JP2770212B2 - Automatic focusing device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [0001] BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to automatic focusing of a camera or the like.
Equipment related, especially when the target object is moving
Occasionally, the imaging lens moves the image plane of the moving object.
Technology for predicting and driving the taking lens.
You. [0002] 2. Description of the Related Art First, FIG.
The configuration of a conventional automatic focus detection device will be described. Stuff
Light emitted from the body a passes through the imaging optical system (photographing lens) L
And usually a film via a quick return mirror M
Guided on finder screen S at a position conjugate with plane F
I will Also, a semi-transparent part in the center of the quick return mirror M
Is transmitted through the sub mirror S. Focusing via M
It is led to the exit means. The focus detection means 101 has a known configuration.
Focus detection optical system, charge storage type image sensor and focus
It consists of a detection calculation unit and an image sensor drive control unit.
Have been. Immediately after the end of charge accumulation, the focus detection
Accordingly, the defocus amount is calculated intermittently. This default
The amount of focus is different from the film conjugate plane which is the predetermined imaging plane.
An amount corresponding to the distance along the optical axis from the image plane of the imaging optical system L
It has become. [0003] The control means 102 is
Receive the data on the defocus amount and
The motor of the moving means 104 is driven and included in the imaging optical system L.
Moving the focusing optics to the desired image plane and the imaging optics
Are controlled so that the image planes of the images coincide with each other. Here is the lens driver
Accurate control of the driving amount of the stage motor by the input signal is possible
If it is not, use a photo interrupter
Monitor means 10 for detecting the amount of movement of the focusing optical system
Using the feedback pulse of No. 3, the control means 102
Controls the driving of the focusing optical system. Of course, pulse mode
The drive amount can be accurately controlled by an input signal like a
The monitoring means detects an input pulse or
May be substituted by equivalent means. [0004] Generally, the entire imaging optical system is moved to automatically focus.
Adjustment is rather rare, and some of the imaging optics
It is common to adjust the focus by moving the focusing optical system of the
It is. In this case, the movement amount of the focusing optical system and the imaging
It does not coincide with the moving amount of the image plane of the optical system. Therefore, actually
Is output from the monitor means 103 to the lens information generation means 105.
The ratio between the number of feedback pulses to be applied and the amount of movement of the image plane
Control unit 102 records the lens information.
The value of this ratio is read from the generating means 105 and the required image plane
The required number of pulses corresponding to the moving amount (defocus amount) of
Calculated feedback pulse is the required number of pulses
Drive will be performed up to this point. [0005] However, this point is not the essence of the present invention.
Relationship, and in the following description, the gist of the present invention will be understood.
The imaging optics is represented by a virtual single lens for ease of use.
In particular, the movement amount of the virtual single lens and the image accompanying the movement amount
The description will be made on the assumption that the movement amounts of the surfaces are equal. Of course
Theory, even in single lens or full lens
If the distance to the body is very close, the above assumption does not hold at all
Therefore, for macro lenses, etc., it corresponds to the amount of lens extension.
The value of the ratio needs to be switched step by step
The function is required for the information generating means 105.
Is to the lens information generating means.
To make the explanation easier to understand,
The description will be made using a target single lens. FIG. 27 shows such a virtual unit for convenience of explanation.
The moving object in which the coordinates are fixed to the lens L
The locus of the imaging plane of the body (solid line P) and the conjugate with the film plane
The trajectory (dotted line Q) of the predetermined image plane is shown on the horizontal axis.
Is the time t, and the vertical axis is the image plane along the optical axis and the virtual unitary.
Indicates the distance to the lens. The coordinates (t_n, X_n)
Charge accumulation start time t of the point detection means 101_nAnd at that time
Position x of predetermined imaging plane_nAnd the coordinates (t_n', X_n')
Is the accumulation end time t_n'And the position of the given image plane at that time
x_n'And the coordinates (t_n ⁰, X_n ⁰) Is the focus detection calculation
End time t_n ⁰And the position x of the predetermined imaging plane at that time_n ⁰The table
I do. FIG. 27 shows the so-called intermittent drive.
This is an illustration of the focus adjustment operation,
Body is a₁', A_Two', A_Three'And follow the movement,
Is a₁, A_Two, A_ThreeIt shows a state of moving. As a result of the first calculation, an image plane a of the object
₁And film side b₁D relating to the difference in distance₁Is time t
₁ ⁰Output from the focus detection means 101 to the control means
Reference numeral 102 denotes the defocus amount D as described above.₁The phase
The drive control of the lens L is performed to kill it.
Also object a₁'Is a_Two′, The time t _Two
The lens to D₁Drive only to stop the lens drive
However, at the midpoint of the next accumulation time, the image plane is already a_Two
And the second calculation result is a_TwoAnd b
_TwoD relating to the difference between the positions of_TwoIs time t_Two ⁰Focus on detection hand
Output from stage 101. Then, the control means 102
Is the defocus amount D as described above._TwoOffset
Control at time t_ThreeStop the lens drive
_ThreeEven if the predetermined image plane is brought to the position of_Two′
Is a_Three′, It is not focused. The same is repeated hereafter as shown in FIG.
It is. If the object did not move, and errors
If there was no, in one cycle of accumulation, operation, drive
The object image plane and the specified image plane should have been able to match.
Yes, the calculation result contains 10 to 20% error in the defocus amount
The object image plane in two or three cycles
The imaging planes could have been approximately matched. [0009] However, as shown in FIG.
First two or three cycles if moving
Then gradually the object image plane a_nAnd the predetermined image plane b_nIs narrow
After that, the distance between the two depends on the moving speed of the object image plane and the
Focusing maintained at a constant value determined by the responsiveness with the dynamic focus adjustment device
The follow-up state continues without achieving. [0010] SUMMARY OF THE INVENTION The following state is solved.
In order to achieve this, the Applicant has
An adjusting device is proposed in JP-A-60-214325.
There, the image plane position at a predetermined time is obtained, and
The target position is calculated by adding the surface movement speed multiplied by the time term.
Out, and drive the lens toward this tracking locus.
After reaching the trace, the trajectory follows the image position every moment
It is driven along. [0011] However, here, there is no control during the mirror up operation.
There is no description about you. Therefore, the present invention involves a mirror-up operation.
A self-tracking operation that takes into account the drive suspension period
It is an object to provide a moving focus detection device. [0013] SUMMARY OF THE INVENTION The present invention provides an automatic focusing apparatus.
The light is used to stop the imaging optical system L during exposure and perform the exposure.
This is an automatic focusing device used for scientific
A predetermined image forming plane of the image forming optical system for forming a light image of
Detects the amount of defocus with the specified imaging plane
Focus detection means 101 and an output from the focus detection means
In addition, the movement locus of the imaging plane accompanying the movement of the target object
And outputs the predicted movement amount (step 34 in FIG. 20)
Prediction means 100; focusing of said moving target object
Was obtained by the prediction means so as to be obtained at the time of the exposure.
The correction amount (step 37 in FIG. 20) to be added to the predicted movement amount is
Correction means 100 for calculating, wherein the target object moves
If it is determined that the mirror is
Also in step 43), the correction amount obtained by the correction means
A predicted trajectory determined from the predicted movement amount (trajectory R in FIG. 18)
Driving the imaging optical system with the goal of
You. The structure of the present invention is mainly based on the sixth embodiment and FIG.
18. [0015] [Function] Immediately before film exposure, lens drive
To move the target object
The specified shift amount is related to the amount of movement of the
Decide. With this configuration, the image plane locus is shifted by the predetermined amount.
Based on the predicted trajectory (FIG. 18; broken line R)
Controlling the focus driving means so that the predetermined image plane is on the predicted locus.
Processing related to mirror up and lens drive stop at one time
Is controlled during the film exposure.
When the specified image plane (film conjugate plane) is matched to the surface locus
So you can shoot in focus. That is, the target object
Since the imaging optical system is driven prior to the movement,
Stop the lens drive after the release signal is
Even if the camera enters the
The image plane of the udon can be matched to the film plane,
It is possible to shoot in a way that suits your needs. [0016] [Embodiment] A first embodiment will be described. This example is cumulative
・ Compensation for tracking without any lens drive during calculation
Driving is performed all at once after the computation is completed and before the accumulation is started.
This is a drive system that completely follows intermittent drive. In addition
Correction drive for canceling body movement is called tracking drive.
And FIG. 1 is a block diagram showing the configuration of the first embodiment.
Except for the presence correction means 100, the description was used for the description of the conventional example.
It is the same as FIG. The correcting means 100 outputs the signal from the focus detecting means 101.
Receiving the force and the output of the monitoring means 103,
No object is determined.
Calculate the correction drive amount for tailing, and control based on this.
The control means drives the lens to track. Figure 2 shows the first implementation
This shows the tracking of the example.
Used. Here, Q is the same as the drive by normal intermittent drive.
And trails the trajectory P of the object image. Q '
FIG. 9 shows a state of tracking drive according to the first embodiment, and shows a trajectory.
Tracking is performed along P, and the state near focus is always maintained.
You can see that you have. The difference between Q 'and Q
This corresponds to the object movement correction amount calculated by the means. Next, the processing contents of the correcting means 100 will be described.
State. Since the lens is not moved during the accumulation / calculation time,
x₁= X₁'= X₁ ^oAnd the time (t₁+ T₁') / 2
Defocus amount at point a₁And point b₁Distance difference (X₁−
x ₁ ^o) At the end of the calculation t₁ ^oFocus detection means 10
If there is no detection error, the defocus amount calculated from 1 is
This value D₁be equivalent to. It is calculated by the focus detection means 100 in the future.
D is the nth defocus amount_nShall be represented by
You. The control means 102 operates at time t₁ ^oFocus on
D calculated by the means 101₁Lens driving means 1 using
04 and the feedback pulse of the monitoring means 103.
The amount of image plane movement by the lens is defocused while counting
Amount D₁Drive is performed until it becomes equal to As mentioned before
In fact, a certain amount of image plane movement ΔB
The amount of lens movement corresponding to f differs for each taking lens.
Feedback pulse that gives the amount of movement of the lens
Δn often differs depending on the photographing lens. So the image plane
Move ΔBf to Δn = K_B* Feed due to the relationship of ΔBf
Conversion coefficient K to convert to back pulse number Δn_BThe lens information
The information is stored in the report generation means 105 and is used for actual driving control.
I will do it. Conversion of image plane movement to the number of feedback pulses
This operation has nothing to do with the essence of the present invention,
And all the following descriptions are converted to image plane movement.
It will be expressed in quantity. Immediately, defocus amount D_nOf course
The correction drive amount C described later_nEverything is a measure of the amount of image plane movement
It is assumed to be described in degrees. Now, the operation end time t
₁ ^oFrom the accumulation start time t_TwoDrive as high as possible
It is also desirable from the viewpoint of responsiveness that the speed is high.
It is necessary to gradually reduce the speed at the edge, so it is fast and
Non-constant speed drive. In the tracking method of the earlier application,
Charge accumulation was performed at the same time.
Because the degree of detection is fast, the detected image is equivalently blurred,
Slight detection of each time of accumulation start, speed change, accumulation end
Error increases the position determination error and
The non-constant speed, etc.
Disadvantages such as the inability to accurately correct the defocus amount.
Was. Therefore, in this embodiment, the high-speed and non-uniform driving
Is almost converged t_TwoTo resume the accumulation. Time t_Two ^o
Second defocus amount D _TwoIs calculated, which is
Accumulation time t_Two~ T_TwoIs the amount of defocus in the middle of '
Point a_TwoAnd point b_Two(X_Two-X_Two ^o)
Equal except for differences. If the object is not moving and detection error
D if the difference is small enough_TwoIs D₁Very small value compared to
In principle, the transfer of the object from the value of this ratio
The presence or absence of the movement can be determined. Actually, the conventional example shown in FIG.
In the following state like the trajectory of Q at n ≧ 3 shown in
Is the calculated defocus amount D for each time_nIs forever
It does not converge to zero and becomes almost constant, and D_n/ D_n-1$ 1
You can see that However, as shown by Q 'in FIG.
Additional correction drive in addition to drive based on defocus amount
Is calculated, the calculated defocus amount D_nSelf
It is impossible to perform the above-mentioned identification using a body. Therefore, as shown in FIG.
Is provided with an insufficient convergence amount calculating means 100a.
P given by the formula as a quantity representing shortage_nCalculate the amount
And take this as a source.         P_n= D_n+ [Previous driving amount] -D_n-1              ...... Here, the expression is the n-th defocus amount D_nWas calculated
Time t_n ^oAt present, D_nIs to the focus detection means 101
The latest defocus amount calculated from D_n-1Is one time ago
Is the defocus amount. [Last drive amount] is time t^o
_n-1~ T_nIf the value X (n-1) actually driven during
Time t^o _n-1X (n−)
Calculated value from which only 1) drive is performed
D_n-1'. Of course, both are equal when there is no error
(D_n'= X (n)). The correction drive amount for tracking is C_nToss
Then, the driving amount D_n′ Becomes D_n'= D_n+ C_n          ............ Using this, the equation can also be expressed as: P_n= D_n+ C_n-1            ……… ’ As can be seen from FIG. 2 according to the definition of the above equation, n ≧
After 3 the convergence shortage amount P_nIs the point a_nAnd point a_n-1
(For example, P_Four= D_Four+ C_Three) Immediately of the object image
It is nothing but a driving amount. The time t when the first calculation result is obtained₁
^oSince there is no previous result at the time of
P₀Is a sufficiently large value, for example, about 1000.₁=
D₁Decide to be So in this case P₁Is the object transfer
It does not correspond to the momentum and P₁/ P ₀  <1. Therefore
In the correcting means 100, the object movement is determined as shown in FIG.
A means 100b is provided to determine whether or not the object has moved. Next, a specific example of the object movement determining means will be described.
Bell. Assuming that there is some detection error,
The tracking drive of the moving object is performed according to the routine shown in FIG.
Insufficient convergence amount P obtained as a result_nTable 1 shows the changes in
You. [0025] [Table 1] In this example, the first insufficient convergence amount P₁Is 10
And n = 2, 3 and P_nQuickly converges but focuses
It doesn't reach to below 05 to 15 which corresponds to the range
And n = 4-7 for P_nIs about 4
It is constant. That is, even if n = 3-4 or later, P_nIf the value of
An object that does not converge within the allowable focus range and becomes almost constant
The story of the movement, P at that time_nValue in one cycle
This corresponds to image plane movement accompanying object movement. From this, P_nIs 5 ≦ P_n/ P_n-1<2
It is judged that the object is moving by entering the range
Can be specified. If the object is stationary,
Defocus amount is 20% 30% of previous defocus amount
In general, Practically object movement discrimination
The means is P_n/ P_n-1And the predetermined constant (threshold) k
It is determined whether the object is moving by comparison. Various errors
Given the effect of, the range of practical values for k is 3 ≦ k ≦ 8 It is considered that 4 ≦ k ≦ 6 is optimal. And P_n
/ P_n-1When ≧ k, the object movement determining means 100b
It is determined that there is. Also in this one cycle
The image plane movement corresponding to the object movement is almost P_nThis is given by
I understand. Therefore, the object movement compensation is added to the compensation means 100.
Positive amount calculating means 100c is provided so that compensation for tracking can be performed.
Positive drive amount C_nIs calculated. That is, the object movement determining means
When it is determined that there is an object movement C_n= P_n When it is determined that there is no object movement C_n= 0 And FIG. 4 conceptually shows a processing flow.
The flowchart of (a) and the processing contents of the first embodiment are specifically described.
Explanation will be made using the flowchart of FIG.
I do. First, in step (1), the initial value setting described above is performed.
U. Start accumulation in step (2), accumulate in step (3)
This is the end, and the charge accumulation type image sensor of the focus detection means
Image output is sent to the focus detection calculation section in the focus detection means.
It is. Next, the focus detection calculation is started in step (4).
The calculation is completed in step (5), and the defocus amount D
_nIs calculated. In this way, the focus detection means
Focus amount D_nIs normally output, the control means 102
Drives the lens based on this data. However
In this embodiment, the defocus amount D_nIs the first correction means
100 receives processing for tracking drive. Step (6) is a step in the correction means 100.
This corresponds to the bundle shortage amount calculating means 100a, and
Foot weight P_nIs calculated. Step (7) is an object movement discriminating hand.
FIG. 4 (b) corresponds to the step 100b.
In (7), P_n  > K * P_{n- 1}  If the object moves
It is determined that it is moving. Step (8) is the object transfer
This corresponds to the motion correction amount calculation means 100c, and
Correction amount C according to the presence or absence of_nIs calculated. Then step
Driving amount D at (9)_n'Is calculated. This D_n′ Before
As described above, the image plane movement amount is converted, and the control means
102 is this D_nAnd the fee from the monitor means 103
As described above, the correspondence between the feedback pulses
Conversion coefficient K stored in generating means 105_BConnect with
Drive control is performed. Next, in step (10b), the next operation is performed.
Store the required values and start driving in step (11b)
Until the driving stop condition in step (12b) is satisfied.
Driving continues. Step again if stop condition is met
Returning to (2), accumulation is resumed. Thus, the first real
In the embodiment, charge accumulation, calculation, and driving are duplicated even when tracking is performed.
Intermittent drive system that is performed sequentially without performing
I have. Therefore, in step (9), D_n'= D_n+ C_nNotation
As described above, the driving amount D_nIs the defocus amount D
_nAnd correction drive amount C for tracking_nGiven as the sum of
ing. If the drive in steps (11b) and (12b)
The drive amount calculated in step (9b) by executing the motion
There is no problem if the goal is fully achieved, but the calculated drive
When there is a difference between the moving amount and the actual driving amount, step (6)
Insufficiency of convergence P in b)_nIs incomplete in the formula
It is necessary to use an expression. And in that case the expression
The value of the previous drive amount is D_n-1′, Not the last
The driving amount X (n-1) at the time of
It is necessary to use the value calculated from the accumulation. As described above, according to the first embodiment, the completely intermittent driving
Movement, so the lens stops during the accumulation time.
Calculation as in the case of JP-A-60-214325.
The accuracy of the output defocus amount does not deteriorate,
Presence / absence of object movement can be determined accurately, enabling accurate tracking drive
It has the effect of becoming effective. It is also a conventional tracking method
In the case of Japanese Patent Application Laid-Open No. 60-214325, the performance is performed even during the accumulation time.
It is assumed that the driving is performed in parallel during the calculation time.
Therefore, the microcomputer becomes multitask
The number of counters and timers is limited, and
Using a computer makes programming difficult or quick.
There was a disadvantage that processing could not be performed. However, in the tracking software of this embodiment, the intermittent drive is used.
Because it is dynamic, it does not become multitasking, and it is absolutely time
Tracking drive can be performed effectively without monitoring
Program construction is easy, and at the same time
The ability of the microcomputer running the software is sufficient to respond
There is an advantage that. Also, the amount of software actually added
4 (a) and 4 (b) show conventional intermittent drive software.
Tracking drive by adding a few calculation steps
Is very compatible with existing software.
You. Next, the object movement discriminating means described in the first embodiment.
In the second embodiment in which the processing of the stage 100b is further refined
explain about. In addition, supplementary information when the object is determined to be moving
Positive drive amount C_nMust be strictly set as described above.
It is not necessary, for example,_nA coefficient close to 1
Overrun in lens drive control.
In the case of only one, the coefficient was set to 1 or less (0.9, 0.8, ...)
In the case of underrun, set the coefficient to 1 or more (1.1, 1.
2,…). [Second Embodiment] As described in the first embodiment, object movement
Tracking correction when the discriminating means determines that there is object movement
Intermittent drive (tracking drive) including
If it is determined that normal intermittent drive (convergence drive) is
U. However, if the judgment accuracy of the
Erroneously determines that there is object movement when there is no object movement
If you do, it is said that the driving operation tends to be hunting
Problems arise. In order to solve this problem, incorrect judgment
A method for preventing the above will be described with reference to FIG. Fig. 5
The flow chart replaces step (7) described in the first embodiment.
Here, in step (1001), the current convergence
Shortage P_nAnd previous convergence deficiency P _n-1Is the same sign
Whether or not the object is moving is determined by checking whether
If not, it is determined that the object has not moved,
In the case of a sign, the next decision step (1002) will certainly
It is determined whether the object is moving. Immediate step
Whether the object moves in two stages (1001) and (1002)
Is determined by P_n  > K * P_n-1    Is satisfied
P_nAnd P_n-1The object is moving when
This is in order to avoid being determined to be. Step (1)
001) and (1002)_n/ P_n-1
> K may be used. Next, the object suddenly comes out of the focus detection visual field.
Let's consider the case. In this case, step (100
When the condition 2) is satisfied (for example, P_n/ P
_n-1≧ 4), P_nDefocus amount based on
Driving, causing a significant overrun.
There is a problem of producing. Solve this problem
Is provided in step (1003) P_n  <
r * P_n-1  R is determined to be about 2 ≦ r ≦ 3
Is a number. Immediately P_n  Is P_n-1  Remarkably compared to
If it becomes larger, that is, if step (1003) is not
If the object moves outside the detection field of view,
Normal without tracking correction to prevent erroneous determination of presence
Intermittent drive (convergence drive). Next, step (1004) will be described.
You. The purpose is that there is no or little object movement near the focus
The convergence deficiency P_nDefocus included
The relative proportion of the volume detection error increases,
Regardless, the determination in step (1002) is affected by the detection error.
Eliminates the possibility of
Make sure that the movement does not become hunting when it converges
It is for. To achieve this purpose, the allowable width δ_a
P_n  > Δ_aIf not, it is determined that there is no object movement.
Set. Where δ_aIs the defocus amount calculation error
And a value that reflects the depth of field
The value is generally about 05 to 2. That is,
Lens is near the in-focus area, and
In the range that can be regarded as P_nIs in tracking drive mode.
Only the convergence drive is performed. This step allows
When it is small, it will not be possible to detect movement, but in this case
Intermittent drive without normal tracking (convergence drive)
There is no problem because there is no follow-up. Also this stage
The tip (1004) is not limited to this position,
It may be before the step (1001). As described above, according to the second embodiment, the object movement
Is more complete, so with the addition of tracking software
No hunting problem during normal operation
Operation is guaranteed. Also in the second embodiment, the first embodiment
As in the example, time measurement is not performed, so a very simple software
This can be dealt with by the software processing. [Third Embodiment] The focus detection means and the lens driver
Discussed timing determined only by stage operating characteristics
However, when considering the camera,
You need to take care. As described in the first and second embodiments.
The drive is an intermittent drive, and the drive itself has a step shape. this
Since the movement of the object changes to Nameraka, the exposure
It is preferable to make the timing of the above appropriate. Reverse
In other words, because the tracking drive is intermittent,
The timing of the exposure,
If the photograph was taken, it would be perfectly in focus
Will be obtained. Timing of such exposure
The method of taking the image will be described in the following third embodiment. What
The tracking method in the third embodiment is the same as that in the first embodiment,
It is assumed that the embodiment is used. Also regarding shutter release
Exposure is independent of the focus detection system
Independent mode that allows (mirror up) and focus determination
Focus priority mode that allows exposure if there is
However, here, we will focus on focusing priority mode. Prior to the detailed description of the third embodiment, a known
Exposure timing method in normal intermittent drive method
Will be described with reference to FIG. Def in step (5)
When the amount of focus is calculated, it is calculated in step (81).
Focusing with a predetermined defocus amount
Compared to the range δ, the magnitude of the defocus amount is larger than δ
In step (82), the lens is driven.
In (83), it is determined that the driving of the predetermined driving amount has been completed.
Then, the next charge accumulation starts. This cycle is 1-3
The process is repeated, and the differential
If you find that the size of the focus amount is smaller than δ,
Goes to step (84) to perform focusing processing. Here
Focusing is the process of turning on the focus indicator and exposing the film.
Means to allow the mirror to move up.
Even if the shutter was released before,
Exposure is not performed until the point of
Then, mirror up and exposure are performed. This is
This is the outline of the focusing priority mode. Normal with no object movement
In this case, there is no particular problem in this processing. But the object moves
If you are out of focus in the above normal processing
However, there is a drawback in that the exposure is performed. this
This will be described with reference to FIG. In the example of FIG. 7, the midpoint of the charge accumulation time Tint is
Since the trajectory P of the image has passed, the calculation end time t_n ^oCalculated
Defocus amount D_nIs D_n  <Δ, figure
The routine enters a focusing process of six steps (84). if
t_n ^oIf shutter release has been performed before, t
_n ^oAt step (84), the exposure is permitted (mirror
Up permission) occurs, the mirror is raised and Mira
Time t after the delay time Tup_expoFill in
Exposure occurs. But at this point it is clear from the figure
Again, since the locus P of the image is far away,
This will result in an out-of-focus photo. That is, D_n  <Δ
Means focusing at the midpoint of the charge accumulation time.
Only the delay time from this moment to the exposure
The image plane will be off. In order to solve this disadvantage, in the third embodiment, FIG.
8 means for performing a focusing process in the correcting means 100
100d, where the tracking drive is performed.
Focusing processing that is most suitable for this is performed. This operation is shown in FIG.
Step (10). (Step (1) ~
(9) is the same as FIG. 4 (a)) Next, this step (10)
Will be described with reference to the flowcharts of FIGS.
You. In step (1101) in FIG.
Whether there is a case or not is determined based on the result of step (7).
If there is no object movement, steps (81) and (8)
2) Go to (83) or (84). This is the same as in FIG.
Descriptions are omitted because they correspond to the numbers. Steps
If it is determined in step (1101) that there is an object movement, step
Driving is started in (1102), and step (1103) is started.
When the predetermined drive stop condition is satisfied in step (110)
Proceeding to 4), the focusing process is performed. Step (11
The content of the focusing process of step 04) is the focusing process of step (84).
Exposure permission and display lighting are the same as the
In the case of, since the object is moving, the display disappears after the predetermined time has elapsed.
It is good to light. In this way, when there is an object movement,
FIG. 7 explains the reason why a photograph in focus can be taken.
You. In the tracking drive method in the first and second embodiments,
According to the above description, during tracking, the instant about the midpoint of the charge accumulation time Tint
Are controlled so that the trajectories of P and Q 'substantially intersect with each other.
ing. From this, if the cycle time is the same,
If the intersection of P and Q 'occurs next time from the end of driving, Tin
It is expected to be later by t / 2. Meanwhile mirror up
The delay from the start to the exposure is Tup, and the time difference between the two is δ
Assuming T, δT = Tup−Tint / 2. Therefore drive
Ended and decided to instruct to resume storage
If the mirror is raised by δT ahead of
At the moment of light, you can almost be near the intersection of P and Q '.
You. By the way, the camera body of Tup ≒ 50msec
In the case of, Tint / 2 changes according to the brightness of the object,
0 to 50 msec in most cases except when it is always dark
0 ≦ δT ≦ 50 msec.
Mirrors earlier than the timing at which accumulation is scheduled to resume for m sec
You can do it up. How to take the timing
As a first example, apply brakes to stop driving the lens.
It takes about 20 msec from the start to the complete stop.
If the image plane moves about 50μ during
The mirror-up timing to the
20 to 30 msec after this,
It can be set to take a picture. timing
As a second example of how to take, the remaining drive amount is converted to image plane movement.
20-30msec to change from 150μ to 50μ
If possible, tie the mirror up when 150μ remains.
The brakes and restart the accumulation when the remaining 50μ
The purpose can be achieved even if the timing is given. FIG. 11 shows the second example.
Will be described. If there is an object movement, step (120)
Lens drive starts from 1) to step (1202)
Is done. The remaining drive amount is δ_PTo δ₀Nana
Δ such that the time until_P, Δ₀The value of
It shall be determined. Drive remaining in step (1203)
The amount is δ with respect to the image plane movement_PIf it is less than the step (12
Go to 04) to perform focusing processing including mirror up permission
U. Of course, the shutter release was done by this point
If not, mirror up even if mirror up permission is issued
Is not executed. Further driving proceeds to step (120).
If the driving stop condition of 5) is satisfied, that is, it corresponds to the remaining driving amount.
Image plane movement is δ₀Step (1206) if
The brake is applied at step (120)
7) Confirm that the main mirror is in mirror down state
When the mirror is down, the operation proceeds to step (2) in FIG.
It returns and accumulation is resumed. The shutter button is
It is in the release state and the mirror mirror in step (1204)
The mirror is actually up
In step (1207), it is determined that the mirror is not in the down state.
Exposure ends and returns to mirror down state
Stop at step (1207) until mirror down
Then, the process proceeds to step (1204) in FIG.
Is done. By the way, the value in step (1203)
δ_PMay be set to a constant value, but the charge accumulation time
To calculate δT according to the value, and according to the value of δT,
δ_PChanging the value of is perfect. Above
Although the case where δT> 0 has been described, depending on the condition, δT
There may be a case where <0 is satisfied. But this place
If the mirror-up timing is the drive end timing
Since it is a direction to delay more, δ from the drive end timing
It is easy to count and delay only T time
You. In the above description, the error of δT is compensated.
Although the method has been strictly described, when δT is 20 to 30 sec.
In most cases, focus processing (mirror
Up permission) and the timing of the storage restart instruction
A practically satisfactory result can be obtained without such a problem. Then
When there is no object movement in step (1201)
The case will be described. In this case, in step (1210)
  D_n′ And δ_fAre compared. For reference this
Since there is no tracking in the loop, D_n'= D_nbe equivalent to. here
At δ_fRepresents the width of one side of the focusing zone, and 50 to 200 μ
It is of the order of magnitude. If D_n'> Δf
In step (1211), the identification flag is set to 0, and the
Driving starts at step (1212). Then step
Step (1213) that satisfies the drive stop condition
In step 14), the brake is applied, and step (1207) is performed.
After that, the accumulation is restarted in step (2). in this way
And return to step (1210) again._n′ <Δ
_fIf it becomes, the process proceeds to step (1215). Stay
In the step (1215), display lighting, mirror up permission etc.
Perform focusing processing. Next, in step (1216), the identification
Add 1 to the lag and remember that you have entered the focusing zone. S
In Step (1217), D_n′ <Δ_c(0 <δ_c<
δ_f) Or ≠ 0, and this condition
When the condition is satisfied, the process proceeds to the next accumulation without driving. Where δ_cIs a value of about 0 to 50 μ,
Provided to stop the lens near the center of the focusing zone
It is a shiki value. Condition is satisfied in step (1217)
If not, ie = 0 and δ_f> D_n′> Δ_c In the case of, driving is started in step (1212), and driving is started.
Stop condition Drive remaining amount <δ_cIs driven until
The brake is applied at step (1214). The essential points of the third embodiment are briefly summarized.
Can be expressed as follows. In other words, the focusing sequential
At least a predetermined value (50 μ to 150)
over about μ), when there is no object movement
Focus processing is not performed if the object is moving.
In the case that the tracking drive has been terminated,
The focusing process is performed sequentially. Thus, in the third embodiment, there is tracking drive.
Focusing processing, that is, lighting of focus indication and mirror
-By changing the timing of the
Exposure at the most effective moment even when there is tracking action
Tracking by intermittent drive.
Focus on a moving object
It is possible to take a photograph that has been taken. In the third embodiment, the time measurement is not necessarily performed.
It is not necessary, and even if it is used, it is the value of the charge accumulation time,
This value is accumulated even in a normal focus detection device without tracking.
Indispensable when controlling time with software
This is often measured, and it can be used
Therefore, in that case, the burden on software will not increase especially.
And it is still intermittent drive.
Multitasking as in the case of 0-214325
However, it is easy to create software as described in the first embodiment.
It has not lost its sexual character. [Fourth Embodiment] In the above third embodiment, the timing of exposure
I explained how to take it. However, the above description applies.
Only the timing of the first exposure,
The timing of the second and subsequent exposures will be different
You need to be careful. That is, before the first exposure
(Accumulation / operation / drive) as one cycle
This is a return operation. Therefore, based on this assumption,
I should have taken the timing. And of the first exposure
The timing between the moment and the moment giving the midpoint of the scheduled storage time
I should have considered the match. However, after the second time in the continuous shooting mode,
(Accumulation / calculation / drive / mirror-up period)
It is necessary to consider the cycle and the repetitive operation,
Moment is in the middle of the mirror-up period,
It is impossible to match the middle of the time with the moment of exposure
is there. The tracking method in such a case is described below.
A description will be given as a fourth embodiment. In this case, the basic tracking
The work is based on the first and second embodiments. The upper part of FIG.
The figure in the right shows the operation of each operation during continuous shooting with the motor drive.
This is an illustration of the imming. The cycle of continuous shooting is T,
The period when the main mirror is up is T _MIn
You. Mirror up period T_MIs the delay time associated with the rise of the mirror
Tup, exposure time Ts, and delay time Tdown due to mirror lowering
Consists of Also, a period T indicated by a broken line in the figure_WDew
This is a period corresponding to film winding after light. Again
The time from the end or the end of calculation to the start of mirror up is T
_DThe lens is driven during this time. Open mirror up
At the start timing, the remaining drive amount is
A constant amount (for example, 50μ or 150μ)
Or). Now, in the case of the configuration shown in FIG.
Focus detection cannot be performed when the
At the same time, accumulation starts, and as shown in FIG.
int, defocus amount when calculation time Tcal passes
Turns out. Lens drive is prohibited during film winding
If the defocus amount is calculated, the winding ends.
Until the start, the driving of the lens is delayed. The first exposure in such an operation mode
Under continuous shooting conditions after the end of light, exposure timing
ΔT between the timing of the₁=
There is a time lag of Tdown + Tint / 2. by the way
As described above, the tracking drive method described in the first embodiment is used.
According to FIG. 12, P and Q ′ are different from the accumulation time as shown in FIG.
Correction drive for tracking is performed so that
Will be. However, the drive is intermittent in this state,
In addition, at the timing of exposure, δT₁Object image during
A photo that is out of focus by the amount Δ that the surface has moved
become. In the fourth embodiment, the timing of exposure is
ΔT between the timing and the midpoint of the accumulation time₁Based on
And consider this as the amount of convergence excess or deficiency.
As shown in FIG.
It is provided in the correction means 100 and can be used for continuous shooting of moving objects.
I will shoot a focused photo while tracking this
It is something to do. That is, the correction drive amount for tracking is Δ
As shown in FIG. 13, the exposure time
The tracking drive is controlled so that P and Q 'overlap during imaging.
Control. Next, a method of calculating Δ will be described. FIG.
As you can see, Δ = P_n* ΔT₁/ T
Cycle time T and accumulation time Tint
Can be determined accurately. But δT₁Neither T
Since the range of possible values is usually limited, α (= δT
₁/ T) with an appropriate constant Δ = P_n* Even if calculated by α
No significant errors occur. In this case δT₁And count T
There is a merit of not having to do it. The value of the constant α is in the range of 1 ≦ α ≦ 5.
Yes It is often about 2. Also, actually figure
The locus of the object image plane described in 13P is not a straight line,
When approaching, the inclination increases with time,
When moving away, the slope changes in a direction that decreases with time
So, P_nThe direction of movement of the object is determined by the sign of
The value of α should be changed accordingly. That is, the object moves away
Let α be α_f, Approaching α to α_nThen α_f
> Α_nLike In this way, the object moves away
When the value of Δ at the time becomes large, the correction drive amount for tracking decreases
This can be matched with the decrease in the inclination of the object trajectory. Next, a specific correction procedure is shown in FIGS.
A more detailed description will be given. FIG. 15 is a flow chart showing a schematic flow.
Between step (8) and step (9) in FIG.
Calculate the convergence excess / deficiency amount Δ as the bundle excess / deficiency amount correcting means 100e.
A step (8) 'for performing the output correction is included. FIG.
The part from step (6) to step (10)
The minutes are shown more specifically. Steps (6), (7) and (8) are the first and second steps.
This is as described in detail in the embodiment. Step (81) (8)
2) (83) corresponds to step (8) ′ in FIG.
Yes, first in step (81), continuous shooting and
It is checked whether the first exposure has been completed. another
By using the expression, the final lens driving operation and the accumulated
Whether there was a mirror up / down during the arithmetic cycle
Can be examined. That is, the first exposure is based on the third embodiment.
Correction of the convergence excess / deficiency amount Δ
Is not performed, so that the process immediately proceeds to step (9). During continuous shooting
In step (82), the convergence excess / deficiency amount Δ is calculated.
In step (83), the tracking correction drive amount C_nTo the above Δ
Is the corrected drive amount C further corrected_nIs calculated,
Move to the next step (9). About the following steps
Are omitted because they are equivalent to those already described. As described above, according to the fourth embodiment, the exposure
Equivalent to the time difference between the moment and the time giving the midpoint of the accumulation time
Since the movement of the object image plane can be corrected,
Even in-focus photos are taken of moving objects
You. Also in this embodiment, basically, the intermittent drive is used.
The CPU uses multi-processing such as parallel processing of accumulation operation and driving.
There is almost no need to perform tasks, and software construction is easy.
is there. As described in the third and fourth embodiments,
At the time of the first exposure and the second
The optimum control conditions are different from the second and subsequent exposures. Follow
It is necessary for the correction means to identify this point and perform optimal control.
You. [Fifth Embodiment] In the third and fourth embodiments, focusing is performed.
Priority mode was assumed. That is, shutter release
The exposure permission signal is output from the focus detection system.
Mirror up was not performed until it was done. That is, tracking
Mirror up at or near the end of driving
A permission signal is output, and the mirror
Had been However, this does not necessarily mean that the intended instant
There is a drawback that a picture cannot be taken in between. So a little pin
The shutter release button is pressed even if the
Independence that mirror is raised in the middle and exposure is performed
Modes are also conceivable. In the fifth embodiment, such an independent mode is used.
Enables photography that is as in-focus as possible
The conditions for setting are described. In the motor drive of a single-lens reflex camera
May be about 5 frames / second at the highest speed. So
The timing in such a case will be described with reference to the upper diagram of FIG.
You. The cycle time T for continuous shooting is 200 msec.
Of the exposure when the mirror up time Tup is 50 msec
Film starting at the end of exposure, with a time of 20 msec
Hoisting time T_wIs 100 msec, the remaining time T_DIs
It is only about 30 msec. Time from braking to stopping
If we expect 10 to 20 msec, the net driving time is
10 to 20 msec, practically impossible to drive the lens
Will be. Therefore, the motor drive as fast as possible
In order to be able to match the focusing operation to Eve,
During the period Tup from the raising of the mirror to just before the exposure,
It is better to allow driving. If Tup ≒ 50m
If there is sec, the time available for driving the lens is T_D+ T
up ≒ About 80m sec, so most of the time tracking
The required driving amount can be covered. Obedience
In order to make the response as fast as possible,
It is important to perform the pitch driving. Now, in the focus priority mode, the focus detection system
The fact that the stem and the timing of exposure have a predetermined relationship
As is clear from the assumption, the state of the focus detection system
Regardless of whether the shutter release button is pressed regardless of
In fact, even in the independent mode where exposure is
Between the timing of system operation and the timing of exposure.
A certain relationship occurs. This allows exposure in independent mode.
Control the focus state to be almost good at the moment of light
It is possible. That is, the above-mentioned certain relationship is
That the charge accumulation starts as soon as the
This is inevitably caused by setting a condition. like this
, The same as described in the fourth embodiment.
Parameter δT₁Has meaning. That is,
Between the last lens drive operation and the current accumulation and calculation cycle
When the mirror is up and down,
The time difference until the middle of this accumulation time is δT₁Becomes
Considering this way, the time for driving the lens is greatly within Tup
Except for the interruption, the fifth embodiment is the same as the fourth embodiment.
It can be seen that the same processing as described above is effective.
Then, the next time for driving the lens is greatly interrupted in Tup
The following describes the effect of this. First, during the time during which the lens can be driven between frames,
ΔZ is the amount of image plane movement that can be₁Then
When the moving speed of the object image plane is 5 frames / sec,
× ΔZ₁Tracking of moving objects equivalent to / sec
In order to improve this ability, you can satisfy the above relationship
It is necessary to increase the driving power. Figure 17 Q 'eight
An example of the case where the condition corresponds to the critical condition like
The brakes were applied and stopped just before the exposure.
You. When the critical condition is slightly exceeded, like Q ''
During exposure, the lens may move.
In such a case, the image plane moves very fast in the first place,
There is no point in stopping the lens strictly,
There is no problem. Further beyond the above critical conditions
When the object movement is large, the tracking
I can't help. As described above, according to the fifth embodiment, the mirror
By allowing the lens drive even after the up, independent mode,
Focused shooting possible even with high-speed motor drive
Ability. In this case, too,
Step 100e operates in the same manner as described in the fourth embodiment.
Is determined in order to correct the convergence excess / deficiency. Of course
Also in the focusing priority mode of the fourth embodiment, δ_PLarge value of
Actively drive the lens during the Tup period
It is also possible to increase the responsiveness. [Sixth Embodiment] In the above embodiments, the time measurement is not necessarily performed.
Not required and can be replaced with representative values when needed
The efficient tracking drive method has been described. Also accumulation time and calculation
I was supposed to be almost equal every time. When actually tracking
Follows the same object, so the above condition is almost satisfied.
You. Also, according to the simulation, there is little variation
Overrun or underunder each time
Although it is a bit difficult, overall it is compared to normal intermittent drive.
It has been found that a sufficiently effective tracking drive is performed. However, in consideration of the variation of the time interval,
If tail software is built, the slight overrun and
Underruns can also be removed, and
In the embodiment, such a case will be described. Until now
In the embodiment, the object movement correction amount P during one cycle_nThe
Although the calculation was used as it is, in this embodiment, as shown in FIG.
In particular, the time T (n-1) during this time is measured to determine the speed of the object movement.
The degree is calculated and the correction shown in FIG.
Undercorrection amount sequential correction means 100f included in the means 100
The tracking drive is finely corrected by using
Drive is not performed during accumulation operation, and it is essentially intermittent drive.
No change. The processing flow will be described with reference to the flowchart of FIG.
This will be explained. Steps (1) to (5) are the same as before
It is like. In the next step (55), the event counter
The drive amount X from the value of Event or the value of the register Regis
(N-1) is calculated. Here, the event counter
Feedback pulse on previous drive from monitoring means
Contains the counting results of In step (6), the convergence insufficient amount calculating means 10
0a is the convergence shortage amount P_nIs given by P_n= D_n+ X (n
-1) -D_n-1Is calculated by Object movement in step (1)
The determination means 100b determines whether or not the object has moved.
The specific method is as described in the first and second embodiments. Stuff
If there is a body movement, proceed to step (31) and flag IDO
= 1. Steps (32) and (33) correspond to this embodiment.
Corresponds to the content of the object movement correction calculating means 100c in
Things. Step (32) is a diagram at the end of the calculation.
The amount corresponding to the difference between 18P and Q 'is used as a correction amount according to the following equation.
Calculated. C_n= P_n* {Tint (n) / 2 + Tcalc (n)} / T (n-1) ... Step (33) is the same as that described in the third and fourth embodiments.
Optimizes exposure timing
In order to further P_n* ΔT / T (n-1) ……… Is corrected. Here, δT depends on the exposure timing.
Amount. In step (34), the driving amount D_n′ Is default
Focus amount D_nAnd drive amount C for tracking_nAs the sum of
Can be Now, when the lens drive starts,
Automatically by setting the number of registers in the register Regis
And output from the monitoring means 103 accordingly.
Feedback pulse at the event counter
The accumulated value of this event counter is
Interrupt when it becomes equal to the value set in the register
(EVC interrupt) is assumed to occur. Drive here
A separate flag is set for the direction, and this flag
The rotation direction of the motor is controlled.
The description is omitted. In step (35), the event counter value
Reset Event to 0 and register Regis (
D_n′ −δ_cSet the number of pulses corresponding to ()). This
Thereby, the lens driving starts automatically. Next,
In step (36), enable the EVC interrupt and wait for the stop interrupt.
One. Step (37) is FIG. 19.
100f, which is calculated in step (32).
Computation was terminated because the correction amount
The correction amount after the time is sequentially corrected. That is, the amount of movement of the object image plane changes during the time ΔT.
formula ΔP_n= P_n* ΔT / T (n-1) ... Given by ΔP every ΔT seconds_n  Equivalent to
Add pulse number to the register or register
Subtract from Whether the addition is made depends on the driving direction that was determined first.
And the direction of movement of the object image plane, ie, D_n'And P_nSign of
Determined by the relationship. Here, ΔT is about 1/10 of the accumulation time
In the following, it is practically assumed that constant-speed driving was performed during accumulation.
Can be considered. In this way, the lens drive is performed by the broken line R in FIG.
And when Q 'and R intersect, the event
Counter value Event and register value Regis are not equal
And an EVC interrupt occurs. EV at step (38)
Set the C interrupt to disabled and set the lens drive in step (39).
Apply a short brake to the motion and stop it,
ΔP_nAlso stops adding to the register. Whether there is an object movement in step (40)
No, and if so, in step (41) mirror
Focusing processing such as up permission and focusing display lighting for a predetermined time
I do. In step (42), the amount required for the next calculation,
P_n, D_nAre stored and n = n + 1. Step (4
In step 3), if the mirror is down, step (2)
If the mirror is up, wait until the mirror is down
Return to step (2). On the other hand, in step (7), it is determined that there is no object movement.
If it is, the normal processing routine after step (45)
to go into. In step (45), a signal indicating that there is no object movement
With the lag, the driving amount D is determined in step (46)._n′
Focus amount D_nIs adopted as it is. Step (4
7) In D_n′ And δ_cD is determined_n′ ≦
δ_cThen, in step (51), focus processing, ie, display of focus
Lighting and mirror up permission are performed. Then step (5)
In step 2), the process jumps to step (42). In step (40)
D_n′> Δ_cIn the case of, the focus indication is turned on in step (48).
If it is lit, turn it off and go to step (49).
Operation starts and in step (50) the EVC interrupt is enabled and
Wait for the end of the movement. As described above, according to the present embodiment, during accumulation,
Time, calculation time, hoisting time, driving time, etc.
Even if it fluctuates slightly, the required drive amount for tracking
Since the change is made, in-focus shooting is always possible.
Note that the three-dot chain line S in FIG.
The exaggerated start is shown, but
Is driven until it crosses the target broken line R,
It shows that there is no problem even if there is. The embodiments described so far have been performed during storage time and during performance.
It is assumed that the lens will not be driven during the calculation time.
In that sense, it is called a tracking drive method that assumes intermittent drive.
I can. In the case of intermittent drive, the processing is time-series
So there is no need for the CPU to multitask, meaning
It can be said that it is very superior. In addition, the lens drive
If the source is shared with the power supply for normal camera operation
For example, at the time of film winding, the motor for driving the lens is
There is a restriction that it must be stopped. this
In that case the lens drive is inevitably intermittent
So far, for systems with such restrictions,
The described intermittent drive tracking method is very suitable. [Seventh Embodiment] However, a system that can always drive the lens is used.
In a system, the movement of an object is substantially continuously
It is also possible to track
The eyes also have the advantage that the movement can be felt in the name. next
In the seventh embodiment, such a continuous tracking method is described.
explain. The point of this embodiment is briefly described as follows.
Assumes the same drive form as normal intermittent drive,
Defocus amount D_nIs calculated, drive by that amount
Accumulates at the end of driving (to be referred to as convergence driving)
To start. And when it is determined that there is an object movement,
Accumulation / calculation is performed for the amount of drive for correction accompanying this object movement.
-Drive uniformly at a constant speed throughout the entire driving period. Follow
Therefore, the movement of the object is tracked by the name melaka. Continuous
The point of tracking is the same as in JP-A-60-214325.
However, the above-mentioned problem contained in this earlier application was solved.
It has been decided. That is, in this case, even during the accumulation period, the lens
Since the movement is performed, it is necessary to correct for that, but the accumulation
Since the movement inside is constant speed, the correction is easy. Also for tracking
Drive (tracking drive) is very fast
Therefore, the effect of the time measurement error on the result is small. Tracking is performed based on such a purpose.
FIG. 21 and FIG. 22 show time charts in the case of the above. FIG.
1 shows a case where there is no mirror-up operation, and FIG.
This shows a case where continuous shooting is performed upon achievement. That is, both
The difference is that accumulation starts immediately after the convergence drive ends.
Or start accumulation after the mirror is up
Is the difference. In this embodiment, under any operating condition
Since the tracking drive is continuing, the mirror has been raised halfway
Even if the state enters, the amount of movement of the object during this period is compensated every moment.
The exposure will be corrected as described in the previous examples.
There is no need for correction to adjust the light timing.
There is a point. Next, FIGS. 22 and 23 showing the same contents are used.
Then, the parameters described therein will be described. Differential
Focus D_nIs known at the end of the operation, but the value
Corresponds to the distance between P and Q 'at the midpoint of the accumulation time
doing. Also, the accumulation time of the n-th time is Tint (n),
The time from the end of the product to the start of this accumulation is T '(n-1),
From the midpoint of the previous accumulation time to the midpoint of the current accumulation time
Let the time be T (n-1). That is [0080] (Equation 1) Then, during each of these times,
Xint (n) and X '(n-
1), X (n-1). Therefore, [0082] (Equation 2) Is as follows. Next, according to the flowchart of FIG.
The operation flow will be described. Step (1) (2) (3)
(4) and (5) are the same as before. Next step (6
0) at intervals of time ΔT, which is sufficiently shorter than the accumulation time.
ΔP that has been performed continuously_nConsiderable pulse to Regis
Stop adding, that is, track during accumulation / calculation
If the drive is being performed, it is stopped. Steps
In (61), T (n-1) and X (n-1) are obtained by the above-described equations.
Is calculated. Start accumulation beforehand if this is possible
And the timer value and event at the end timing.
It is necessary to read the value of the counter
The values Tint (n), T '(n-1), Xint (n),
X '(n-1) can be calculated, so that T (n-1), X (n
-1) can be calculated. Next, in step (62), FIG.
The insufficient convergence amount P is calculated by the insufficient convergence amount calculating means 100a._nBut
The following equation P_n= D_n+ X (n-1) -D_n-1Is calculated by
You. Step (63) corresponds to the object movement determining means 100b.
In response, the object is moved by the method described in the first and second embodiments.
The presence or absence of movement is determined. Step if there is object movement
In step (64), the flag corresponding to movement is set to IDO = 1.
In step (65), the object moves per unit time ΔT
Quantity ΔP_nIs calculated by the following equation. ΔP_n= P_n* ΔT /
T (n-1) In this embodiment, this step (6
5) corresponds to the object movement correction amount calculation means 100c. S
In Step (66), the content of the event counter is set to zero.
(Event = 0), D is the register_n  Equivalent pulse
Set the number. At this moment, the lens drive starts automatically
However, the moving direction of the lens is D_nSeparately depending on the sign of
The lugs are controlled by this. Next, in step (67), the convergence drive ends.
Of the EVC interrupt that satisfies the condition (1). Step (6
In 8), ΔP every ΔT_nAdd a considerable number of pulses to Regis
Operation, and this operation is performed in the next cycle.
This is continued until step (60) is reached. Ie step
From (60) to step (60) of the next cycle is always
The target drive amount increases (decreases) at a constant speed.
You. As a result, during periods other than the convergence driving, that is, during accumulation / calculation,
Of course, the lens drive at a constant speed during the mirror up
Will be done. Convergence drive is opened from step (66).
Has been started, but the feedback from the monitoring
Of the event counter accumulating the clock pulse E
vent equals register value Regis, EVC interrupt
Occurs. In response to this, the process proceeds to step (70),
The subsequent EVC interrupt is disabled. After this register
-Since the value Regis continues to increase at a constant speed,
The lens keeps constant speed tracking drive and event counter
The value of Event also increases accordingly, ie, Regis
値 Increase the value while keeping the Event equilibrium state. This
State corresponds to the constant speed tracking drive. Then step
At (71), it is determined that the object has moved (IDO = 1), and
Defocus amount D in step (72)_n  <Δ_qIs determined
Then, a focusing process is performed in step (73). The content is
Mirror up permission, focusing display lighting for a certain period, and the like. Next, in step (74), the time
Necessary data is stored, and n = n + 1. Step (7
In 5), it is determined whether or not the mirror is currently down.
If the mirror is down, go to step (2).
Start the next accumulation. No object movement in step (63)
If yes, go to step (76),
Description is omitted because it is equivalent to FIG. As described above, in this embodiment, the object movement
When the lens is detected, the lens
It is driven at a high speed, and the
After the calculation is completed, the high-speed non-
A constant speed drive (convergence drive) may be performed. And convergent drive
The next constant speed drive starts at the end of the
Will be resumed. Because of this type of drive,
Not only gives a good impression of following the Nameraka to the movement,
Even if the accumulation timing changes due to mirror up etc.
Without any problems, you can always take pictures that are in focus. [Eighth Embodiment] Next, in order to determine again whether or not the object has moved.
This method will be described in the following eighth embodiment. object
Regarding the contents of the movement determining means 100b, the simplest
In the first embodiment, a simple shape was described, and the accuracy was further improved.
The method has been described in the second embodiment. Here, the time element is added.
Taste allows for more accurate determination of object movement
The method will be described. Steps (1005) and (1005) in FIG.
6) corresponds to this. That is, the step (1002) is strict.
Step (10)
05) [0088] (Equation 3) The step (1003) is performed at a strict speed.
It is in step (1006) that the degree was compared. [0090] (Equation 4) Is as follows. Steps (1002), (100
Step 5) mainly determines the presence or absence of the movement of the object, and performs step (100).
3) and (1006) mainly determine whether or not the subject is out of position.
I have. In actual use, step (1002),
Step (1003) instead of (1003)
5) and (1006) may be used, but as shown in FIG.
You may use both as a lease. In that case
In steps (1002) and (1003), step (10)
05) and (1006)
0 <k <k ′ <1 <r ′ <r. Specific values of these coefficients are k ≒
3-5, k '≒ 5-7, r' ≒ 4-2, r ≒ 2
Approximately 3 is good. According to the eighth embodiment,
Since the comparison is performed by obtaining the moving speed of the body more precisely,
There is an advantage that the accuracy of determining the presence or absence of motion is improved. [0093] As described above, according to the present invention, the target object
Is moving, even if it moves
Predict the movement locus of the imaging plane of the imaging optical system
Calculation, and furthermore, the focus of the moving target object is
The prediction calculated by the prediction means so as to be obtained at the time of exposure.
The correction amount to be added to the measured movement amount is calculated. And said
When moving the target object based on the predicted movement amount and the correction amount,
Controls the imaging optics even before the mirror is raised
(That is, forward drive control),
Has the effect of That is, when moving the target object,
Since the imaging optical system is always driven by dynamic control,
No matter what the lens (mirror up) is,
You can get a matching photo. Specifically, the shooting motion of optical equipment
At the time of work, in the time from release to exposure
Move the target object in consideration of the period during which lens driving is stopped
Drive the imaging optics earlier than the trajectory of the image plane
Movement (driven according to the predicted trajectory),
Even when the imaging optical system is stopped, just pin
This has the effect of allowing you to take pictures that are suitable for shooting. Simply,
If the imaging optical system was driven according to the trace,
There is a period during which the imaging optics is stopped in relation to the release signal
Can follow the movement of the target object during the lens stop period.
Instead, the exposure is out of focus at the time of exposure. This
For shooting operation with lens stop just before exposure like
It is the present invention that enables proper driving of the imaging optical system
It is.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a block diagram of an automatic focusing device according to a first embodiment of the present invention. FIG. 2 is a relationship diagram showing how an automatic focus adjustment device tracks a subject. FIG. 3 is a detailed block diagram of a correction unit of the automatic focusing adjustment device. FIG. 4 is a flowchart of the automatic focusing adjustment device. FIG. 5 is a flowchart of an automatic focusing adjustment apparatus according to a second embodiment of the present invention. FIG. 6 is an explanatory diagram showing exposure timing in a known normal intermittent driving method. FIG. 7 is a relationship diagram showing a state of a focusing operation of the automatic focusing device in the intermittent driving method. FIG. 8 is a block diagram of a correction unit of the automatic focusing device according to the third embodiment of the present invention. FIG. 9 is a flowchart of the automatic focusing adjustment device. FIG. 10 is a detailed explanatory diagram of S10 in FIG. 9; FIG. 11 is a detailed explanatory diagram of S10 in FIG. 9; FIG. 12 is a relationship diagram showing timings at the time of continuous photographing by the motor drive device of the automatic focusing adjustment device using the tracking drive method according to the first embodiment. FIG. 13 is a relationship diagram showing timings at the time of the continuous shooting of the automatic focusing apparatus according to the fourth embodiment of the present application. FIG. 14 is a block diagram of a correction unit of the automatic focusing adjustment device. FIG. 15 is a flowchart of the automatic focusing device. FIG. 16 is a detailed view of a flowchart of the automatic focusing apparatus. FIG. 17 is a relationship diagram showing timing at the time of photographing of the automatic focusing apparatus according to the fifth embodiment of the present application. FIG. 18 is a relationship diagram illustrating timings at the time of shooting of the automatic focusing apparatus according to the sixth embodiment of the present application. FIG. 19 is a block diagram of a correction unit of the automatic focusing apparatus. FIG. 20 shows a flowchart of the automatic focusing apparatus. FIG. 21 is a relationship diagram showing a timing at the time of photographing of the automatic focusing device when there is no mirror up. FIG. 22 is a relationship diagram showing a timing at the time of photographing of the automatic focusing apparatus when there is a mirror up; FIG. 23 is a relationship diagram showing timings at the time of photographing of the automatic focusing device. FIG. 24 is a block diagram of a correction unit of the automatic focusing apparatus. FIG. 25 is a flowchart of the automatic focusing device. FIG. 26 is a block diagram of a conventional automatic focusing device. FIG. 27 is a relationship diagram showing timings at the time of photographing of the automatic focusing device. [Description of Signs] 100: Correcting means 101: Focus detecting means 102: Control means 103: Monitoring means 104: Lens driving means 105: Lens information generating means

Claims (1)

(57) [Claims] In an automatic focus adjustment device used for an optical device that performs exposure by stopping an imaging optical system during exposure, a predetermined imaging surface that is predetermined as an imaging surface of the imaging optical system that forms a light image of a target object focus detection means for defocusing amount iterative detection and, in response to an output from said focus detecting means, the predicted amount of movement <br/> by predicting the movement locus of the imaging plane due to the movement of the target object Prediction means for outputting the focus of the moving target object at the time of the exposure.
In addition to the predicted movement amount obtained by the prediction means,
Correction means for calculating the amount of correction to be performed, and when it is determined that the target object is moving,
Even before the correction, the correction amount obtained by the correction means and the
The imaging light is aimed at a predicted trajectory determined from the predicted movement amount.
An automatic focus adjustment device that drives academic systems . 2. The correction amount is an amount about the image plane movement amount that depends on the movement of the target object for a given time, the predicted transfer
2. The automatic focusing apparatus according to claim 1 , wherein the correction amount is added to or subtracted from a moving amount . 3. The predetermined time is a predetermined operation during a camera sequence,
3. The automatic focusing apparatus according to claim 2, wherein the time interval is a time interval related to a stop of driving of the image forming optical system with respect to an operation from the start of mirror up to exposure.