JPH0694985A - Camera autofocus device - Google Patents

Abstract

(57) [Summary] [Purpose] To improve the stability and accuracy of subject tracking correction. According to the defocus amount and focus detection time interval information obtained a plurality of times in the past by the focus detection means, the change in the image plane position of the subject after a predetermined time is obtained by approximating the determined function, and after a predetermined time. In the calculation means PRS for controlling the lens drive means LPRS, LMTR so as to match the image plane position of the subject and the image plane position of the photographing lens, the weighting considering the reliability of the change speed of the image plane position obtained by the focus detection is performed. Setting means PRS for setting a function that approximates the average is provided, and in order to evaluate the reliability of the changing speed of the image plane position, the time interval at which the detected focus detection is performed and the focus detection accuracy are used. There is.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION The present invention is provided in a camera,
The present invention relates to an improvement of an automatic focus adjustment device capable of performing follow-up control on a moving subject.

[0002]

2. Description of the Related Art Conventionally, most automatic focus adjustment systems for single-lens reflex cameras attempt to focus on a subject by repeating a cycle of "focus detection (sensor signal input, focus detection calculation) and lens drive". It is a thing.

In such a conventional apparatus, when the defocus amount obtained by the focus detection after driving the lens is zero, it is determined that the focus is achieved.

However, at the time of actual photographing, it is not necessary for the photographing lens to accurately come to the in-focus position (the defocus amount does not have to be zero), the open F number of the lens and the aperture value at the time of photographing. Therefore, if the focus shift amount (defocus amount) is equal to or less than the allowable amount (image plane depth), it can be regarded as the focus, so the focus position is not a single point but is given as a focus zone having a width, and focus detection is performed. If the defocus amount detected by the means is within the focus zone, it is determined to be in focus,
The focus was displayed.

Then, by repeating the cycle of "focus detection (sensor signal input, focus detection calculation), lens drive" as described above, the object is focused, and the lens drive amount in each cycle. Is based on the defocus amount at the time of performing focus detection in that cycle, and this is expected to eliminate the defocus amount at the time of focus detection at the end of lens driving.

Therefore, as a matter of course, the focus detection and the lens drive require a considerable amount of time, so that in the case of a subject having a large movement, the defocus amount changes during the focus detection and the lens drive, The defocus amount to be eliminated may be significantly different from the detected defocus amount, resulting in a problem that the subject is out of focus at the end of lens driving.

As an automatic focus adjustment system for the purpose of solving the above problems, the applicant of the present application has filed Japanese Patent Application No. 62-328233.
I am proposing an issue.

The gist of this proposal is to predict the defocus change caused by the movement of the object from the detected defocus change and the time interval in each cycle, and to correct the lens drive amount (hereinafter referred to as "following"). This is referred to as correction), and from the viewpoint of focusing accuracy at the end of lens driving, the above method is expected to improve the above problem.

However, when the tracking correction is performed, the following problems occur. The tracking correction will be described below with reference to the drawings.

FIG. 8 is a diagram for explaining a lens driving method by follow-up correction, in which the horizontal axis represents time t and the vertical axis represents the image plane position x of the subject.

A curve x (t) represented by a solid line means an image plane position at time t of a subject approaching the camera in the optical axis direction when the taking lens is at infinity. L (t) represented by a broken line means the position of the photographing lens at time t, and when x (t) and L (t) match, the lens is in focus. Then, [ti, ti '] is a focus detection operation, and [ti', ti + 1] is a lens driving operation. Further, in the conventional example shown in the same figure, it is assumed that the image plane position changes according to a quadratic function. That is, at the time t3, the image plane positions (t1, x
1) If (t1, x2) (t3, x3) is known, the time after TL (AF time lag + release time lag) from time t3 based on the following equation x (t) = at ² + bt + c (1) image plane position x ₄ in t ₄ in which can be predicted.

However, what the camera can actually detect is not the image plane positions x1, x2 and x3, but the defocus amount D.
F1, DF2, DF3, and lens drive amounts DL1 and DL2 converted into image plane movement amounts. Further, the time t ₄ is a future value to the last, and in reality, it is a value that changes in accordance with a change in the accumulation time of the accumulation type sensor depending on the subject brightness. I assume.

T4-t3 = TL = TM2 + (release time lag) Based on the above assumption, the lens drive amount DL3 calculated from the focus detection result at the time t3 is obtained as follows.

X (t) = at2 + bt + c (2) Then, considering (t1, L1) in the figure as the origin, t1 = 0 x1 = DF1 (3) t2 = TM1 x2 = DF2 + DL1 (4) t3 = TM1 + TM2 x3 = DF3 + DL1 + DL2 (5) Substituting the equations (3), (4), and (5) into the equation (2), a, b, and c are substituted. When asked, Therefore, the lens driving amount DL3 converted into the image plane movement amount at the time t4 is as follows: DL3 = x4-L3 = x4-x3 + DF3 = a {(TM1 + TM2 + TL) ^2- (TM1 + TM2) ² } + b.TL + DF3 (9) Motomaru.

However, in the control in which the tracking correction is performed,
There is a problem that a prediction error (correction error) caused by a focus detection error or a lens drive error becomes several times to ten or more times the focus detection error or the lens drive error.

FIG. 9 is a diagram for explaining the principle of occurrence of this prediction error.

What the camera actually detects is the drive amount and the defocus amount, and the image plane position of the subject is calculated based on these values. Therefore, the focus detection error and the lens driving error become the detection error of the image plane position.
When an image plane position detection error of δ1 occurs at time t2 with respect to the actual image plane position x (t) written by a solid line as shown in FIG. 9, the image plane position detected by the camera is one point difference. It becomes x '(t) written in a line. At this time,
Time t at which the last focus detection was performed for the detection error δ1
The difference in the image plane position after a predetermined time TL from 3 is δ _P ,
This is the prediction error.

As described above, the prediction error δ _P becomes several times as large as the error δ1 at the time of detection, and in some cases, becomes more than ten times. As described above, in the follow-up predictive control, a detection error of a level that does not cause a problem in the conventional control is greatly enlarged, so that the prediction error exceeds a permissible level.

As one of means for solving such a problem, Japanese Patent Laid-Open No. 63-5316 proposes to reduce the influence of disturbance noise by weighted averaging and perform stable follow-up prediction control. .

This is to obtain the average value of the average image plane moving speed LWR (1) obtained by the previous calculation and the image plane moving speed WR (0) detected this time, as in the following equation (10). In this case, as in formula (11), the specific gravity of the image plane moving speed detected this time is increased to emphasize the follow-up performance for a subject whose change in image plane movement is large.

Note that LWR (0) is the average image plane moving speed obtained this time, LWR (1) is the average image plane moving speed obtained last time, and WR (0) is the image plane moving speed detected this time.

By performing the averaging process in this way, the influence of the error contained in the image plane moving velocity detected this time is "1/2" in the above equation (10) and "1/2" in the above equation (11). It can be reduced to 3/4 ", which enables stable prediction control that is not easily affected by the detection error.

[0023]

However, in the above-described conventional example, the following problems occur because the weighting used when performing weighted averaging is always constant if the usage conditions are the same. is doing.

FIG. 10 shows the relationship between the image plane position and time when the image plane position is constant, that is, when focus detection is performed on a stationary subject. At this time, two values TA and TB (TA <T
In B), the image plane moving velocities V1 and V2 detected when the same focus detection error δ1 occurs are as follows.

V1 = δ1 / TA (12) V2 = δ1 / TB (13) Here, since the subject is not moving, V1 and V2 are detection errors of the image plane moving speed. , Its size is "V1>V"
2 ”. Here, even if the focus detection error is the same, if the focus detection time interval is different, the detection error of the image plane moving speed is also different,
The smaller the focus detection time interval, the larger the detection error of the image plane moving speed.

As described above, if the weighted averaging process is performed without evaluating the error component included in the detected image plane moving velocity, the averaging process appropriate for the situation cannot be performed, and depending on the situation, it may be excessive. There are problems such as deterioration of tracking performance due to averaging, and insufficient stability due to insufficient averaging.

One of the reasons why the weighting of the averaging process is insufficient or excessive is that the focus detection time interval changes significantly, which occurs when the lens is driven or not driven. . This is because if the lens drive is performed during focus detection, the focus detection time interval increases by the lens drive time required for lens drive, and otherwise the focus detection time interval does not increase. Is likely to occur when the subject has a slow speed and the lens drive amount is not so large.

(Object of the Invention) It is an object of the present invention to provide an automatic focus adjusting device for a camera capable of improving the stability and accuracy of tracking correction of an object.

[0029]

According to the present invention, an image plane position of a subject after a predetermined time is determined based on a defocus amount and focus detection time interval information obtained a plurality of times by the focus detection means. And the reliability of the speed of change of the image plane position obtained by focus detection in the arithmetic means for controlling the lens driving means so as to match the image plane position of the subject and the image plane position of the photographing lens after a predetermined time. In order to evaluate the reliability of the changing speed of the image plane position, the time interval at which the detected focus is detected, the focus detection accuracy (each The reliability of the defocus amount) is used.

[0030]

DESCRIPTION OF THE PREFERRED EMBODIMENTS FIG. 1 is a block diagram showing the essential structure of a camera equipped with an automatic focusing device according to a first embodiment of the present invention.

In FIG. 1, PRS is a camera control device, for example, a CPU (central processing unit), ROM, R
It is a one-chip microcomputer having an AM and A / D conversion function. This control device (hereinafter referred to as a computer) PRS performs a series of camera operations such as an automatic exposure control function, an automatic focus detection function, and film winding according to a camera sequence program stored in a ROM. For that purpose, the computer PRS is provided with the synchronous communication signals SO, SI, SCLK and the communication selection signals CLCM, CS.
The DR and CDR are used to communicate with peripheral circuits and lenses in the camera body to control the operation of each circuit and lens.

SO is a data signal output from the computer PRS, SI is a data signal input to the computer PRS, and SCLK is a synchronous clock of the signals SO and SI.

LCM is a lens communication buffer circuit,
When the camera is operating, power is supplied to the lens power supply terminal and the selection signal CL from the computer PRS is supplied.
When CM is at a high potential level (hereinafter referred to as "H"), it serves as a communication buffer between the camera and the lens. That is, the computer PRS sets the selection signal CLCM to "H" to set the SCL
When certain data is sent from SO in synchronization with K, LC
M outputs the buffer signals LCVK and DCL of SCLK and SO to the lens via the camera / lens indirect points. At the same time, the buffer signal of the signal DLC from the lens is output as SI, and the computer PRS inputs the above SI as data from the lens in synchronization with SCLK.

SDR is a drive circuit of a line sensor device SNS for focus detection which is composed of a CCD or the like, and has a signal C
Selected when SDR is "H", SO, SI, SCL
It is controlled from the computer PRS using K.

Signals CK are CCD driving clocks φ1, φ
2 is a clock for generating 2, and the signal INTEND is a signal for notifying the computer PRS that the accumulation operation is completed.

The output signal OS of the sensor device SNS is a time-series image signal synchronized with the clocks φ1 and φ2, is amplified by the amplifier circuit in the drive circuit SDR, and is then output to the computer PRS as the image signal AOS. Computer PR
S inputs the image signal AOS from the analog input terminal, and CK
R / S after synchronizing A / D with the internal A / D conversion function
The data is sequentially stored at a predetermined address of AM.

Similarly, the SAGC which is the output signal of the sensor device SNS is the output of the AGC (Auto Gain Control) sensor in the sensor device SNS, which is input to the drive circuit SDR and the image at the sensor device SNS is input. Used for signal accumulation control.

SPC is a photometric sensor for exposure control that receives light from a subject through a taking lens, and its output SSPC is input to an analog input terminal of a computer PRS, and after A / D conversion, according to a predetermined program. Used for automatic exposure control (AE).

DDR is a switch detection and display circuit, which is selected when the signal CDRD is "H", and SO, S
It is controlled from the computer PRS using I and SCLK. That is, the display of the display member DPS of the camera is switched based on the data sent from the computer PRS, and the on / off states of various operation members of the camera are notified to the computer PRS by communication.

The switches SW1 and SW2 are interlocked with a release button (not shown), and are the first release button.
The switch SW1 is turned on by pressing the step, and the switch SW2 is turned on by pressing the second step. As will be described later, the computer PRS performs photometry and automatic focus adjustment operations when the switch SW1 is turned on, and exposure control and film winding are performed when the switch SW2 is turned on.
The switch SW2 is connected to the "interruption input terminal" of the computer PRS, and even when the program is executed when the switch SW1 is on, an interrupt is generated by turning on the switch SW2, and it is possible to immediately shift to a predetermined interrupt program.

MTR1 is a film feeding motor, MTR2 is a motor for mirror up / down and shutter spring charging, and forward / reverse control is performed by respective drive circuits MDR1 and MDR2. Computer PRS to M
Signals M1F and M1R input to DR1 and MDR2
, M2F, M2R are signals for motor control.

MG1 and MG2 are shutter front and rear curtain running start magnets, respectively, which are energized by signals SMG1 and SMG2 and amplification transistors TR1 and TR2, and shutter control is performed by a computer PRS.

Since the switch detection and display circuit, the motor drive circuits MDR1 and MDR2, and the shutter control are not directly related to the present invention, detailed description thereof will be omitted.

The signal DCL input to the in-lens control circuit LPRS in synchronization with LCK is data of a command from the camera to the lens FLNS, and the operation of the lens in response to the command is predetermined.

The in-lens control circuit LPRS analyzes the command in accordance with a predetermined procedure, and performs focus adjustment and diaphragm control operations, and various lens parameters (open F number, focal length, defocus amount versus extension) from the output DLC. Output the quantity coefficient, etc.).

In the embodiment, an example of a zoom lens is shown, and when a focus adjustment command is sent from the camera, the focus adjustment motor LMTR is controlled by the signals LMF, according to the drive amount and direction sent at the same time. Controlled by LMR,
Focusing is performed by moving the optical system in the optical axis direction. The movement amount of the optical system is the pulse signal SENCF of the encoder circuit ENCF.
Is monitored and counted by the counter in the lens control circuit LPRS, and when the predetermined movement is completed, the circuit LPRS itself sets the signals LMF and LMR to "L" to stop the motor LMTR.

Therefore, once the focus adjustment command is sent from the camera, the control device PRS in the camera does not need to be involved in lens driving at all until the lens driving is completed.

When an aperture control command is sent from the camera, a stepping motor DMTR known for driving the aperture is driven in accordance with the number of aperture steps sent at the same time. Since the stepping motor can be open-controlled, it does not require an encoder for monitoring the operation.

ENCZ is an encoder circuit attached to the zoom optical system, and the in-lens control circuit LPRS inputs the signal SENCZ from the encoder circuit ENCZ to detect the zoom position. The lens parameter at each zoom position is stored in the in-lens control circuit LPRS, and when there is a request from the computer PRS on the camera side, the parameter corresponding to the current zoom position is sent to the camera.

The operation of the camera having the above structure will be described with reference to the flowchart of FIG.

When a power switch (not shown) is turned on, power supply to the computer PRS is started, and the computer PRS starts executing the sequence program stored in the ROM.

When the execution of the program is started by the above operation, the program proceeds to step (002) through step (002).
, The state of the switch SW1 which is turned on when the release button is pressed in the first step is detected.
When W1 is OFF, the process proceeds to step (003) to clear and initialize all the control flag variables set in the RAM in the computer PRS.

The steps (002) and (003) are
The process is repeated until the switch SW1 is turned on or the power switch is turned off. When the switch SW1 is turned on, the process moves from step (002) to step (005).

In step (005), a "photometry" subroutine for exposure control is executed. The computer PRS is the output S of the photometric sensor SPC shown in FIG.
The SPC is input to the analog input terminal, A / D converted, and the optimum shutter control value and aperture control value are calculated from the digital photometric value and stored in a predetermined address of the RAM. Then, during the release operation, the shutter and the diaphragm are controlled based on these values.

Then, in step (006),
The "image signal input" subroutine is executed. Although the flow of this subroutine is shown in FIG. 3, a detailed description will be given later, but in this step (006), the computer PRS inputs an image signal from the focus detection sensor device SNS.

In the next step (007), the defocus amount DE of the photographing lens is calculated based on the input image signal.
Calculate F. The specific calculation method is disclosed by the applicant of the present application in Japanese Patent Application No. 61-160824, and the detailed description thereof is omitted here.

In step (008), a "prediction calculation" subroutine is executed. This "prediction calculation" subroutine corrects the lens drive amount and will be described in detail later.

In the next step (009), the "lens drive" subroutine is executed, and the previous step (00
The lens is driven based on the lens driving amount corrected in 8). Details of this "lens drive" subroutine will be described later with reference to FIG.

After the lens driving is completed, the step (00
2), until the SW1 is turned off or the release switch SW2 (not shown) is turned on, the above step (00
5) to (009) are repeatedly executed, and preferable focus adjustment is also performed on the moving subject.

Now, when the release button is pushed further,
When the switch SW2 is turned on, the process immediately proceeds to step (010) in any step by the interrupt function to start the release operation.

In step (011), it is determined whether or not the lens drive is being executed. If the lens drive is in progress, the process proceeds to step (012), a lens drive stop command is sent to stop the lens, and step (013). Proceed to. On the other hand, if the lens is not driven, the process immediately proceeds to step (013).

In step (013), the quick return mirror of the camera is raised. This is executed by controlling the motor control signals M2F and M2R shown in FIG. In the next step (014), the aperture control value already stored in the "photometry" subroutine of the previous step (005) is sent as an SO signal to the in-lens control circuit LPRS via the circuit LCM to control the aperture. Let it be done.

Whether or not the mirror up and diaphragm control performed in the steps (013) and (014) are completed is detected in the step (015). The mirror up is not shown in the figure and is detected. It can be confirmed by a switch, and the aperture control confirms by communication whether or not the lens has been driven to a predetermined aperture value. If any of them is not completed, the process waits at this step and the state detection is continued. When both control ends are confirmed, the process proceeds to step (016).

In step (016), the shutter is controlled at the shutter time already stored in the "photometry" subroutine of the previous step (005) to expose the film.

When the shutter control is completed, the process proceeds to the next step (017), in which a command is sent to the lens by the above-mentioned communication operation so as to open the diaphragm, and then in step (018). , Down the mirror. The mirror down is executed by controlling the motor MTR2 using the motor control signals M2F and M2R as in the mirror up.

In the next step (019), like the step (015), the completion of the mirror down and the aperture opening is waited for. When both the mirror down and the aperture opening control are completed, the process proceeds to step (020).

At step (020), one frame of film is wound by properly controlling the motor control signals M1F and M1R shown in FIG.

The above is the entire sequence of the camera that has performed the predictive AF.

Next, the "image signal input" subroutine called in step (006) of FIG. 2 will be described with reference to the flowchart of FIG.

"Image signal input" is an operation executed at the beginning of a new focus detection operation, and when this subroutine is called, step (101) is passed through step (10).
In 2), the timer value TIMER of the free-running timer possessed by the computer PRS itself is set to the storage area T on the RAM.
By storing it in N, the start time of the focus detection operation is stored.

In the next step (103), the contents of the memories TM1 and TM2 for recording the time intervals in the lens drive amount correction equations (6), (7) and (9) are updated. Before executing this step (103), the memory TM
1 and TM2 record the time intervals of the focus detection operation two times before and the previous time, and the memory TN1 stores the time when the previous focus detection operation was started.

Therefore, in this step (103), (T
N1-TN) represents the time interval of the focus detection operation from the previous time to this time, and this is recorded in the memory TM2. Further, the data immediately before the execution of "TM2 ← (TN1-TN)", that is, the focus time interval from the last two times to the previous time is recorded in the memory TM1. In this step (103), the memory TM1 always stores the time-preceding time interval data, and the memory TM2 stores the previous time interval data.

In the next step (104), the sensor device SNS is caused to start accumulating an optical image. Specifically, the computer PRS sends a "storage start command" to the drive circuit SDR by communication, and in response to this, the drive circuit SDR causes the drive circuit SDR to clear the clear signal CLR of the photoelectric conversion element section of the sensor device SNS.
To "L" to start charge accumulation.

In step (105), the timer value of the free-running timer is stored in the variable TI to store the current time.

In the next step (106), the state of the input INTEND terminal of the computer PRS is detected to check whether or not the accumulation is completed. The sensor drive circuit SDR sets the signal INTEND to "L" at the same time as the accumulation starts, and the sensor device S
The AGC signal SAGC from the NS is monitored and the signal S
When AGC reaches a predetermined level, the signal INTEND is set to "H", and at the same time, the charge transfer signal SH is set to "H" for a predetermined time,
It has a structure for transferring the charges of the photoelectric conversion element section to the CCD section. Here, if the INTEND terminal is "H", it means that the accumulation is completed, and thus the process proceeds to step (110),
If it is "L", it means that the accumulation is not completed yet, and thus the process proceeds to step (107).

In step (107), the time TI stored in step (105) is subtracted from the timer value TIMER of the free-running timer, and this is stored in the variable TE. Therefore, the variable TE stores the time from the start of accumulation until this time, so-called accumulation time.

In the next step (108), the variable TE is compared with the constant MAXINT. If the variable TE is less than the constant MAXINT, the process returns to step (106) and waits for the end of accumulation again. On the other hand, when the variable TE becomes equal to or larger than the constant MAXINT, the process proceeds to step (109).

In step (109), the accumulation is forcibly ended. Computer PRS for forced termination
Is executed by sending a "storage end command" from the drive circuit to the drive circuit SDR. The drive circuit SDR is a computer PR
When the "storage end command" is sent from S, the charge transfer signal SH is set to "H" for a predetermined time and the charge of the photoelectric conversion unit is CCed.
Transfer to section D. The accumulation of the sensor is completed by the flow up to this step (109).

In the next step (110), the signal OS of the sensor device SNS is amplified by the drive circuit SDR, A / D conversion of the image signal AOS and RA of its digital signal are carried out.
Store in M. More specifically, the drive circuit SDR generates CCD drive clocks φ1 and φ2 in synchronization with the clock CK from the computer PRS and supplies them to the sensor device SNS.
The CCD section is driven by 1 and φ2, and the charge in the CCD is output as an image signal from the output OS in time series. This signal is amplified by the amplifier inside the drive circuit SDR and then input to the analog input terminal of the computer PRS as the image signal AOS. The computer PRS performs A / D conversion in synchronization with the clock CK output by itself,
The digital image signal after A / D conversion is sequentially stored in a predetermined address of the RAM.

When the input of the image signal is completed in this way, this "image signal input" subroutine is returned in step (111).

Next, the "lens drive" subroutine called in step (009) of FIG. 2 will be described with reference to the flowchart of FIG.

When this subroutine is executed, in step (202) through step (201), communication with the lens is made and two data "S" and "PTH" are input. "S" is a "coefficient of the defocus amount versus the amount of extension of the focus adjustment lens" peculiar to the photographing lens. In the case of a zoom lens, “S” is determined by the control circuit LPRS according to each zoom position detected by the encoder ENCZ.
“RTH” is the amount of extension of the focus adjustment lens per one output pulse from the encoder ENCF that is interlocked with the movement of the focus adjustment lens FLNS in the optical axis direction.

Therefore, the defocus amount D for the focus adjustment
A value obtained by converting the extension amount of the focus adjustment lens into the number of output pulses of the encoder by L, the above S, and PTH, that is, the so-called lens drive amount FP is given by the following equation.

FP = DL × S / PTH Step (203) executes the above equation as it is.

In the next step (204), the lens drive amount FP obtained in the above step (203) is sent to the lens to drive the focus adjustment lens (in the case of the whole extension type single lens, the entire photographing lens). Command.

In step (205), the lens drive amount F instructed in step (204) by communicating with the lens.
It is detected whether or not the driving of P is finished, and when the driving is finished, the routine proceeds to step (206) and the "lens driving" subroutine is returned.

In this lens drive completion detection, as described above, the counter in the in-lens control circuit LPRS counts the pulse signal of the encoder ENCF, and it is determined whether or not the counted value matches the lens drive amount FP. It is executed by detecting the above communication.

Next, the "prediction calculation" subroutine called in step (008) of FIG. 2 will be described with reference to the flowchart of FIG.

In the "prediction calculation" subroutine, stable prediction control is performed by weighted averaging in consideration of the focus detection time interval when the image plane moving speed at which prediction control tends to become unstable is small. .

When this subroutine is executed, it is determined in step (302) through step (301) whether or not to count up a counter COUNT for determining whether or not the data required for prediction has been accumulated. To do. In the present embodiment, when the focus detection data and the lens driving data are accumulated three or more times, that is,
If "COUNT>2", the predictive calculation is possible, and there is no need to further count up. Therefore, "COUNT>
If “2”, the process proceeds to step (304). Also,
If "COUNT≤2", the process proceeds to step (303), the counter COUNT is counted up, and then the process proceeds to step (304).

In step (304), the data for the current prediction calculation is updated.

That is, the prediction calculation is performed in the above (6), (7),
Since it is performed based on the equations (8) and (9), the data is the defocus amount DF3 of this time in FIG.
The defocus amounts DF2 and DF1 of the previous and the previous two times, the lens drive amount DL1 of the two previous times, the lens drive amount DL2 of the previous time,
Time interval TM between the focus detection two times before and the previous focus detection
1. The time interval TM2 from the previous focus detection to the current focus detection and the expected time lag TL are required.

Therefore, in this step (304), the defocus amount D detected this time each time focus detection is performed.
F is input to the storage area DF3 on the RAM, the previous defocus amount is input to the storage area DF2, the previous defocus amount is input to the storage area DF1, and the lens drive amount DL is converted to the previous image plane movement amount. To the storage area DL2, and to the storage area DL1 the lens drive amount DL1 calculated in terms of the image plane movement amount two times before.
To the storage area VA.
By inputting 0, the data in each storage area is updated to the data required for the current prediction calculation.

In step (305), the average image plane moving speed V1 from the previous focus detection to the current focus detection is calculated by using each data updated in the above step (304).
Is calculated according to the following equation.

V1 = (DL2 + DF3−DF2) / TM2 (14) In the next step (306), the average image plane moving speed VA is calculated by the weighted average processing weighted by the focus detection time interval. . Here, the calculation formula of the weighted average is as follows.

VA = {(TX−TM2) · VA0 + TM2 · V1} / TX (15) In the above equation (15), TX is a predetermined value, and this TX is T
Since the value may be larger than M2, the normal focus detection time interval (time required for focus detection and lens driving) is "1.
Since it is “00 to 150 msec”, for example, “TX
= 500 msec ". Further, since the release operation is further performed during focus detection during continuous shooting, the focus detection time interval becomes longer accordingly, so the value of TX may be changed to a larger value during continuous shooting.

Further, VA0 is the average image plane moving speed calculated last time, which is zero in the initial state. When the time interval TM2 from the previous focus detection to the current focus detection is small, the weight of VA0 is large, and when TM2 is large, the weight of V1 is small.

Thus, the detected image plane moving speed V1
When the focus detection time interval TM2 at which the reliability of the image is reduced is small, the weight of the time interval TM2 of low reliability is decreased and the weight of the average image plane moving speed VA0 is increased. Therefore, the average image plane moving speed VA is reliable. When the focus detection time interval TM2 is large and a reliable image plane moving speed V1 is obtained, the weight of the image plane moving speed V1 is increased and the response of the tracking prediction control is increased. Acts to increase speed.

In the first embodiment, the reliability of the image plane moving speed detected by the focus detection time interval is evaluated, and the weighting of the weighted average is controlled by these.

In the next step (307), the estimated time lag TL required for performing the prediction calculation is calculated. The estimated AF time lag (time required for focus adjustment) at this time is calculated from the previous focus detection to the current time. The sum of TM2 and the release time lag TR is calculated as TL under the assumption that the time interval TM2 required for focus detection is almost the same.

In the next step (308), it is judged whether or not the average image plane moving speed VA obtained in the above step (306) is larger than a predetermined value AX, and when it is larger, the process proceeds to step (309), If not, the process proceeds to step (310). In this step, when the image plane moving speed is high, the process proceeds to step (309) in order to perform the predictive control of the quadratic function with emphasis on the tracking performance, and when the image plane moving speed is low, the process proceeds to step (310). Performs predictive control with an emphasis on stability.

When the process moves to step (309),
In this step, it is determined whether or not the data necessary for performing the predictive control using the quadratic function is accumulated.
The content of the counter COUNT is "3", that is, "COUNT
If it is ">2", the predictive control using the quadratic function is possible, so the process proceeds to step (311), and if not, the process proceeds to step (310).

In steps (311) and (312), each of the storage areas DF1 to DF3, DL1, DL2, TM.
1, based on the data stored in TM2, (6),
A and B representing the a and b terms of the equation (7) are obtained, and step (3
Go to 13).

In the next step (313), the data of each storage means and steps (307), (311),
Based on the calculated value of (312), the calculated value of the equation (9) is calculated, and the lens drive amount DL converted in terms of the image plane movement amount this time is calculated.
When this step is completed, the process proceeds to step (316).

Step (308) or step (3)
When the process proceeds from step 09) to step (310), it is determined in step (310) whether or not the data necessary for the predictive control using the linear function is accumulated. In this step, if the content of the counter COUNT representing the amount of data accumulated is "1", it is judged that the predictive control cannot be performed because the data is accumulated only once, and the step (31
If the content of the counter COUNT is "2" or "3", the predictive control is possible, so step (3) is performed.
Go to 14).

At step (314), the lens driving amount DL which is in focus after the time TL is calculated using the average image plane moving speed VA obtained at step (306) as in the following equation.

DL = DF3 + VA · TL (16) In step (315), since the predictive control is not performed, the lens drive amount DL is the defocus amount DF obtained by the focus detection this time.

In the next step (316), the above step (313) or steps (314), (31)
Lens drive amount DL obtained in 5) and shooting lens opening F
The number FN and the predetermined coefficient δ (in this embodiment, the circle of least confusion
0.035mm) product (FN · δ) is compared, and “DL <FN ·
If δ ”, the process proceeds to step (317), and if not, the process returns at step (318).

The above (FN · δ) represents the depth of field, and in step (316) the lens drive amount DL is smaller than the depth of field (FN · δ), that is, the lens drive amount DL is within the depth. This is a value, and when it is determined that the in-focus state can be maintained without lens driving and there is no need for lens driving, the lens driving amount DL is set to 0, and lens driving is prohibited. As a result, unnecessary microlens driving is not performed, and both the feeling of use and power consumption can be improved.

Further, in this embodiment, FN is the open F number of the taking lens, but there is no problem in using this as the taking aperture value, and δ is not limited to 0.035 mm. When this step is completed, the next step (31
This subroutine is returned at 8).

As described above, when the image plane moving speed is low and the predictive control is likely to be unstable due to the lens driving or not, the predictive control is performed using the image plane moving speed subjected to the weighted average processing. Since the reliability of the detected image plane movement speed becomes low when the lens is not driven (the focus detection time interval becomes short), the weighting of the detected image plane movement speed is reduced in such a case. Since the weighted average is performed, more stable follow-up prediction control can be performed.

(Second Embodiment) The second embodiment of the present invention assumes a camera configured to perform a weighted average process in consideration of focus detection accuracy, and the overall configuration of the camera is the first. Since it is the same as that of the first embodiment, only the "prediction calculation" subroutine having different contents will be described here.

FIG. 6 is a flow chart showing the "prediction calculation" subroutine in the second embodiment of the present invention. The feature of this embodiment is that the reliability of the image plane moving speed detected by the focus detection accuracy and the focus detection time interval. The weighting average is controlled by evaluating the sex.

From step (401) to step (40
The flow up to 5) is the same as step (301) to step (305) in FIG. 5 in the first embodiment, and the description of this part is omitted here.

In step (406), the focus detection accuracy reliability is evaluated from the contrast of the image signal and the correlation between the two images when the stored defocus amounts DF2 and DF3 are detected, and the focus detection accuracy reliability of DF2 and DF3 is evaluated. If the reliability is high, the process proceeds to step (407). If not, that is, if the reliability of either or both of DF1 and DF2 is low, the process proceeds to step (408).

In step (407), since the reliability of focus detection is high, the same weighted average processing as in step (306) of FIG. 5 is performed, and the process proceeds to step (409).
Since the processing here is the same as that of the first embodiment, detailed description will be omitted.

In step (408), DF2,
It is expected that one or both focus detection errors of DF3 are large. Therefore, the image plane moving velocity V detected this time is detected.
The error of 1 is also expected to be large. Therefore, in this step, the weighting of the image plane moving velocity V1 detected this time is made small (1/2 in the embodiment), and the weighted average processing is performed by the following equation. When this step is completed, the process proceeds to step (409).

From step (409) to step (42)
The flow up to 0) is the same as the flow from step (307) to step (318) in FIG. 5 in the first embodiment, so the description thereof is omitted here.

As described above, in the second embodiment, since the weighted average processing is performed with weighting taking into account the focus detection accuracy as well, it is more stable in a situation where it is more unstable than in the first embodiment. Predictive control can be performed.

(Third Embodiment) In the third embodiment of the present invention, the weighted average is determined depending on whether the reliability of the defocus amount detected last time or this time is high, only one is high, or both are low. The weighting of is changed.

FIG. 7 is a flow chart showing a "prediction calculation" subroutine in the third embodiment of the present invention. Steps (501) to (505) are the same as steps (301) in FIG. 5 in the first embodiment. Since it is the same as the step (305), the description is omitted here. Also, from step (512) to step (52
The flow up to 4) is also the same as the flow from step (306) to step (318) in FIG.

From step (506) to step (51
Up to 1), weighting is set according to the focus detection state.

That is, in step (506),
The reliability of detection accuracy at the time of focus detection of the previously stored defocus amount DF2 is evaluated, and if the reliability is high, the process proceeds to step (507), and if not, the step (5
08).

In step (507), the reliability of the detection accuracy of the defocus amount DF3 detected this time is evaluated. If the reliability is high, the process proceeds to step (509), and if not, the process proceeds to step (510). Transition.

In step (508), the same determination as in step (507) is performed, and the defocus amount DF3
If the reliability of is high, the process proceeds to step (510), and if not, the process proceeds to step (511).

That is, if the reliability of the defocus amount detected last time and this time is both high, step (50
If the reliability of either one is high, the process proceeds to step (510), and if the reliability of both is low, the process proceeds to step (511).

At step (509), since the focus detection accuracy is high, the smallest value of TX is set as "TX = 500 msec" in order to increase the weighting of the image plane moving speed obtained this time. ing.

In step (510), since the reliability of the focus detection accuracy is slightly lowered, "TX = 75" is set in order to slightly lower the weighting of the image plane moving speed obtained this time.
The value of TX is set a little larger as "0 msec".

In step (511), since the focus detection accuracy is the worst, in order to further reduce the weighting of the image plane moving speed obtained this time, the value of TX is set to be larger with "TX = 1000 msec".

As described above, by changing the set value of TX and controlling the weighting of the weighted average depending on the reliability of focus detection accuracy, there is a possibility that unstable control using unreliable data is performed. In this case, the weighting of the weighted average is changed so that stable prediction control can be performed, and in the case of highly reliable data, the weighting of highly reliable data can be increased so that predictive control with good response can be performed. It is configured in.

In this embodiment, the set values of TX are "500 msec", "750 msec", "1000".
However, this is a typical value only, and values other than these values may be used, and these values may be changed depending on the shooting state.

According to each of the embodiments described above, the reliability of the image plane moving speed is evaluated by the items that affect the reliability of the focus detecting accuracy and the detection accuracy of the image plane moving speed such as the focus detection time interval. Since the weighted average weighting is changed based on the result, control that emphasizes stability is performed when there is a high possibility that unstable predictive control is likely to occur, and control that emphasizes followability when reliable data is obtained. Can be easily achieved.

Although the reliability of focus detection and the focus detection time interval are used as parameters, other parameters, for example, when the deterioration of the lens driving accuracy such as inversion of the lens driving direction is likely to occur, or the focus detection is performed. When various errors are likely to occur, such as zooming during or during lens driving,
It is clear that the present invention is effective even if it is configured to perform the weighted average processing with emphasis on stability.

[0134]

As described above, according to the present invention,
Based on the defocus amount and the time interval information of focus detection obtained a plurality of times in the past by the focus detection means, the image plane position change of the subject after a predetermined time is obtained by approximating the determined function, and the change of the subject after the predetermined time is performed. In the calculating means for controlling the lens driving means so as to match the image plane position with the image plane position of the photographing lens, a function approximated by a weighted average considering the reliability of the change speed of the image plane position obtained by the focus detection is calculated. Setting means for setting is provided, and in order to evaluate the reliability of the change speed of the image plane position, the time interval at which the detected focus detection is performed and the focus detection accuracy are used.

Therefore, it is possible to improve the stability and accuracy of the tracking correction of the subject.

[Brief description of drawings]

FIG. 1 is a block diagram showing a schematic configuration of an automatic focus adjustment device for a camera according to a first embodiment of the present invention.

2 is a flowchart showing the overall operation of the camera of FIG.

FIG. 3 is a flowchart showing details of an “image signal input” subroutine of FIG.

FIG. 4 is a flowchart showing details of a “lens driving” subroutine of FIG.

5 is a flowchart showing details of a "prediction calculation" subroutine in FIG.

FIG. 6 is a flowchart showing details of a “prediction calculation” subroutine in the second embodiment of the present invention.

FIG. 7 is a flow chart showing details of a “prediction calculation” subroutine in the third embodiment of the present invention.

FIG. 8 is a diagram for explaining a prediction principle of conventional follow-up prediction control.

FIG. 9 is a diagram for explaining problems in conventional predictive control.

FIG. 10 is a diagram for explaining a problem of the conventional weighted average processing.

[Explanation of symbols]

 PRS computer SNS sensor device SDR drive circuit LCM lens communication buffer circuit LPRS in-lens control circuit LMTR focus adjustment motor FLNS lens

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁵ Identification code Internal reference number FI Technical display location H04N 5/232 HC

Claims (4)

[Claims] 1. A focus detecting means for repeatedly detecting a defocus amount of a photographing lens, a lens driving means for driving a photographing lens each time a defocus amount detecting operation is performed, and a plurality of past determinations by the focus detecting means. Based on the obtained defocus amount and focus detection time interval information, the change in the image plane position of the subject after a predetermined time is obtained by approximating the determined function, and the image plane position of the subject and the image plane of the photographing lens after the predetermined time In an automatic focus adjusting device for a camera, which comprises a calculating means for controlling the lens driving means so as to match the positions , a weighting considering the reliability of a changing speed of an image plane position obtained by focus detection in the calculating means. An automatic focus adjusting device for a camera, characterized by comprising setting means for setting an approximate function by an average. 2. The setting means is means for performing a weighted average of the changing speeds of the image plane positions weighted by the focus detection time intervals as a weighted average considering reliability. Automatic focus adjustment device for cameras. 3. The camera according to claim 1, wherein the setting means is means for performing a weighted average of the changing speeds of the image plane positions weighted by focus detection accuracy as a weighted average considering reliability. Automatic focus adjustment device. 4. The setting means is means for performing a weighted average of the changing speeds of the image plane position weighted by the focus detection time interval and the focus detection accuracy, as a weighted average considering reliability. Item 1. The automatic focus adjustment device for a camera according to item 1.