in the focus zone, and in this case
Is displayed while the lens is being driven.   FIG. 8 shows the calculation of the following correction in step S627.
This correction method is described in detail in Fig. 18.
It will be described using FIG.   Currently calculated C_ThreeAt the end of the TM3, already mentioned
U, the last integration I_TwoAnd this integration I_ThreeIntegral center point TMI with
The movement amount WR of the subject for one cycle between L and TMI has already been obtained.
And also the integral I_ThreeAt the midpoint TMI
The scrap amount DFc has also been determined. However, in the figure, line A₁₈
Indicates a true subject movement, and the detected defocus amount is
As already mentioned, for reasons such as electrical noise and dark output
Has more error. Where integral I₁, I_Two, I_ThreeDefoe in
Let the amount of scum be DFa, DFb, DFc and determine
If so, the subject movement amount WR naturally has an error. Time
Point TM_ThreeThe object movement amount WR detected in₁₈Like
Line A indicating that the subject is moving, indicating true movement
₁₈Generates a large error as compared with. In this embodiment,
Filtering and detection results as described in
In order to absorb the variation of₁₈Indicated by
As a result, it is possible to suppress errors. this
This will be described in more detail with reference to FIG. In FIG.
The vertical axis indicates the amount of defocus, and the horizontal axis indicates time.
You. A for true movement of the subject₁₉Shown in the Defoca
Are shown by white circles. The lens is for simplicity
It is assumed that it has stopped. In the figure, at time TM3, the integral I
_Two, I_ThreeThe change in defocus of the subject is detected using the detection value at
If this is the case, the variation in the detected values at these two points
Affected, C₁₉When the subject is moving as shown
I will refuse. With filtering, the integral I_Twoprevious
The value detected by integration is also used to detect defocus changes.
Will be. That is, the integral I_-Five~ I_ThreeAt the time of
Straightened D₁₉′
The occurrence of errors can be suppressed. Straight line D₁₉′ Slope
The average moving amount LWR (0) is shown. Straight line D₁₉Is a straight line
D₁₉′ And integral I_ThreeIs a straight line that passes the detection value of
Line D in the figure₁₈Corresponding to Returning to Fig. 18,
The subject moves further even while the lens is
To correct the defocus amount using the current time TM3 as the target
It would be pointless even though. Then, the next integral I_FourAnd operation C_FourBut
TM at the end of calculation of new defocus amount DFd
Correcting with the target of 3 'is efficient and appropriate
I can say. Therefore, the actual defocus amount DFc is
The relative movement amount X of the subject between the time TMI and the time TM3 '
Should be corrected. From time TMIL to TMI
The moving speed of the subject is LWR (0)
I have been asked.   The movement amount of the subject from time TMI to time TM3 is (TMI-T
M3) × LWR (0), focus detection cycle is almost constant
If TM3−TM3 ′ = TML−TMI is approximated, TM3 becomes TM
The movement of the 3 'subject is   (TMIL−TMI) × LWR (0).   Therefore, the movement amount of the subject from time TMI to time TM3 ′
X is   X = {(TMIL−TMI) + (TMI−TM3)} × LWR (0)     = (TMIL-TM3) x LWR (0).   The obtained movement amount X of the subject is determined in step S627-1.
Is set as the correction amount SWR in the next step S627-
In 2, the lens drive amount ERR is corrected by this correction amount SWR
Is done.   Note that here, the lens is
The case of catching up is discussed, but the subject is
Even when moving in the direction
Similar control is performed.   FIG. 9 is a table showing steps S604 and S630 in FIG.
FIG.   In step S901, the lens extension amount absolute counter LN
Addition value of SC count value and lens drive amount ERR
Set in distance counter DSCN. Extend this lens
The quantity absolute counter LNSC is 0 when the lens is in the ∞ position.
And counts in proportion to the amount of lens extension.
Therefore, the distance counter DSCN indicates the distance to the subject.
A count value corresponding to the distance is input. Step S902
Then, the count value of the distance counter DSCN and the distance
Distance conversion data Dkl specific to the lens
Is multiplied and converted to data called the display parameter DSPS.
Is done. This data is not distance data, but
The distance display data is sent to the display device.
The distance is displayed in step S903.   Referring back to FIG. 4, it is determined that the lens is being driven in step S410.
The operation when the determination is made will be described in detail with reference to FIG.   In steps S701 to S703, the driving direction of the lens and the movement
Defocus direction after minute correction is compared and direction is reversed
Then, the process proceeds to step S704, and the following flag is reset.
You. This means that the lens catches up with the subject during
The tracking mode has been canceled because
You. Then, in step S704 ', the past data
WR (1), WR (2), LWR (1) are cleared and the next
After the driving of the lens is stopped in step S705,
The process returns to step S6, and the distance measurement calculation is performed again. one
If the driving direction of the lens is the same as the previous driving direction
If it is, the process proceeds to step S706, and the movement amount WR of the subject has already been described.
Thus, ERR + ITI-LERR. And
In step S706 ', a similar filtering process is performed.
It is. In the next step S707, follow by judging the flag
Mode is determined, and if it is in the following mode, step
The process proceeds to S708 to S714.
The description is omitted because it is the same as step S618 and subsequent steps in
You. On the other hand, if the mode is not the following mode, the process proceeds to step S715.
The average moving amount LWR (0) is smaller than the predetermined value AB.
Or a small predetermined value AA, it is determined whether it is large.
Proceed to steps S712 and S714, and the non-update flag and follow-up correction
The flag is reset, while the average moving distance LWR (0)
Is greater than or equal to the predetermined value AA, the following flag is set in step S716.
Is set, and then in steps S711 and S713,
New flag reset and tracking correction flag set
It is. This point is different from the stopped mode.
In high-speed movement mode, high-speed control is required.
The tracking correction flag is set so that the tracking is corrected
You. After step S713 and step S714, the process of FIG.
Proceed to step S624.   FIG. 10 illustrates steps S424 to S427 in FIG. 4 in detail.
Well written.   First, in step S1001, the state of the non-update flag is determined.
If the flag has been reset, go to step S1002.
It is determined whether the lens is currently being driven and
If it is medium, capture the subject data in step S1003
Lens movement amount CTC (= MI-T
3) is calculated, and in step S1004, the lens is driven.
The quantity ERR is corrected. That is, the time of data input and the operation
A lens that has a count error from the time when the end result was obtained
The movement amount CTC is corrected, and then the process proceeds to step S1005.
No. If the lens is not being driven, the above steps S1002, S100
3 skips. In step S1005, the lens driving camera
Determines whether the und value ERR is in the near zone NZC.
If it is within the near zone NZC, the
Drive the lens at low speed to increase focus accuracy.
The lens drive count value ERR
Is set in the event counter EVTCNT. On the other hand,
If it is outside the near zone NZC, in step S1007,
Speed is set and the lens drive count value ERR
Is set in the event counter EVTCNT. In step S1008
Is used to calculate the offset for the next event count.
The count value of the vent counter EVTCNT is used as the lens drive amount SERR
Is set as For AF drive in next step S1009
After the motor is energized, return to step S6 in FIG.
I do.   FIG. 11 shows an interrupt for controlling the lens drive amount.
It shows only the routine, and
Execute this routine every time a pulse is output from the coder
I do.   First, in step S1101, an event indicating the lens drive amount is set.
The count value is decremented by one. In step S1102, the
The event count value becomes 0 and the target lens drive amount is driven.
Is completed, and if it becomes 0, the process proceeds to step S1103.
Thus, the motor is stopped, and thereafter, the process returns. Rita
If the result of the stopped operation is within the focusing zone after
The focus is displayed.   FIG. 12 shows that the relay routine is executed during the execution of the AF routine shown in FIG.
Switch S by pressing_TwoOccurs when is turned on
An interrupt routine is shown.   In step S1201, it is determined whether the mode is the following mode.
If the following flag is set, go to step S1202.
The tracking correction of the lens drive amount is calculated, and the exposure starts.
The lens is driven based on the result of this correction. this
The correction is performed from the time the release signal is input until the start of exposure.
The amount of movement of the subject during the
This is to correct for the counter EVTCNT. Tracking correction
If the lag has been reset, go to step S1203.
Then, the driving of the lens is stopped. This is during follow-up mode
Is not possible because the lens is relatively close to the subject.
Controlled by speed, but in normal mode not following mode
In that case, high speed is also possible, in this case the lens
Even if the stop signal is issued, it takes a little
Takes a long time and delays up to the release time lag
Stop the lens unless it is in tracking mode.
Stopped.   Here, during the release in step S1202,
The optional correction calculation will be described with reference to FIG.   In step S1301, RT determines the camera-specific relay.
Time lag time, which is a constant value. LWR (0)
Is the subject movement per unit time filtered
And the defocus amount, so LWR (0) · RT
Indicates the amount of tracking delay during the release time lag.
This value is used as the moving defocus amount SWR of the subject.
You. In the next step S1302, the event counter EVTCNT
The calculated defocus amount SWR is added to the count value of
Is calculated and corrected.   Now, returning to FIG. 12, in step S1204, the in-focus state
Is turned off. Then in step S1205
The raising of the reflection mirror 105 is started, and in the next step S1206
Aperture control is performed via the exposure control circuit 121. S
At step S1207, the raising of the reflection mirror 105 is completed.
If the determination is completed, the process proceeds to step S1208, where the lens
Driving is stopped. Thus, the reflection mirror 105 rises
In the tracking mode until the exposure starts.
Lens driving is continued. Exposure starts in step S1209
Switch S in the next step S1210._FiveIs turned on
It is determined whether the exposure has been completed. Once exposure is complete,
At itch 1211, automatic film winding starts,
Subsequently, in step S1212, the reflection mirror 105 is lowered.
It is. In step S1212, switch S_FourDepending on the condition of
The camera determines whether the shooting mode is
For example, the process returns to step S3 in FIG.
Is made. On the other hand, if it is the single shooting mode, step S121
4, proceed to the loop that repeats only the photometry of S1215, and
Operation or switch S₁Wait for is turned off.   As explained above, filtering
Defocus changes due to subject movement
And defocus due to subject movement
If the change in volume is small, the specific gravity of the past data
Variation due to the increased averaging process
Is absorbed and stable tracking correction becomes possible.
If there is a large change in the amount of
The averaging process with emphasis on
Focus adjustment is possible even for fast-moving subjects
Can do it. [The invention's effect]   In the present invention, it is possible to follow a fast-moving subject.
Focus adjustment and moving the subject
The amount of change in the focus detection value caused by the
By performing different arithmetic processing,
If the subject moves fast, pursue responsiveness,
Pursuit of stability over responsiveness when subject movement is slow
This makes it possible to control according to the situation.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a diagram showing one embodiment of a block configuration of a camera to which the automatic focusing device of the present invention is applied, FIG. 2 is a block diagram showing a control circuit in FIG. Figures 3 to 14
14A is a flowchart showing a control operation in FIG. 1, and FIGS. 14B to 14D are flowcharts showing another embodiment of the filtering process in FIG. 14A. FIG. 15 is a diagram for explaining the calculation of the driving amount of the lens, FIGS. 16, 18, and 19 are diagrams used for easily explaining the operation of the tracking correction, and FIG. FIG. 9 is a diagram used to explain an operation of a filtering process. 101, 102: clutch, 103: lens circuit, 104: reading circuit, 105: reflecting mirror, 107: CCD image sensor, 108: interface circuit, 109: microcomputer, 110: slip mechanism, 111 ... … Drive mechanism,
112: transmission mechanism, 113: encoder, 114: motor driver, 115: display circuit, 121: exposure control circuit, 12
2 ...... photometric circuit, 124 ...... one frame hoisting circuit, 125 ...... exposure setting circuit, MO1, MO2 ...... motor, S ₁ to S ₅ ...... switch.

   ────────────────────────────────────────────────── ─── Continuation of front page        Panel     Referee Takehiko Katayose     Judge Kimio Yoshino     Judge Ichiro Kotani                (56) References JP-A-59-68713 (JP, A)                 JP-A-60-214325 (JP, A)                 JP-A-61-269106 (JP, A)                 JP-A-59-142506 (JP, A)                 Shokai Sho 61-72967 (JP, U)

Claims (1)

(57) [Claims] An automatic focus adjustment device that adjusts the focus of the taking lens based on the detection result of the focus detecting means that performs focus detection on the expected focal plane of the taking lens based on outputs from the light receiving elements by a plurality of pixels. Control means for driving the means to output a focus detection value; and, based on a plurality of focus detection values detected in chronological order by the control means, calculating a change amount of the focus detection value due to movement of the subject. Calculating means for outputting the data in chronological order according to the magnitude of the change amount output in chronological order from the first calculating means. Selecting means for selecting arithmetic processings having different specific gravities; second calculating means for performing the arithmetic processing selected by the selecting means on the amount of change output in time series from the first calculating means; Based on the detection value and the processed change amount calculated by the second calculating means, an automatic focusing apparatus characterized by having a drive means for driving the photographing lens. 2. 2. The automatic focus adjustment device according to claim 1, wherein the calculation process selected by the selection unit includes an average process of a change amount of the focus detection value output in time series. . 3. 3. The automatic focus adjustment device according to claim 2, wherein the averaging process includes a simple averaging process of a change amount of the focus detection value output in time series. 4. 3. The automatic focus adjustment device according to claim 2, wherein the averaging process includes a weighted averaging process of a change amount of the focus detection value output in time series. 5. 2. The averaging process according to claim 2, wherein the averaging process includes an averaging process performed using a processed change amount that has been subjected to the averaging process in the past and a change amount output from the first calculation unit. An automatic focusing device according to the item. 6. The arithmetic processing selected by the selection means includes a first processing in which the preceding data is emphasized in time series and a second processing in which the subsequent data is emphasized in time series.
Wherein the selecting means selects the first processing when the moving speed of the subject is low, and selects the second processing when the moving speed of the subject is high. 2. The automatic focusing device according to claim 1, wherein