JP2809117B2 - 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 an automatic focusing device such as a camera.

[0002]

2. Description of the Related Art Conventionally, the amount of defocus of a photographing lens is repeatedly detected, and it is determined whether or not the subject is moving based on the current and past defocus amounts. The amount of correction for correcting the position of the photographing lens with respect to the moving subject is determined based on the amount of defocus, and the amount of drive of the photographing lens is further determined based on the amount obtained by adding the amount of correction to the defocus amount. 2. Description of the Related Art There is known a technique for driving a photographing lens without delay with respect to a moving subject by driving a lens, a technique called so-called tracking or following drive. For example, the above technique is disclosed in Japanese Patent Application Laid-Open No. 60-214325 by the present applicant.

[0003]

The prior art as described above has the following disadvantages. When determining whether or not a subject has moved based on a focus detection signal having low reliability, an erroneous determination may be performed. Also,
In calculating the lens drive amount for tracking the movement of the subject, the error in the lens drive amount was large, and a focused photograph could not be taken. When switching between tracking drive and normal drive according to the presence or absence of subject movement, the determination of subject movement was performed uniformly, so depending on conditions, tracking drive would be excessively performed,
There are drawbacks in that the drive becomes unstable and that the tracking of the subject cannot be tracked because it is difficult to enter the tracking drive. In particular, when a conventional automatic focus detection device is incorporated in a camera capable of continuous shooting, focus detection is performed during shooting during continuous shooting, and the focus detection cycle changes suddenly before and during continuous shooting. Therefore, the determination of the movement of the subject was unstable.

[0004]

[0005]

[0006]

According to the present invention, a focus adjustment state of a subject image formed by a main optical system is repeatedly detected, and a focus detection signal is generated in time series according to the detected focus adjustment state. Focus information detecting means for detecting, based on the plurality of focus detection signals generated in time series, moving information detecting means for detecting moving information on a temporal change of a subject image plane accompanying the movement of the subject, and a predetermined value for the moving information. A subject movement determining means for comparing the presence or absence of a subject movement; and a driving means for driving the main optical system by a drive control method according to the determination result of the subject movement determining means, wherein the subject movement determining means , The predetermined value is changed.

[0007]

In the present invention, the predetermined value for determining whether or not the subject has moved is changed according to conditions (characteristics of the optical system, reliability, the result of the previous determination, the number of frames during continuous shooting, etc.).

[0008]

【Example】

(First Embodiment) FIG. 1 shows an embodiment in which the present invention is applied to an interchangeable lens type single-lens reflex camera.
The interchangeable lens 10 can be mounted detachably. In a state where the lens 20 is mounted, the photographic light beam coming from the subject
The light is partially reflected by a main mirror 21 provided on the camera body 20 through the camera body 20 and guided to a finder (not shown). At the same time, another part of the photographing light beam is transmitted through the main mirror 21 and reflected by the sub-mirror 22, thereby being guided to the autofocus module 23 (hereinafter, referred to as an AF module) as a light beam for focus detection. FIG. 2 shows a configuration example of the AF module 23. In FIG. 2, the AF module includes a focus detection optical system 24 including a field lens 27 and a pair of re-imaging lenses 28A and 28B and a CCD (charge coupled device) 25 having a pair of light receiving units 29A and 29B. .

In the above configuration, the photographing lens 11
A light beam passing through a pair of regions 18A and 18B symmetric with respect to the optical axis 17 included in the exit pupil 16 of the
A primary image is formed in the vicinity, and a pair of secondary images is formed on a pair of light receiving portions of the CCD 25 by the field lens 27 and the re-imaging photographing lenses 28A and 28B. When the primary image coincides with a film conjugate plane (not shown), the relative position of the pair of secondary images in the direction in which the light receiving sections are arranged on the CCD 25 has a predetermined value determined by the configuration of the focus detection optical system. The pair of light receiving units 29A and 29B are respectively n light receiving elements ai and b.
i (i = 1 to n) and arranged so that the outputs of the corresponding light receiving elements (a ₁ and b ₁ , a ₂ and b ₂ ...) when the primary image coincides with the film conjugate plane are equal. I have. When the primary image is formed on a plane shifted from the film conjugate plane, the relative position of the pair of secondary images on the CCD 25 is determined by the shift direction of the primary image in the optical axis direction (that is, the front focus or the rear focus). The value is changed from the predetermined value in the case of the coincidence. For example, in the case of the front focus, the positional relationship between the pair of secondary images is relatively widened, and in the case of the rear focus, it is narrowed. Light receiving unit 29
The light receiving elements ai and bi forming A and 29B are constituted by charge storage elements such as fort diodes,
By performing the charge accumulation for the charge accumulation time corresponding to the illuminance on the CCD 25, the output of the light receiving element can be set to an output level suitable for processing described later.

Returning to FIG. 1, the description will be continued. The sensor control means 26 receives the charge accumulation start and end commands from the port P4 of the AFCPU 30 and controls the charge accumulation start and end of the CCD 25 by supplying a control signal corresponding to the command to the CCD 25, and also outputs a transfer clock signal to the CC 25.
D25 to the light receiving element output signal
Transfer to In addition, generation of a light receiving element output signal and synchronization signal
The signal is sent to the port P4 of the FCPU 30, and the AF CPU 30 synchronizes with the generated synchronizing signal by the built-in AD conversion means.
After the D conversion is started, the output of the light receiving element is sampled and A / D converted at the port P3 every cycle time of the transfer clock, and the A / D conversion data (2n) corresponding to the number of the light receiving elements is obtained.
G), a well-known focus detection calculation described later is performed based on the data to obtain a defocus amount between the first image and the film conjugate plane. The AF CPU 30 controls the display mode of the AF display means 40 using the port P5 based on the focus detection calculation result. For example, in the case of a front focus, a triangle display unit 41, in the case of a rear focus, a triangle display unit 43, and in the case of focus, a circle display unit 4
2. If the focus cannot be detected, the AF CPU 30 controls the cross display unit 44 to be activated. AFC
The PU 30 controls the AF motor 50 based on the focus detection calculation result.
The driving direction and the driving amount are controlled to move the photographing lens 11 to the focal point. First, the AF CPU 30 changes the port P2 to A according to the sign of the defocus amount (front focus, rear focus).
A drive signal for rotating the F motor 50 in a direction in which the taking lens 11 approaches the focal point is generated. The rotational motion of the AF motor is transmitted to a body-side coupling 53 provided on a mount portion of the body 20 and the lens 10 via a body transmission system 51 including a gear and the like built in the body 20.

The rotational motion transmitted to the coupling 53 on the body side is further transmitted to the lens transmission system 12 composed of a coupling 14 on the lens side and a gear built in the lens 10 to be fitted thereto. Shooting lens 11
Moves in the focusing direction. The amount of driving of the AF motor 50 is converted into the number of pulses by an encoder 52 constituted by a photo interrupter or the like, and the amount of rotation of the gear or the like of the body transmission system 51 is fed back to the AF CPU 30 from the port P1. The AFCPU 30 has a body transmission system 51
By controlling the driving amount of the AF motor 50, that is, the number of pulses fed back from the encoder 52, according to parameters such as the reduction ratio of the lens transmission system 12, the photographing lens 11 can be moved by a predetermined moving amount. AFC
The PU 30 has a built-in pulse counter for counting the number of pulses input from the port P1 and a comparison register for comparing the contents of the pulse counter with the pulse counter. It has a function that includes

The AF CPU 30 executes the AF in the following order.
The driving amount of the motor 50 is controlled. First, before the driving of the AF motor 50 is started, the contents of the pulse counter are cleared and the desired number of pulses is set in the comparison register. Next, the AF motor 5
0 is started. The rotation of the AF motor 50 causes the encoder 52 to generate a pulse and count up to a pulse counter. When the content of the pulse counter matches the comparison register, an interrupt occurs and the AF CPU executes A
Stop the F motor. In this way, the AF motor is driven and controlled by a desired number of pulses. Further, the AFCPU 30 has a built-in timer for measuring time, and also has a timer interrupt function of interrupting at regular intervals. AF
As described above, the CPU 30 mainly has a function of controlling the AF operation.

A main CP for mainly controlling a camera sequence exposure operation (AE) is provided inside the body 20.
There is U70. The main CPU 70 inputs information on exposure such as subject brightness, film sensitivity, aperture value, shutter speed and the like from the AE information means 85 from the port Q12,
An aperture value, a shutter speed, and the like are determined based on the E information. The main CPU 70 displays information such as the determined aperture value and shutter speed on the display unit 86 through the port Q13, and sets the information as the aperture value and shutter speed in the shooting operation. The main CPU 70 controls the port Q8 in the photographing operation.
The mirror control means 81 controls the up and down operations of the main mirror 21. Further, the control unit controls the diaphragm control means 83 through the port Q10 to control the diaphragm mechanism in the lens 10 (not shown). The port Q9 operates the shutter control means 82 to control a shutter mechanism (not shown). Upon completion of the photographing operation, the main CPU 70 controls the hoisting charge control means 84 through Q11 to operate a hoisting charging mechanism (not shown) in preparation for the next photographing operation. The above is the outline of the operation of the main CPU 70.

The lens 10 has a lens CPU 13 built therein. The lens CPU 13 includes AE-related information such as an open F value required for the main CPU 70 and a coupling 14 per unit movement amount of the photographing lens 11 required for the AF CPU 30. The AF-related information such as the number of rotations is transmitted to the body-side communication bus 64 via the lens-side contact 15 provided with the mount unit and the body-side contact 63. AFCPU 30 is lens C
The AF-related information from the PU 13 is received from the port P6 connected to the communication bus 64. Further, the main CPU 70 receives the AF-related information from the lens CPU 13 from the port Q1 connected to the communication bus 64. The main CPU 70 and the AF CPU 30 are connected to the port Q1 via the communication bus 64, respectively.
And P6, it is possible to input / output various information to / from each other. There are also directly connected input / output signal (IO signal) lines other than the communication bus 64 between the main CPU 70 and the AF CPU 30.

AF is an AF permission signal, and the main CPU 7
The data is sent from the port Q2 of 0 to the port P7 of the AFCPU 30. When the AF permission signal (AF) is ON (hereinafter, ON), the AF CPU 30 permits the AF motor 50 to be driven,
Driving is prohibited when it is OFF. The AF permission signal (AF)
Hoisting charge control of main CPU 70 and AFCPU 3
The AF motor drive of 0 is performed at the same time, and is used for the purpose of preventing a problem from occurring beyond the power supply capability of a power source such as a battery (not shown). That is, the main CPU 30 turns off the AF permission signal (hereinafter referred to as O) during the winding charging operation.
As FF), the driving of the AF motor 50 of the AF CPU 30 is prohibited to prevent the winding charging operation and the AF motor driving from being performed at the same time.

MR is a mirror up signal.
From port Q3 of PU70 to port P8 of AFCPU30
Sent to When the mirror up signal (MR) is ON, it indicates that the mirror is up and the transition between up and down is OFF.
Indicates that the mirror is down. Mirror up signal (M
R) is used for adjusting the timing of the drive delay start after the CCD accumulation start mirror of the AFCPU 30 is raised.
RL is a release permission signal, which is sent from the port P9 of the AFCPU 30 to the port Q4 of the main CPU 70.
When the release permission signal (RL) is ON, the main CPU 70
Is permitted, and prohibited when OFF. The release permission signal (RL) is transmitted to the AF
It is used to adjust the timing between the tracking operation control and the photographing operation control of the main CPU 70, and to prohibit the photographing operation before focusing in the one-shot AF mode.

RB is a release button signal, which is a release button 6 which is an external operation member provided on the body 20.
The operation state information of 0 is sent to the port P10 of the AF CPU 30 and the port Q5 of the main CPU 70. When the release button signal (RB) is ON, fully press the release button, OF
F indicates non-full press. Release button signal (RB)
Indicates activation of the photographing operation control of the main CPU 70 and A
Used for controlling the tracking operation of the FCPU 30.

DM is a frame speed mode signal.
The information of the frame speed mode selection state of the frame speed mode selection means 61 which is an external member provided at
11 and port Q6 of the main CPU 70. The frame speed mode indicated by the frame speed mode signal (DM) is C1, C2, S
C1 is a high-speed continuous shooting mode in which the frame speed is prioritized, and while the release button 60 is fully pressed, the shooting operation is immediately shifted to the next shooting operation while the release button 60 is fully pressed. AF operation is hardly performed. And C
Reference numeral 2 denotes a normal continuous shooting mode in which the AF operation is performed at least once between the shooting operation and the next shooting operation while the release button 60 is fully pressed, and the frame speed is slower than the frame speed mode C1. S is a single photographing mode, and the photographing operation is performed only once when the release button 60 is fully pressed.

Reference numeral FM denotes a focus mode signal, which is used to output information of a focus mode selection state of a focus mode selection means 62 which is an external operation member provided on the body 20 to the AF mode.
It is sent to the port P12 of the CPU 30 and the port Q7 of the main CPU 70. There are three types of focus modes represented by the focus mode signal (FM): C, O, and M. C is a continuous AF mode in which the photographing lens 11 is servoed to a focused point based on the always detected defocus amount. is there. O is a one-shot mode in which the photographing lens 1
This mode is a mode in which the servo of the photographing lens 11 is not performed after reaching the focal point of 1. M is a manual mode, in which the servo of the taking lens 11 is not performed and the display means 4
This is a mode in which the focus detection result is displayed only with 0. Table 1 summarizes the IO signals described above.

Next, the relationship between the operations of the AF CPU 30 and the main CPU 70 and the combination of the frame speed mode and the focus mode will be described. When the focus mode is manual (M), the AF CPU 30 does not drive the AF motor 50, so that the AF permission signal (AF) becomes unnecessary for the AF CPU 30. The AF CPU 30 is a main CPU 70
After detecting that the mirror-up signal (MR) is OFF, the accumulation of the CCD is started. Main CPU
70 performs a shooting operation in accordance with the frame speed mode when the shutter button is fully pressed, regardless of the release permission signal (RL) of the AFCPU 30. Focus mode is one-shot AF
In the case of (O), the AF CPU 30 outputs the AF permission signal (AF)
Only when is ON, the AF motor 50 is driven, and at the same time, it is detected that the mirror-up signal (MR) is OFF, and then the accumulation of the CCD is started. Fix the drive. Further, the main CPU 70 issues a release permission signal (R
The shooting operation can be started when L) is ON and the release button signal (RB) is ON. Therefore, when the focus mode is the one-shot AF, the frame speed modes C1 and C2 effectively perform almost the same operation.

When the focus mode is continuous AF (C) and the frame speed mode is C1 or S, the AF CPU 30 drives the AF motor 50 only when the AF permission signal (AF) is ON, and simultaneously turns off the mirror up signal (MR). , The accumulation of the CCD is started. In this case, the display drive is updated even after the photographing lens 11 reaches the focal point. In this case, the main CPU 70 performs the photographing operation according to the frame speed mode C1 or S when the release button signal (RB) is fully pressed regardless of the release permission signal (RL). Therefore, when the focus mode is C and the frame speed mode is C1, the main CPU 70 does not provide a margin time during the photographing operation, and the AF CPU 30
The time during which 0 can be driven is a short time from the completion of the winding charge to the start of the next winding charge.

When the focus mode is C and the frame speed mode is C2
Is a special mode (tracking mode) for a tracking operation optimized for a dynamic subject to be described later.
The CPU 30 performs the AF only when the AF permission signal (AF) is ON.
The motor 50 can be driven and the mirror up signal (MR) is OF
The point at which the accumulation of the CCD is started upon detecting that it has become F is the same as in the above-described mode selection. AF motor 50
Use the tracking algorithm described below when calculating the drive amount of
When it is determined that the subject is a dynamic subject, the driving amount of the AF motor 50 is determined as the sum of the defocus amount and the tracking correction amount.
When 0 is driven, the AF display mode is also changed. AFCP
U30 is the release button signal (RB) in the tracking mode
When the button is fully pressed, the drive time of one time of the AF motor 50 is limited to a predetermined time, and a release permission signal (RL) is turned ON after a predetermined time from the start of driving to adjust the timing of the AF operation and the photographing operation. When the main CPU 70 is in the tracking mode (the focus mode is C and the frame speed mode is C
2) The photographing operation is started when the release permission signal (RL) is ON and the release button signal (RB) is ON.

Table 2 summarizes the relationship between the combination of the frame speed mode and the focus mode and the tracking operation. From Table 2, the tracking mode for determining the drive amount of the AF motor 50 by adding the tracking correction amount to the normal defocus amount for the dynamic subject is selected only when the focus mode is C and the frame speed mode is C2. Will be. The operation of the AF CPU 30 and the main CPU 70 in the tracking mode will be described in more detail with reference to FIGS. FIG. 3 is a diagram showing the relationship between the movements of the subject and the photographing lens when the release button is fully pressed in the tracking mode and the operations of the AF CPU 30 and the main CPU 70. The vertical axis represents the lens position Z, and the horizontal axis represents time t. A solid line L1 is a locus of an ideal position of the photographing lens 11 necessary for always forming a subject image on a film surface when the subject is continuously moving. The alternate long and short dash line L2
Is the locus of the actual shooting lens 11 moving. At time t0 when the main CPU finishes the mirror down operation, the taking lens 11 is stopped and the lens position is at Z0. The AFCPU starts accumulation of the CCD at time t0 and ends the accumulation at time t7. AFCPU starts C from time t7.
A / D conversion of CD data and calculation of focus detection are started. In the tracking mode, when a dynamic object is determined by the tracking algorithm as described later, the photographing lens is driven according to an amount obtained by adding a tracking correction amount to a defocus amount obtained by a focus detection calculation as a static object. Here, a defocus amount corresponding to the difference (ΔZ2 = Z2-Z0) between the solid line L1 and the one-dot chain line L2 and a tracking correction amount (previous time) at time t10, which is a midpoint between time t0 and time t7, are added. The object (tracking defocus amount) is obtained at time t1.

On the other hand, the main CPU starts the charging / winding operation from time t0 and ends at time t2. When the charging and hoisting operation is completed at time t2, the AFCPU starts motor driving, sets the tracking defocus amount as a new tracking correction amount, and adds this to the defocus amount (ΔZ1 = Z1).
The photographing lens 11 is moved by -Z0). Further, the main C is shifted from a time t4 after a predetermined time from the motor drive start time t2.
The PU starts a mirror-up operation. The AFCPU forcibly ends the motor driving at time t5, which is a predetermined time after the motor driving start time t2. The main CPU ends the mirror-up operation at time t8, which is a predetermined time after time t4, and starts the shutter operation. Then, at time t9, the shutter operation ends, the mirror down operation starts, and at time t6, the mirror down operation ends. The main CPU starts the charge winding operation again at time t6, and the AF CPU starts the next CCD accumulation operation.

As described above, when photographing in the tracking mode,
Since the time for focus detection and the time for driving the motor are always set between the photographing operations, the timing of the shutter operation is close to the end of the motor driving. Therefore, as shown in FIG. You can shoot in places where there is little deviation, and you can get in-focus photos. In the above description, for simplicity, the case where the required tracking defocus amount ΔZ1 has just been driven during the predetermined AF motor drive time (t2 to t5) has been shown. Time t before the end of driving time (t2 to t5)
It is easier to control the drive to end at 5 'or t5 "and to keep the AF motor drive stopped for the remaining time. In any case, the predetermined motor drive time (t2 to t)
The driving of the required tracking defocus amount ΔZ1 during 5) is completed. The length of AF motor drive time is
For example, the time is set to be about 100 ms, for example, so that a defocus amount of 3 to 4 mm can be driven during that time. By keeping the AF motor driving time constant and starting the mirror up at a time t4 after a certain time from the motor driving start time t2, accurate tracking photographing can be performed.

In other words, the cycle time (t0 to t6) becomes constant regardless of the drive amount, and the defocus amount calculated for focus detection is calculated repeatedly in the cycle of this cycle time. The determination of the presence or absence of the movement of the subject, the calculation of the tracking correction amount, and the like can be easily and accurately performed. Furthermore, mirror up is performed at time t4 after a predetermined time has elapsed after motor drive start time t2,
The time interval is set so that the driving of the required driving amount is completed by the time t8 when the shutter operation starts after the mirror-up is completed and the exposure starts, so that the time from the start of motor driving to the exposure is always maintained. (T2 to t8) becomes constant, and accurate prediction driving can be performed so that L1 and L2 intersect at the moment of the coming exposure. In short, if the subject is moving and the speed is also various, the above t2
If the time t8 is not constant, a change in the lens drive amount corresponding to the movement of the subject in the time interval of the change must be calculated and corrected by some method, and L1 and L2 intersect at the moment of exposure. Or it will be very difficult to control to match. Therefore, it is important to set the AF motor drive time to a fixed value and to start the mirror up after a fixed time from the start of the motor drive in order to enhance the tracking performance. FIG. 4 is an operation flowchart showing the operation of the AF CPU in the tracking mode in more detail based on the relationship between each internal flag and each IO signal.

The delay flag (DLYFLG) is a flag indicating a motor delay driving state after the mirror is raised. When the switch is ON, the delay drive is being performed, and when the switch is OFF, the delay drive is not performed. The drive state flag (MOVFLG) is a flag indicating the motor drive state, and is ON when the motor is being driven, and is stopped when OFF. The tracking delay flag (PDYFLG) is a flag indicating a motor tracking delay driving state from the start of driving the motor in the tracking mode to the start of mirror up.
In the case of FF, the tracking delay is not driven. The mirror flag (MIRFLG) is a flag indicating a state before and after the mirror-up in the tracking mode.
FF is after mirror up. In FIG. 4, the frame speed mode is C2 and the focus mode is C.
In short, the tracking mode is selected, and the release button signal (RB) is fully pressed (ON). AFCPU
Ends the CCD accumulated focus detection calculation at time t1 (O
FF) and waits for AF permission from the main CPU.

At time t2, when the main CPU completes the hoisting charge and turns the AF permission signal on (ON), the AF
The CPU detects this and starts driving the motor based on the result of the focus detection calculation. At the same time, the drive status flag (MOVF
LG) (ON), the tracking delay flag (PDY)
FLG) during the tracking delay (ON), the mirror flag (M
IRFLG) before mirror-up (ON).
At time t2, the AFCPU starts tracking the tracking delay time (T1), and ends the tracking at time t3.
And a release permission signal (R
L) is permitted (ON). The tracking delay time (T
By providing 1), it is possible to secure a constant AF motor driving time until the mirror-up operation starts. Main C
The PU detects that the release permission signal (RL) has been permitted (ON), starts the mirror-up operation at time t4, and simultaneously raises the mirror-up signal (MR) (O).
N).

The AFCPU detects that the mirror-up signal (MR) is up (ON) and disables (OFF) the release permission signal (RL). At the same time, the mirror flag (MIRFLG) is reset after the mirror is raised (OFF), and the delay flag (DLYFLG) is set to a delay (ON). The AFCPU starts measuring the delay time (T2) at time t4, and when the time measurement ends at time t5, resets the delay flag (DLYFLG) to outside the delay (OFF). Is forcibly terminating the motor drive and stopping the drive state flag (MOVFLG) (O
FF). The AFCPU thereafter waits for the mirror-up signal (MR) to go down (OFF).
When the main CPU finishes a series of mirror-up operation, shutter operation, and mirror-down operation started at time t4, it turns down (OFF) the mirror-up signal (MR) at time t6. The AFCPU detects this and starts the next CCD accumulation operation.

As described above, in the tracking mode, the focus detection and the AF motor driving operation are always performed once during the photographing operation, and the AF motor driving time is the maximum tracking delay time (T1) + driving delay time (T2). , It is possible to sufficiently track a fast-moving subject. The above is an outline of the temporal flow of the operation of the AFCPU in the tracking mode. Next, specific programs and operations of the AF CPU and the main CPU in the embodiment of the present invention will be described. First, the program of the main CPU will be described with reference to the flowcharts of FIGS. The main CPU has a built-in timer and has a timer interrupt function. The program is composed of a main program shown in FIG. 5 and a timer interrupt program shown in FIG. In FIG. 5, the main program first performs initialization in # 100. That is, the IO signal mirror up signal (MR) for the AFCPU is reduced (OFF), and the AF permission signal (AF)
Is permitted (ON). Further, the timer is set so that the timer is interrupted every predetermined time, for example, every 50 ms, and the timer interrupt is permitted.

Next, at step # 105, the control waits until the release button (RB) is fully pressed (ON). Release button (R
When B) is fully pressed (ON), the process proceeds to step # 110 to test whether the focus mode (FM) is the manual mode (M). The process jumps to the photographing operation process after 130. If not manual in # 110, focus mode (FM) is continuous AF (C) in # 115
And if it is not a continuous AF, that is, in the case of one shot, the flow proceeds to # 125. If the continuous AF is performed in # 115, it is tested in # 120 whether the frame speed mode is the normal continuous shooting (C2). If not, the process jumps to # 130. # 12 if the frame speed mode is C2 in # 120
Go to 5. In step # 125, the process waits until the release permission signal (RL) is turned on (ON). In steps # 110 to # 125, the focus mode is one-shot AF or the focus mode is continuous A
AFC only when the frame speed mode is C2, ie, the tracking mode in F
After waiting for the release permission signal from the PU, the process proceeds to the photographing operation process after # 130, and in the other mode setting, the photographing operation process after # 130 is immediately performed.

In step # 130, the mirror-up signal (MR) is turned up (ON), and in step # 135, the aperture control is performed so that the aperture value becomes the target aperture value based on the result of the AE operation performed in the timer interrupt processing described later, and at the same time, the mirror-up is performed. Perform control. In step # 140, the shutter control is performed at the shutter speed obtained by the AE calculation. In step # 145, mirror down control is performed and aperture control is performed to open the aperture. In # 150, the mirror up signal (MR) is turned down (OFF), and then in # 155, the AF permission signal (A
F) is prohibited (OFF), and the winding charge control is performed in # 160. AF is completed at # 165 when the winding charge is completed.
The permission signal (AF) is permitted (ON). In # 170, it is tested whether the frame speed mode is the normal continuous shooting (C2),
If it is C2, the process proceeds to # 175, and after a delay of a predetermined time, returns to # 105 again. If it is not C2 in # 170, the frame speed mode is single (S) in # 180.
Is tested, and if it is single, the process waits until the release button signal (RB) is not fully pressed (OFF) at # 185, and returns to # 105 where it is no longer fully pressed. # 18
If it is not 0 and it is not single, that is, if it is high-speed continuous shooting (C1), the process immediately returns to # 105, and the next shooting operation sequence is repeated. FIG. 6 shows a timer interrupt program of the main CPU. The timer interrupt program is activated at predetermined time intervals (for example, 50 ms) while the main CPU is executing the main program.

First, when the timer is interrupted, the camera communicates with the lens CPU 13 shown in FIG. 1 through the communication bus 64 at # 200, and AE information (set aperture value, focal length, etc.) of the lens.
Take in. Next, at step # 205, AE information (photometric value, film sensitivity, etc.) of the body is collected from the AE information means 85 of FIG. In # 210, the AE information of the lens and the AE of the body
The AE calculation is performed based on the information to determine a target aperture value, a shutter speed, and the like. In # 215, the result obtained by the AE operation is displayed on the display means 86 of FIG. 1, and in # 220, the process returns to the main program. The above is the program operation of the main CPU. Next, the program of the AFCPU will be described. The AFCPU has a built-in memory for storing A / D conversion data of the CCD output, a timer and a pulse counter, and has a timer interrupt function and a pulse counter interrupt function. Table 3 shows the names and meanings of the flags used in the AFCPU program. Table 4 shows the names and contents of data used in the AFCPU program. FIG. 7 shows AF
The program outline of the CPU is shown. The AFCPU program is a main program and two interrupt programs (timer interrupt program and pulse counter interrupt program)
It is composed of The main program is (1) ~
It consists of the module of (11) and has a large loop structure.

In the main program, first, (1) various flags, data, and signals are initialized by the initialization module. Next, (2) the accumulation preprocessing module determines whether or not CCD accumulation can be started. If accumulation is possible (the mirror is down and the AF motor is stopped),
(3) The CCD accumulation control module controls the start, end and accumulation time of CCD accumulation. (4) The CCD output AD conversion module stores the CCD data obtained by AD converting the CCD output in the internal memory. (5) The AF algorithm module performs a predetermined focus detection calculation on the stored data to calculate a defocus amount for a static subject. (6) In the lens information reading module, the lens CPU
And lens AF information necessary for driving the motor and the like. (7) The tracking algorithm module determines whether or not the subject is a dynamic subject. If the subject is determined to be a dynamic subject, the tracking correction amount is added to the defocus amount for the static subject, and the motor drive amount (the tracking Drive amount). (8) The focus determination / display module determines the focus state (whether or not the defocus amount is within the focus zone) and displays the determination result on the AF display means 40 in FIG. (9) The AF permission standby module waits for the AF permission signal (AF) sent from the main CPU to be permitted (ON) in the tracking mode. (10) The drive control module converts the defocus amount into a pulse number, sets the pulse number data in the comparison register, and starts driving the AF motor in the focusing direction. (11) The AGC (auto gain control) calculation module determines the next CCD accumulation time (INTT) based on the CCD data obtained this time so that the next CCD data will have an appropriate value. Return to the preprocessing module. The outline of the main program of the AF CPU has been described above, and the accumulation operation of the CCD and the operation of driving the photographing lens by driving the AF motor are time-independent sequences. (12) The timer interrupt module detects changes in various IO signals, sets and resets flags in accordance with the changes, manages drive delay time, and detects lens ends. (13) In the pulse counter interrupt module, the drive stop processing of the AF motor is performed.

Next, the operation of each module will be described in detail. FIG. 8 is a flowchart of (1) Initialize Module. The AF CPU starts processing from # 230 by turning on or resetting the power. A in # 230
Various flags and data used in the program of the FCPU are initialized. The initial values of the flags and data are as shown in Tables 3 and 4. In Table 4, those whose initial values are blank are those that do not require initialization. And C
The initial value of the storage time (INTT) of the CD is set to a predetermined value IZ (for example, 1 ms). Next, in step # 235, the release permission signal (RL) is prohibited (OFF). This is to prevent the photographing operation from being performed even if the release button is fully pressed immediately after the power is turned on in the one-shot AF mode or the tracking mode. In # 240, the AF display unit 40 of FIG.
Of the display units 41, 42, 43 and 44 are turned off. In # 245, the AF motor is initialized (stopped). In step # 250, a command for sweeping out the electric charge accumulated in the light-receiving unit transfer section of the CCD and for bringing the CCD into an accumulation end state is sent to the sensor control means 26 of FIG. 1 to initialize the CCD. In # 255, the timers built in the AFCPU are set. The timer interruption is performed every predetermined time (for example, every 1 ms). In step # 260, acceptance of the timer interrupt is permitted. AF in # 265
The pulse counter interrupt for stopping the motor drive is prohibited (2) and the process proceeds to the storage preprocessing module.

(2) The description after the storage preprocessing module is as follows.
Because of the loop structure as described above, the description will be made not as an operation when the power is turned on but as an operation when the loop is rotated several times. FIG. 9 shows a flowchart of the (2) storage preprocessing module. In step # 270, a test is performed to determine whether the mode is the tracking mode. If the mode is not the tracking mode, the process jumps to step # 276. If the tracking mode has been set in # 270, it is tested whether or not the mirror-up operation has been completed in # 275. If it is determined that there is, the process proceeds to # 276. # 270 and # 27 above
The process 5 is a process for performing the focus detection calculation and the AF motor driving operation only once during the photographing operation when the release button is fully pressed in the tracking mode as shown in the operation time chart of FIG. is there. Immediately, during the full pressing in the tracking mode, the mirror flag (MIRFLG) is set to ON at the time of starting the AF motor driving, as described later, and the next CCD accumulation start is always the mirror flag (MIRFLG) by mirror up. Is turned off. In the mode other than the tracking mode, if there is enough time, the focus detection calculation and the AF motor drive are performed any number of times during the shooting operation.
Skip 75. The reason why the focus detection calculation is always performed only once during the photographing operation in the tracking mode is as follows. In short, in tracking during continuous shooting, the tracking correction amount is calculated so that the subject is in focus at the moment of exposure,
Such drive control of the lens is as described above.

In order to achieve the above object with high accuracy, it is desirable that exposure, accumulation operation, driving, exposure, accumulation operation, driving... Are repeatedly performed at predetermined time intervals. If the number of times the accumulation calculation is performed between the exposure and the drive is different each time, the cycle time fluctuates, and the process of performing the accurate moving object determination and obtaining the accurate tracking correction amount becomes very complicated or difficult. In # 276, scanning is in progress (SCAFLG
Is ON), and if scanning is in progress, A of # 280
The flow jumps to # 285 without waiting for the stop of the driving of the F motor. This is to permit the scan drive operation of the AF motor and the accumulation operation of the CCD to be performed temporally in parallel only during scanning. If it is determined in step # 276 that the image is not scanned, the process proceeds to step # 280. In step # 280, the process waits for the AF motor to stop (MOVFLG is OFF).
Proceed to 85. This is because the accumulation operation of the CCD and A
This is for separating the driving operation of the F motor temporally. #
At 285, a test is performed to determine whether or not tracking is currently being performed. If tracking is not being performed, the process jumps to step # 305. Tracking (PRSFLG
(ON) means that the subject is determined to be a dynamic subject in the tracking algorithm described later, and the AF motor is driven with the drive amount of the AF motor being (normal drive amount + tracking correction amount). If the subject is being tracked in # 285, the subject is approaching further in # 290 (the tracking drive amount DR
IV <0, that is, the driving direction is the closest direction)
If it is approaching, jump to # 305. If the object is not approaching in # 290, that is, if the subject is moving away, it is tested in # 295 whether the remaining amount of driving (scheduled pulse number ETM-pulse count number ECNT up to the present) is larger than a predetermined amount EX, If not, jump to # 305. If it is determined in step # 295 that the remaining drive amount is large, the flow advances to step # 300 to clear the tracking correction amount (COMP =
0). The tracking correction amount is data used in a tracking algorithm described later. To summarize the processes from # 285 to # 300, the process is to clear the tracking correction amount when tracking is being performed, the subject is moving away, and the remaining driving amount when the AF motor is stopped is large.

FIG. 1 shows the reason for performing such processing.
Explanation will be made using 0. In FIG. 10, the solid line indicates the ideal locus of the lens position of the photographing lens 11 for always forming the subject image on the film surface when the subject is moving away at a constant speed. The trajectory of the movement of the taking lens in the case of pressing is shown. At time t0, the position of the photographic lens 11 stops at Z0 and the accumulation of the CCD is started. At time t1 at the midpoint of the accumulation time, the solid line and the dashed line almost intersect, and the defocus amount for the static subject becomes almost zero and tracking is performed. The tracking drive amount (Z2) determined as a dynamic subject by the algorithm and added with the tracking correction amount
−Z0) is considered when driving is started from time t2. In the case of full pressing during tracking, as described in the operation time chart of FIG. 4, the entire driving time of the AF motor is limited to a predetermined time (approximately T1 + T2), so that the tracking drive amount is large and a predetermined time after time t2. If the predetermined lens position Z2 has not been reached by the time t3, the driving of the AF motor is forcibly terminated at the time t3 lens position Z1. Assuming that the CCD accumulation is started again at time t3 and the position of the solid line is Z3 at time t4 at the midpoint of the accumulation time, the defocus amount for the static subject is (Z3
−Z1). On the other hand, when it is determined again as a dynamic subject in the tracking algorithm, the tracking drive amount to which the same tracking correction amount as the previous time is added is (Z4-Z3).
When the driving is started at time t5, the driving time is much longer than the solid line as shown by the broken line. However, if it is determined at time t5 that the object is not a dynamic object, the lens is driven from time t5 by a drive amount (Z3-Z1) corresponding to the defocus amount for the static object, and the lens position Z at time t6.
Since it reaches 3, it does not go too far from the solid line.

Therefore, in the processes from # 285 to # 300 in FIG. 9, when the remaining amount of the driving amount becomes large during tracking (FIG. 10).
In the case where the AF motor is forcibly stopped at time t3), the tracking correction amount is cleared. In the tracking algorithm described below, if the tracking correction amount is cleared during tracking, the dynamic subject is not determined at that time. After the time t5, the subject becomes out of tracking, and the photographing lens can be moved along the locus of the dashed line. The moving direction of the subject is tested at # 290 in FIG. 9. The tracking correction amount is not cleared during the approach. When the subject approaches at a constant speed, the ideal lens movement increases as the approach approaches. This is because there are few problems such as excessive shooting lens movement. On the other hand, when the subject moves away, the ideal movement of the lens decreases as the distance increases, but the above-mentioned excessive movement poses a problem. Of course, the approaching approach may be the same as the approaching approach. The predetermined amount EX of # 295 can be determined to be a constant amount by an experiment, or can be changed according to various conditions (lens focal length, focus detection cycle time = focus detection time + driving time, etc.).

Returning to FIG. 9, the (2) accumulation preprocessing module will be described again. In step # 305, the main CPU waits for the mirror-up signal (MR) to go down (OFF) in preparation for the next (3) CCD accumulation control module. FIG. 11 is a flowchart of (3) the CCD accumulation control module. Before proceeding to # 320, it is confirmed that the drive of the motor is stopped except during scanning, and it is further confirmed that the mirror is down and the CCD can be accumulated. In step # 320, the sensor control means 26 shown in FIG.
A storage start command of the CD is issued to start the storage of the CCD.
In # 325, except for the first CCD accumulation, (11) described later.
Storage time (INT) determined by the AGC operation module
Time T). In the case of the first CCD accumulation, (1) the accumulation time (I) initially set by the initialization module
NTT = IZ). The time is measured by a timer built in the AFCPU or a timer created on software. When the measurement of the accumulation time in # 325 is completed, # 330
Then, a CCD accumulation end command is issued to the sensor control means 26 to terminate the accumulation of the CCD, and the process proceeds to (4) CCD output AD conversion module.

FIG. 12 is a flowchart of the (4) CCD output AD conversion module. In step # 340, the AF CPU starts AD conversion of the CCD output sent from the CCD 25 in synchronization with the CCD output synchronization signal sent from the sensor control means 26. In step # 345, the CCD output is AD-converted a predetermined number of times (2n times) in synchronism with the CCD output transfer block sent from the sensor control means 26.
Store the data in internal memory. Here, a pair of CCD data is A (1) to A (n) B (1) to B (n) and A (1)
A (2) and B (2)... A (n) and B (n) are output data of the corresponding light receiving elements of the pair of light receiving units 29A and 29B in FIG. When the storage of the CCD data is completed, the process proceeds to the next (5) AF algorithm module. FIG. 13 is a flowchart of (5) the algorithm module.
In step # 360, the known CCD disclosed in Japanese Patent Application Laid-Open No. 60-37513 according to the present application uses 2n CCD data A (1) to A (n) B (1) to B (n) stored in the internal memory. By performing a correlation operation, a relative lateral shift amount (SHIFT) of the pair of subject images on the CCD 25 in FIG. 2 and a parameter (SLOP) indicating the reliability of the obtained lateral shift amount are obtained. FIG.
4, a well-known correlation calculation will be briefly described with reference to FIG. First, the correlation operation of equation (1) is performed to obtain the correlation amount C (L) between the CCD data.

(Equation 1) However, in the equation (1), L is an integer and is a relative shift amount in a case where a light receiving element pitch of a pair of CCD data is used as a unit. Further, it is assumed that the integration operation of the expression (1) is executed in a range where CCD data exists.

The calculation result of the equation (1) indicates that the correlation between a pair of CCD data is as shown in FIG. At a high shift amount L, the correlation amount C (L) is minimized. However, the relative shift amount L is actually equal to the light receiving units 29A, 29A.
Since it is determined based on data discretely obtained from the light receiving elements constituting B, the correlation amount C (L) also becomes discrete. Therefore, the minimum prime value C (L) _{MIN of} the correlation amount C (L) is not always directly obtained from the correlation amount C (L) obtained by the calculation. Therefore, the minimum value C (L) _MIN of the correlation amount C (L) is obtained by using the three-point interpolation method shown in FIG. That is, if the minimum value of the discretely obtained correlation amount C (L) is obtained when the relative shift amount L is L = x,
Correlation amounts C (L) corresponding to the relative shift amounts x-1 and x + 1 before and after that are C (x-1), C (x), and C (x + 1).
become. Therefore, first, a straight line H connecting the minimum correlation amount C (x) and the larger correlation amount (C (x + 1) in FIG. 9) among the remaining two correlation amounts C (x−1) and C (x + 1) is drawn. ,
Then, a straight line J passing through the remaining correlation amount C (x-1) and having a slope opposite to the straight line H is drawn to obtain an intersection W between these two straight lines H and J. The coordinates of this intersection W are the relative shift amount x
_m and the correlation amount C (x _m ), and the coordinates can represent the minimum relative shift amount x _m and the minimum correlation amount C (x _m ) in the continuous relative shift amount. . If such a three-point interpolation method is expressed by an arithmetic expression, the minimum relative cyst amount x _m is given by the following expression.

(Equation 2) And the correlation amount C
(X _m ) is

(Equation 3) Can be expressed as

Here, in the expressions (2-1) and (2-2), D is a relative shift amount... X-1, x, x + 1.
・ The following equation is the deviation between each data

(Equation 4) Can be expressed as Also, (2-1) and (2-
In the equation (2), SLOP is a relative shift amount x-1, x,
Correlation amounts C (x-1), C (x), C corresponding to x + 1
The larger deviation among the deviations between (x + 1) is represented by the following expression: SLOP = MAX (C (x + 1) -C (x), C (x-1) -C (x)) (4) Can be represented.

The arithmetic expressions represented by the expressions (1) to (4) are as follows:
The relative shift amount x _m represents the relative shift amount of a pair of CCD data, and if the pitch of the light receiving elements is y, the CCD 25
Relative lateral shift SH between two subject images formed on the top
IFT can be expressed as SHIFT = y × x _m (5). The defocus amount DEF on the focal plane can be expressed by the following equation: DEF = KX × SHIFT (6) Here, KX is a coefficient determined by the configuration conditions of the focus detection optical system shown in FIG. The larger the value of the parameter SLOP obtained by the equation (4), the deeper the depression of the correlation amount C (L) shown in FIG. 14, that is, the larger the correlation. Therefore, the reliability of the obtained defocus amount DEF is high. It is shown that. Returning to FIG. 13 again, the description will be continued. In step # 360, the shift amount (SHIFT) and the reliability (SLOP) are obtained as described above. In # 365, it is tested whether or not the shift amount (SHIFT) has been obtained. That is, if no dent is found even if the shift amount L is shifted to the maximum shift amount (5 in the figure) in FIG. 14, the shift amount (SHIFT) cannot be obtained. If the shift amount (SHIFT) is not determined in # 365, the process proceeds to # 385. When the shift amount is obtained in # 365, whether the reliability of the defocus amount (DEF) obtained in # 370 is reliable (SLOP is equal to or more than a predetermined value SX) or not is determined. Go to # 385. If it is determined that there is reliability in # 370, the low contrast flag (LOCF) is determined in # 375.
LG) is reset (OFF) and focus detection is not disabled, and in # 380, the obtained shift amount (SHIF) is determined.
From (T), the defocus amount (DEF) is obtained by equation (6), and the process proceeds to the next (6) lens information reading module.
On the other hand, if it is determined in step # 365 that the shift amount has not been obtained or if it is determined in step # 370 that there is no reliability, the process proceeds to step # 385, and the low contrast flag (LOCFLG) is set.
Is set (ON), it is determined that the focus cannot be detected, and the process proceeds to the next (6) lens information reading module.

FIG. 16 is a flowchart of the (6) lens information reading module. In step # 390, communication with the lens CPU 13 is performed through the communication bus 64 of FIG.
The lens AF information required by U is fetched and stored in the internal memory. For example, the lens CPU 13 sends data such as a pulse conversion coefficient KL required when converting the defocus amount (DEF) to the number of pulses, a focal length FL of the lens, and information on whether or not the lens is AF-capable from the lens CPU 13 to the AF CPU. In step # 395, it is tested whether the attached lens is an AF lens (an AF-enabled lens) based on the acquired lens information. If it is determined that the lens is an AF lens, in step # 405, an AF lens flag (AFLFLG) is set (ON). )
Then, proceed to (7) tracking algorithm. If it is determined that the AF lens is not the AF lens in # 395, the AF lens flag is reset (OFF) in # 400, and the process proceeds to the next (7) tracking algorithm.

FIG. 17 is a flowchart of the tracking algorithm module (7). Blocks # 410 to # 425 are used to count the number of shots while the release button is fully pressed. In # 410, a test is performed to determine whether the relay button signal (RB) is fully pressed (ON).
If it is not fully pressed, the shot counter (PC
ENT) is cleared to 0, and the flow advances to # 430. # 4
If it is determined that the number of shots has been fully pressed in step # 10, it is further tested in step # 415 whether the number of shots is less than 3 (PCOUNT <3). Proceed to. If it is less than 3 in # 415, the process proceeds to # 420, and 1 is added to the number of shots (PCOUNT = PCOUNT + 1). The data of the shot counter (PCOUNT) is used only in the tracking mode for tracking determination and the like described later. In the tracking mode, during full press, CCD accumulation and focus detection are always ensured once during the shooting operation, and once during the shooting operation (7)
Since the tracking algorithm module is executed, the number of shots is counted by the blocks # 410 to # 425 described above. Blocks # 430 to # 515 are blocks for determining whether or not to perform the tracking operation in the tracking mode.

Table 5 summarizes the conditions for performing the tracking operation. Hereinafter, the determination of the tracking operation will be described in order.
In # 430, is the tracking mode currently set (PMDFLG
ON) Test. Note that the tracking mode flag (PMDFL
G) is updated by periodically examining the combination of the focus mode and the frame speed mode in the timer interrupt processing described later. If the mode is not the tracking mode in # 430, the process proceeds to # 545 without performing the tracking operation. If the tracking mode is set in # 430, whether focus detection is impossible in # 435 (LOCF
LG ON) Test. If the focus cannot be detected, the process proceeds to step # 545 without performing the tracking operation. If the focus detection is possible in step # 435, the flow advances to step # 440 to determine whether the obtained defocus amount (DEF) is reliable, that is, the parameter SLOP representing the reliability obtained by equation (4) is equal to the predetermined value S.
Test for Z or greater. Of course, the predetermined value SZ is # 3
65 has a value larger than the predetermined value SX used. # 44
When it is determined that the reliability is not 0 and the tracking operation is not performed, the process proceeds to # 545. The reason for making such a determination is that if there is no reliability, the obtained defocus amount DEF also contains a large amount of error, and if the tracking operation described below is performed as it is, the photographing lens will be unstable even for a static subject ( This is to prevent such unstable operation from occurring since hunting occurs. In step # 440, the reliability is determined based on the value of the parameter SLOP obtained by the equation (4). However, the present invention is not limited to this, and any process that can determine the reliability may be used. For example, the reliability may be determined by comparing the magnitude of the contrast information CONT obtained by Expression (7) with a predetermined value.

(Equation 5) In the equation (7), A (i) is CCD data, and l is a predetermined integer. If it is determined in step # 440 that there is reliability, the process proceeds to step # 445, and scanning is being performed (SCAFLG O
N) or test. If it is determined in step # 445 that a scan is being performed, the process proceeds to step # 545 without performing the tracking operation.

The defocus amount obtained during scanning is a defocus amount obtained based on the CCD data obtained by accumulating the CCD while the photographing lens is being driven. Therefore, the defocus amount includes many errors. Performing the operation tends to cause unstable operation. # 445 is a determination for preventing such a problem. If the scanning is not being performed in # 445, the flow proceeds to # 450 to test whether the previous driving was performed (DRVFLG is ON), and if the driving was not performed, the control jumps to # 545 without performing the tracking operation. The reason for this is that the tracking operation is an operation that adds a tracking correction amount based on the assumption that the shooting lens has moved, as described later, so if the tracking operation is performed immediately from the state where the shooting lens is stationary, the tracking correction will be performed. This is because the operation does not work well and the operation becomes unstable. Therefore, the normal drive (drive without tracking correction) is always performed once during the transition from the state where the photographing lens is stationary to the tracking operation, and unstable operation can be prevented. If it is determined in # 450 that the previous driving has been performed, the process proceeds to # 455, and the first driving (REVFLG becomes O) after the previous driving has reversed the driving direction.
N).

The drive inversion plug (REVFLG) is a flag that is set when the drive direction is inverted in the drive control module (10) described later. If it is determined that the previous driving is the first driving after the driving direction is reversed, the control jumps to # 545 without performing the tracking operation. The reason for branching at # 455 will be described with reference to FIG. FIG.
FIG. 8 is a diagram illustrating the movement of the photographing lens when there is no determination at 455, and a solid line indicates a position (focusing position) of the photographing lens for making a subject image coincide with a film surface with respect to a stationary subject. The two-dot chain line indicates the actual movement of the photographing lens. When the focus is approached by the drive D0 from the defocused position of the photographing lens and stopped by passing through the focal point due to an error, the defocus amount obtained at this position was DEF0. Next, drive D1 is performed from this position toward the in-focus position based on the defocus amount DEF0.
However, since this drive D1 is the first drive after inversion,
Due to the backlash of the body transmission system 51 and the lens transmission system 12 of FIG. 1, as shown by the broken line, the lens is not driven to the in-focus position, and stops at a position separated by the backlash from the in-focus position.
When the defocus amount obtained at this position becomes DEF1,
Since the in-focus position has not been reached despite the previous drive, the tracking operation is started, and the drive amount of the next drive D2 is equivalent to twice the defocus amount DEF1, and has passed the in-focus position. I will. After that, the same operation is repeated, and hunting occurs near the in-focus position.

On the other hand, in the case of FIG. 18B, since the tracking operation is not performed in the second driving D2 after the reversal, the driving amount corresponds to the defocus amount DEF1 and reaches the in-focus position. Can be. In the above description, when the driving direction of the photographing lens is reversed, the two driving operations after the reversal are prohibited from performing the tracking operation. However, the driving operation is not limited to two times and may be any predetermined number of times. In addition, if a predetermined amount or more of drive is performed in the first drive after the inversion, the tracking operation may be permitted in the second drive, or the tracking may be performed if the accumulated drive amount after the inversion becomes a predetermined amount or more. The operation may be permitted. The tracking operation may be prohibited for a predetermined time after the reversal. As explained above #
The process 455 is a process for preventing unstable operation due to backlash at the time of drive inversion. If it is determined in # 455 that the previous drive is not the first drive after the drive direction has been reversed, the flow proceeds to # 460 to test whether the vehicle is currently being tracked (PRSFLG is ON). The tracking-in-progress flag is set (ON) when a tracking operation is performed as described later, that is, when it is determined that the object is a dynamic object and driving is performed by a tracking drive amount obtained by adding a tracking correction amount to a defocus amount for a static object. . If tracking is not being performed in # 460, the tracking correction amount (COMP) is cleared to 0 in # 465, and the process proceeds to # 480.
If it is determined in step # 460 that tracking is being performed, the flow advances to step # 470 to test whether the sign of the defocus amount (DEF) obtained this time and the sign of the previous tracking defocus amount (PLST) are the same. If it is the same sign, proceed to # 480,
If the signs are different, the process proceeds to step # 475 to test whether the absolute value (| DEF |) of the defocus amount is larger than the predetermined value DX. If it is determined in step # 475 that it is large, the process jumps to step # 545 without performing the tracking operation. If it is determined in step # 475 that the size is small, the process proceeds to step # 480. # 47
The processes from 0 to # 475 are processes for speeding up the response at the end of the tracking operation, which will be described with reference to FIG.

In FIG. 19, the solid line is an ideal locus of the photographing lens position for making the subject image coincide with the film surface with respect to the subject, and the one-dot chain line is the actual locus of the movement of the photographing lens. In the tracking operation before the release is performed, tracking correction is performed so that the defocus amount (DEF), which is the focus detection result, becomes 0 with respect to the dynamic subject, as described later, and driving is performed. Therefore, the operation is performed so that the solid line and the dashed line intersect at the middle point of the accumulation time of the CCD. (The drawing is drawn assuming that the accumulation has been performed after the driving is completed, that is, the accumulation time is almost zero.)
When the subject stops suddenly during the tracking operation, as shown in FIG. 19, the sign of the previous tracking defocus amount (PLST) and the current defocus amount (DEF) are inverted and the absolute values thereof are considerably large.

However, even in such a case, the current tracking defocus amount (PRED) is equal to the current defocus amount (DED).
F) to which the tracking correction amount (COMP) has been added, which is the same sign and the same size as the previous tracking defocus amount (PLST).
Even in the determination processing of No. 15, the tracking operation is started without determining that tracking is impossible, and the photographing lens further overruns from the in-focus position as shown by a broken line in FIG.
Therefore, in such a case, the tracking operation is not performed by # 470 and # 475, and as shown in FIG.
If the camera has passed the focusing position, it is possible to reach the focusing position in the next drive without performing the tracking operation. In # 480, the current tracking defocus amount (PRE
D) is calculated as the sum of the current defocus amount (DEF) and the previous tracking correction amount (COMP). In # 485, it is tested whether the sign of the current tracking defocus amount (PRED) and the sign of the previous tracking defocus amount (PLST) are the same sign. If the sign is different, the process proceeds to # 545 without performing the tracking operation. Jump. This is because, when the tracking direction is reversed, the tracking operation is temporarily stopped to perform normal driving, so that unstable operation (hunting, overrun, etc.) is prevented when the movement of the subject is reversed. In the case of the same sign in # 485, the process proceeds to # 490, and the absolute value (| PRED + PL) of the sum of the current tracking defocus amount (PRED) and the previous tracking defocus amount (PLST)
ST |) is greater than or equal to a predetermined value δ (for example, 200 μm). If the value is equal to or smaller than the predetermined value δ, the process jumps to # 545 without performing the tracking operation. The process of # 490 is
In the vicinity of the focus position, the tracking defocus amount and the error value included in the tracking defocus amount are substantially the same, and if a tracking operation is performed using this, an unstable operation (hunting, overrun, etc.) occurs near the focus position. The purpose is to prevent this. The predetermined value δ is experimentally determined to be a constant value,
It can also be changed according to various conditions (lens focal length, whether tracking is in progress, reliability of defocus amount, etc.).

In particular, it is effective to provide a predetermined value δ1 during tracking and a predetermined value δ2 (> δ1) during non-tracking in order to ensure stability. It is also possible to determine whether tracking is possible only by the absolute value of the current tracking defocus amount (PRED) instead of # 490, but # 4
By taking the sum with the previous tracking defocus amount (PLST) as in 90, the influence of an error included in the tracking defocus amount can be reduced, and a more stable tracking operation can be guaranteed. If it is determined in step # 490 that the value is equal to or larger than the predetermined value δ, the process proceeds to step # 495. The processes of # 495 to # 515 are processes for determining whether or not the tracking operation is possible according to the ratio between the current tracking defocus amount (PRED) and the previous tracking defocus amount (PLST). As described above, during the tracking operation, the defocus amount (DEF) becomes almost 0 and the tracking correction amount CO
MP becomes almost constant. Therefore, the current tracking defocus amount (PRED) and the previous tracking defocus amount (PLS)
The ratio to T) is ideally approximately 1. Tracking correction is a correction performed on the assumption that the subject is moving at a substantially constant speed. Even if the speed of the subject changes suddenly, such tracking correction may cause unstable operation (hunting, overrun, etc.). May cause. When the speed of the subject changes, the value of the tracking defocus amount changes accordingly.
The value of the ratio between the current and previous tracking defocus amounts also increases or decreases from 1.

Therefore, in the processing from # 495 to # 515, the tracking operation is performed only when the ratio is within a certain range including 1 to prevent the unstable operation due to a sudden change in the object speed. In steps # 495 to # 505, the value of the upper limit (r) of the ratio is changed depending on the value of the shot counter (PCOUNT). The reason will be described with reference to FIG. In the figure, the solid line is the locus of the movement of the photographing lens required to match the image of the moving object with the film surface, and the chain line is the locus of the actual movement of the photographing lens during the tracking operation. Before the full press, the AFCPU operates the CCD accumulation and focus detection calculation
The driving operation of the F motor is repeated, and the cycle is almost constant at F0, and the shot counter (PCOUNT)
Is 0. After full press, the shot counter (PCOUNT) becomes 1, and the photographing operation is performed after the driving operation. Before the photographing operation is performed, the driving is completed, that is, the CCD is accumulated. On the other hand, after the full press, the CCD is accumulated after the photographing operation. Therefore, the defocus amount before the full press is almost zero. The initial defocus amount (DEF2) after full pressing has a large value. Therefore, in the processing of the tracking algorithm module performed for the first time after the first photographing operation (7), that is, the number of shots is the second (P
When COUNT = 2), the above-described tracking defocus amount (PRED) at this time is equal to the previous tracking defocus amount (PLS).
T), the tracking operation is unnecessarily deviated unless the upper limit of the ratio (r) is set too large. Therefore, only when the shot counter is 2 (PCOUNT = 2), the upper limit (r) of the ratio is set to a value (R) larger than the normal value (RS).
L, RL> RS).

The above is the contents of the processing from # 495 to # 505. In # 495, it is tested whether or not after the first photographing operation (release). If it is not the first time (PCOUNT = 2), the processing proceeds to # 505. The upper limit value (r) of the advance ratio is set to a predetermined value R.
S (for example, 3), and the process proceeds to # 510. The first (PC
If OUNT = 2), the process proceeds to # 500, where the upper limit value (r) of the ratio is set to a predetermined value RL (for example, 6), and the process proceeds to # 510.
In # 510, as described above, it is tested whether the absolute value (| PRED |) of the current tracking defocus amount is r times or less of the absolute value (| PLST |) of the previous tracking defocus amount.
If it is not r times or less, it is determined that tracking is impossible, and the routine jumps to # 545 without performing the tracking operation. If it is less than r times, the process proceeds to # 545, where a predetermined value K (for example, 1/2) of the absolute value of the current tracking defocus amount (| PRED |) and the previous absolute value of the tracking defocus amount (| PLST |) Test if it is more than twice. If it is not K times or more, it is determined that tracking is impossible, and the control jumps to # 545 without performing the tracking operation. If it is K times or more, it is determined that tracking is possible, and the process proceeds to # 520. The comparison parameters δ, r, and k in the above-described determination processing of # 490, # 510, and # 515 have been described as the predetermined values.
Hysteresis of a predetermined width may be provided for these parameters depending on whether tracking is being performed or not. The hysteresis is set so that the tracking operation is hard to escape during the tracking operation, and the tracking operation is hard to enter when the tracking operation is not performed.

For example, δ is δ1 during tracking and δ2 outside tracking.
(> Δ1), r is RL1 or RS1 during tracking, RL2 (<RL1) or RS2 (<RS1) outside tracking, k is k1 during tracking, and k2 (> k1) outside tracking. By providing the hysteresis in this manner, the transition between the tracking operation and the tracking operation can be performed stably. Steps # 520 to # 540 are calculation processes for the tracking operation. # 520
Is a process for determining the coefficient α when the tracking correction amount (COMP) is calculated by multiplying the current tracking defocus amount (PRED) by the coefficient α. The contents of the process of # 520 will be described with reference to FIG. In FIG. 20, before the full press, the cycle consisting of the CCD accumulation and focus detection calculation operation and the AF motor driving operation is almost constant at F0, so that the coefficient α is approximately 1 is appropriate. After the full press, the shooting operation is included in the cycle, so that the cycle is F1, F2,
It becomes longer than F3 and the cycle F0 before full pressing. Before the full press, the coefficient α is determined so that the solid line and the dashed line intersect at the middle point of the accumulation time of the CCD. After the full press, the coefficient α is determined so that the solid line and the dash-dot line intersect at the middle point of exposure during photography (indicated by E in the figure). In the figure, the period from the start of CCD accumulation to the middle point of exposure during the photographing operation is represented by F1 'and F1'.
2 'and F3', the coefficient α is almost proportional to the ratio between the previous cycle and the cycle up to the middle point of the current exposure. When the shot counter is 0 (PCONET = 0), the coefficient α
Is F0 / F0, F1 in case of 1 and 1 (PCOUNT = 1)
When 1 '/ F0 = 1.5, 2 (PCOUNT = 2), F2' / F1 = 0.9 to 1, and when 3 (PCOUNT = 3), F3 '/ F2 = 0.8 to 1 is suitable. ing. It is desirable that the coefficient α be changed in accordance with the direction of movement of the subject and the focal length of the lens.

The reason will be described with reference to FIG. FIG.
In FIG. 1 (A), the solid line indicates the trajectory of the movement of the photographing lens for making the subject image always coincide with the film surface when the subject approaches from ∞. It is a locus of the movement of the taking lens. When the subject approaches, the movement of the taking lens increases as the subject approaches. On the other hand, when the subject goes away, the movement of the taking lens becomes smaller as the subject goes away. Therefore, the coefficient α for determining the tracking correction amount
Is preferably set to be smaller when the subject moves away than when the subject approaches. FIG. 21B shows the trajectory of the taking lens with respect to the approaching subject, the solid line shows the trajectory of the movement of the taking lens having a short focal length, and the dashed line shows the trajectory of the movement of the taking lens having a long focal length. ing. When the focal length is long, the movement of the photographing lens is constant from infinity ∞ to a close distance. On the other hand, when the focal length is short, the locus of the movement of the photographing lens rises sharply as the focal distance approaches. Therefore, it is preferable to set the coefficient α to be smaller for a lens having a long focal length than for a lens having a short focal length.

For the above reasons, in # 520, there are three shots: the number of shots (PCOUNT), whether the focal length (FL) of the lens is larger or smaller than a predetermined value (FX), and the moving direction of the subject (the sign of the tracking defocus amount) The coefficient α is determined according to the parameters as shown in the table. In # 525, #
The tracking correction amount (COMP) is obtained by multiplying the coefficient α determined in 520 by the current tracking defocus amount (PRED). Further, even if the lens has the same focal length, the way of acceleration is different as shown by the solid line in FIG. Therefore, in order to treat the object more strictly, it is better to determine the value of α in consideration of the distance information of the lens. For example, when the subject approaches, the value of α is increased so that it is closer to the object and the correction amount is increased. Good to take.

Next, a method of obtaining the value of α without using the focal length information or distance information of the lens will be described. For this purpose, the value of PRED / PLST = β is calculated using the above-mentioned PLST and PRED. When the image plane moving speed is constant, β = 1, β> 1 during acceleration, and β <1 during deceleration. Therefore, the value of α can be determined using β. In this case, for example, the value of α is determined from the value of β using the following table.

[Table 1] The value of α is slightly smaller than the value of β, which depends on the time interval between the exposure and the accumulation time, and is generally

(Equation 6) Becomes In step # 530, the current tracking drive amount (DRIV) is determined by adding the tracking correction amount (COMP) obtained in step # 525 to the current defocus amount (DEF) in order to perform the tracking operation. In # 535, the previous (final) tracking defocus amount (PLST) is replaced by the current tracking defocus amount (PRE).
Prepare for the next tracking process / judgment by replacing with D). # 5
At 40, it is determined that the tracking operation is being performed (PRSFLG ON), and the process proceeds to the next (8) focus determination / display module. The above is the processing when the tracking operation is performed. Steps # 545 and # 550 are processing when it is determined that tracking is impossible. In step # 545, the current defocus amount (DEF) is adopted as the previous (final) tracking defocus amount (PLST) to prepare for the next tracking process / determination. In step # 550, it is determined that the tracking operation is not being performed, the tracking flag is reset (PRSFLG is turned off), and the process proceeds to the next (8) focusing determination / display module.

FIG. 22 is a flowchart of the (8) focus determination / display module. In # 560, it is tested whether the focus mode is the one-shot mode (ONEFLG ON). If the focus mode is the one-shot mode, # 565 is skipped and the process proceeds to # 570. If the mode is not the one-shot mode, that is, if the mode is the continuous AF mode or the manual mode, the fixed flag is reset at # 565 (FI
XFLG is turned off) so that the drive display is not fixed even after being once focused. In # 570, it is tested whether the drive display is fixed (FIXFLG is ON), and if it is fixed, the process passes the following processing and proceeds to the next (9) AF permission standby module. If the drive display is fixed in # 570, the flow advances to # 575 to determine whether focus detection has been impossible (LO
If CFLG is ON) and the focus cannot be detected, # 6
Proceeding to 10, it is determined that the image is out of focus (FZCFLG OFF), and the display unit 44 (X) of the AF display means 40 is activated and displayed at # 615. If the focus cannot be detected in step # 575, the scan prohibition flag is set in step # 580 so that no further scanning is performed (NSCF).
LG is ON) and the tracking mode is set at # 585 (PMD
FLG is ON) Test. If the mode is the tracking mode, the process proceeds to step # 590 to determine whether or not tracking is being performed (PRSFLG is O
N) Test.

If tracking is in progress, a test is performed at # 595 to determine whether the lens is at the lens end (LLMFLG is ON). If it is the lens end, it is determined that out of focus at # 600 ((10) so that the drive is always performed by the drive control module), the focus flag is reset (FZCFLG is turned ON), and further displayed at # 605. Both the units 41 and 43 are activated to indicate that the tracking operation is being performed in a display mode different from the other display states (in-focus and out-of-focus), and the process proceeds to the next (9) AF permission standby module. The processes from # 620 to # 640 are processes for determining the focusing zone for the focus determination. If the tracking mode is not set in # 585, the narrow zone (from the non-focusing state to the focusing state) (ZONEN) is set to Z1 (for example, 50 μm), and a wide zone (ZONEW) when going out of focus to out of focus is set to Z2 (for example, 150 μm).
The process proceeds to the determination process after 645. On the other hand, when tracking is not being performed in # 590 and when it is determined that the lens is at the lens end in # 595, the process proceeds to # 625, and a narrow zone (ZONEN) is set.
To Z3 (for example, 50 μm), wide zone (ZONEW)
Is set to Z4 (for example, 100 μm). By setting Z4 smaller than Z2, the response of driving in the tracking mode can be enhanced. In # 630, a test is performed to determine whether the luminance is low (LOLFLG is ON). If the luminance is low, the process proceeds to # 640. If the brightness was not low, # 635
To test whether the reliability is high (SLOP is equal to or more than a predetermined value SY). If the reliability is high, the flow directly goes to # 645. In step # 635, the reliability determination parameter SLO
Although P is used, the contrast parameter CONT obtained by Expression (7) may be used.

On the other hand, when it is determined that the reliability is low,
If the brightness is low at 630, a wide zone (ZONEW) is set to Z5 (for example, 200 μm) larger than Z4 at # 640.
m) and proceeds to # 645. # 630 to # 640
Is a process for setting a focusing zone with emphasis on stability over responsiveness when low luminance and reliability are low in the tracking mode. If it is determined in # 645 that focus was previously achieved (FZCFLG is ON) and it is out of focus, the process proceeds to # 650 and the absolute value of the current defocus amount (|
DEF |) and a narrow zone (ZONEN). If it is in focus, the absolute value (| DEF |) of the current defocus amount is compared with the wide zone (ZONEW) in # 680. If it is determined in steps # 650 and # 680 that the image is out of the zone, the flow advances to step # 685 to determine that the image is out of focus.
FF), and in # 690, the current defocus amount (DE
The sign of F) is determined.

If the sign is positive (previous pin), the flow advances to step # 695 to activate the triangular mark on the display section 41 to display the front focus state and to proceed to the next module. When the sign is negative (back focus), the triangular mark on the display unit 43 is activated in # 700 to display the back focus state, and the process proceeds to the next module. If it is determined in # 650 and # 680 that the area is within the zone, the focusing flag (FZCFLG) is set (ON) in # 655. In step # 660, it is tested whether the focus mode is the one-shot mode (ONEFLG is ON). If not, the process jumps to step # 675. If the mode is the one-shot mode, the fixing flag is set (FIXFLG is turned ON) in # 665, and the subsequent driving and display are fixed. In step # 670, the release permission signal (RL) is permitted (ON) to notify the main CPU of the focus release permission in the one-shot mode.
Go to 75. In step # 675, the focus mark on the display unit 42 is activated to display the focus, and the process proceeds to the next (9) AF permission standby module.

FIG. 23 is a flowchart of the (9) AF permission waiting module. In step # 710, a test is performed to determine whether the mode is the tracking mode (PMDFLG is ON). If the mode is the tracking mode, the process waits for the AF permission signal (AF) to be permitted (ON) in # 715, and if not, repeats # 710 and # 715. When the AF permission signal is permitted, the process proceeds to # 720 and a test is performed to determine whether or not the release is being performed (PCOUNT ≠ 0). (10) Set the tracking delay state (PDYFLG ON) from the start of driving by the drive control module until the relay permission signal (RL) is permitted (ON), and at the same time, the tracking delay The time T1 is set to the tracking delay (PR
(SDLY) and proceeds to # 730. If relaying is not being performed in # 720, processing in # 725 is not performed and #
Proceed to 730. In # 730, is the tracking operation in progress (PR
(SFLG is ON) If the tracking operation is in progress,
In step # 375, the current defocus amount (DEF) is replaced with the tracking drive amount (DRIV), and the next (10) drive control module performs driving with the tracking drive amount instead of the defocus amount. If tracking is not being performed in # 730, the process directly proceeds to (10) drive control module.

FIG. 24 is a flowchart (10) of the drive control module. At # 740, an AF permission signal (AF)
Tests whether is permitted (ON), and if not, #
The process proceeds from 865 to the non-drive processing of # 880. If the AF permission signal indicates permission, the flow proceeds to step # 745, and a test is performed to determine whether the AF mode is set (AFMFLG is ON). The AF mode flag (AMFLG) is set when the mounted lens is an AF-capable lens and the focus mode is not manual. If not in AF mode,
The process proceeds to the non-drive processing after # 865. In the case of the AF mode, whether the drive is fixed at # 750 (FIXF
(LG is ON) Test. If the drive was fixed,
The process proceeds to the non-drive processing after # 865. If the drive is not fixed, the process proceeds to # 755, and whether the camera is in focus (FZCF
(LG is ON) Test.

If the camera is in focus, the flow advances to the non-drive processing in and after # 865. If the camera is out of focus, the process proceeds to step # 760, and a test is performed to determine whether focus detection is impossible (LOCFLG is ON). If the focus cannot be detected, the process proceeds to the scan drive process after # 830. If the focus detection has not been possible, the process proceeds to the driving process after # 765. From # 765 to #
In step 765, the sign of the current defocus amount (DEF) is compared with the sign of the final defocus amount (DEFLST). Sets the drive inversion flag (REVFLG is turned ON) in # 770, and proceeds to # 780. If the signs are the same, the drive inversion flag is reset (REVFLG is turned off) in # 775, and the flow proceeds to # 780. In # 780, the final defocus amount (DEFLST)
Is updated with the current defocus amount (DEF). In # 785, the current defocus amount (DE
F) The driving amount of the AF motor in accordance with F)
The expected pulse number (ETM) fed back from 2 is calculated using the following equation. ETM = KL × KB × | DEF | (8) In equation (8), the coefficient KL is the number of rotations of the lens-side coupling 14 per unit defocus amount of the image plane of the photographing lens, and the coefficient KB is the body-side coupling. The number of pulses generated by the encoder 52 per one rotation of 53 is shown.
Therefore, according to equation (8), when the image plane of the photographing lens is moved by the current defocus amount (DEF), the encoder 5
That is, the number of pulses for which 2 should be generated is obtained. The calculated expected pulse number is set in a comparison register as a set value of a pulse counter interrupt described later. In # 790, scan drive is not performed (SCAFLG is OFF), and # 7
At 95, drive this time (DRVFLG is ON),
At # 800, the driving state is set to driving (MOVFLG is ON).

Next, in # 805, the pulse counter is cleared (ECNT = 0) before the driving starts. Steps # 810 to # 820 are processes for determining the driving direction according to the sign of the current defocus amount (DEF) and starting driving. In # 810, the current defocus amount (DEF)
Is checked, and if it is the previous pin (the sign is positive), the AF motor starts to be driven in the far direction at # 815. In the case of the rear focus (the sign is negative), the process proceeds to # 820, and the AF motor starts to be driven in the near direction. At the end of the driving process, the interrupt of the pulse counter is permitted in # 825, and the process proceeds to the next (11) AGC operation module. On the other hand, if it is determined in step # 760 that the focus cannot be detected, the flow advances to step # 830 to test whether scanning is prohibited (NSCFLG is ON). If the scan is prohibited in # 830, the scan drive is not performed and # 86
The process proceeds to the non-driving process after 5.

If the scan is not prohibited, # 83
In step 5, it is tested whether scanning is currently being performed (SCAFLG is ON). If a scan is being performed, the process jumps to # 860 without performing the scan drive start process. If the scan is not being performed, a scan drive start process from # 840 to # 855 is performed. Scanning at # 840 (SCA
FLG is turned ON) and the current drive is performed in # 845 (D
RVFLG is turned ON), and driving is being performed at # 850 (MOVF
LG is ON). In # 855, the AF motor starts to be driven in a predetermined direction to start scanning, and the lens end counter is cleared in # 856 (LCOUN).
T = 0), and then proceeds to # 860. In step # 860, the interruption of the pulse counter is prohibited (11), and the process proceeds to the AGC operation module. # 740, # 745, # 750, # 755, #
If the process proceeds from step 830 to step # 865, the scan flag is reset (SCAFLG is turned off). In # 870, it is determined that there is no drive this time (DRVFLG is OFF), and in # 875, it is not driving (MOVFLG is OFF). At the end of the non-drive processing, the drive of the AF motor is stopped at # 880, and the process proceeds to the next module.

FIG. 25 is a flowchart of the (11) AGC operation module. First, at step # 890, the next accumulation time (INTT) of the CCD is calculated by equation (9). INTT = INTT × IX / MAX (9) In equation (9), INTT on the right side is the current accumulation time,
X is the target value of the maximum value of the CCD data, and MAX is the current C value.
This is the maximum value of the CD data. According to equation (9), the next accumulation time (INTT on the left side) is set such that the maximum value of the next CCD data becomes the target value IX. In # 895, it is tested whether the next accumulation time is equal to or longer than the predetermined value IY, that is, whether the brightness is low. If the value is equal to or more than the predetermined value IY, the process proceeds to # 900, and the low luminance flag is set (LOLFL).
G is turned ON), and the process returns to (2) pre-storage processing. If the value is not equal to or more than the predetermined value IY, the process proceeds to step # 905, where the low luminance flag is reset (LOLFLG is turned off), and the process returns to (2) pre-storage process. In the above processing, whether or not the luminance is low is determined on the basis of the accumulation time of the CCD. It may be determined whether or not the brightness is low. Although IY is set to a predetermined value, a predetermined hysteresis may be provided depending on whether or not the brightness was low at the previous time. The above is one cycle of the accumulation calculation and driving of the main program, and this processing is repeated.

FIG. 26 is a flowchart of the timer interrupt program. The timer interrupt program is activated and executed at regular intervals during the operation of the main program. # 910 to # 925 are the AF mode flags (A
FMFLG) is updated. In # 910, whether the attached lens is an AF lens (AFLFLG is O
N) Test. If the lens is an AF lens, # 915
And the focus mode signal (FM) is one shot A
Test for F or continuous AF (C or O). If it is the one-shot AF or the continuous AF, the process proceeds to # 920, and the AF mode flag (AMFLG) is set (OMFLG).
N), and proceed to # 930. If it is not an AF lens in # 910, and if the focus mode signal (F
If M) is in the manual mode, the process proceeds to step # 925, where the AF mode flag (AMFLG) is reset (OF
F) and proceed to # 930. The processes from # 930 to # 940 are processes for updating the one-shot flag. In # 930, a test is performed to determine whether the focus mode signal (FM) is in the one-shot AF mode. If the mode is the one-shot mode, the one-shot flag is set (O
NEFLG is turned ON), and the process proceeds to # 945. If the mode is not the one-shot mode, the one-shot flag is reset (ONEFLG is turned off), and the process proceeds to # 945. # 9
Blocks 45 to # 1005 are the full-press, tracking delay, and mirror-up processing in the tracking mode. # 945
In step # 965, it is determined whether or not the tracking mode is set. In step # 945, the frame speed mode signal (DM)
Is normal continuous shooting (C2), and if not, the process proceeds to # 965 as the non-tracking mode. If it is the normal continuous shooting (C2), the process proceeds to step # 950, and a test is performed to determine whether the mode is the AF mode (AFMFLG is ON). If not in the AF mode, the process proceeds to # 965 as the non-tracking mode. If the mode is the AF mode, the process proceeds to # 955, and a test is performed to determine whether the mode is the one-shot mode (ONEFLG is ON). If the mode is the one-shot mode, the process proceeds to # 965 as the non-tracking mode. If the mode is not the one-shot mode, the process proceeds to # 960 as the tracking mode. After all, when the focus mode is the continuous AF, the frame speed mode is C2, and the attached lens is the AF lens, the process proceeds to # 960, the tracking mode is set (PMDFLG is ON), and thereafter,
Tracking mode processing is performed.

On the other hand, in other combinations of modes, the non-tracking mode and the tracking mode are reset (PMDFL
G is turned OFF), and the tracking mode processing is not performed.
The process proceeds to mirror-up processing of No. 10. When the tracking mode is set in # 960, the release button signal (R
S) is fully pressed (ON), and if it is fully pressed, the full-press processing in the tracking mode of # 975 is skipped and # 9 is skipped.
Go to 80. If not fully pressed, the tracking delay flag is reset in # 975 (PDYFLG is turned off)
The mirror flag is reset (MIRFLG is turned off), the release permission signal (RL) is prohibited (OFF), and the operation during full pressing in the tracking mode is reset. # 98
Steps 0 to # 995 are processes for timing the tracking delay time in the tracking mode and for permitting release at the end of the tracking delay. Is tracking delay in # 980 (PD
If YFLG is ON) and the tracking is not being delayed, the processing after # 985 is not performed, and the process proceeds to # 1000.
If the tracking delay is being performed, the tracking delay time is reduced by 1 in # 985 (PRSDLY = PRSDLY-
1). For example, if the timer interruption is performed every 1 ms and the tracking delay time is 45 ms, (9) PRDLY = 45 is set in the AF permission standby module, and this is reduced by 1 for each timer interruption. It becomes 0. In # 990, the tracking delay ends (PRSDL
Y = 0), and if not finished, # 1000
Exit.

In the case of termination, delay termination processing is performed in # 995, and the tracking delay flag is reset (PDYFL).
G is turned off), the release permission signal (RL) is permitted (ON), the release permission is transmitted to the main CPU, and the routine proceeds to # 1000. # 1000 and # 1005
Is a mirror-up process in the tracking mode, and # 1
If the mirror up signal (MR) is up (ON) at 000 and it is down, the process directly goes to # 1010. If the mirror is up, the mirror flag is reset (MIRFLG is turned off) in # 1005, and (2) the storage preprocessing module enables the process to proceed to the post-mirror up, and prohibits the subsequent release. for,
# 1 by setting the release permission signal (RL) to prohibition (OFF)
Go to 010. # 1010 to # 1050 are blocks for the mirror-up process. If the mirror-up signal (MR) is up (ON) in # 1010 and the test is down, the mirror-up flag is reset (RL) in # 1055.
SFLG is turned OFF), and it is determined that the mirror is down.
Exit to 60. If the mirror is being raised, the fixed flag is reset (FIXFLG) in # 1050 in order to release the drive display fixed by focusing in the one-shot mode and the release permission by performing a single shooting operation (mirror up). OFF) and release permission signal (RL)
Is prohibited (OFF). In # 1020, is the mirror up during the previous timer interrupt (RLSFLG is O
N) Test, if mirror is up, # 1 as it is
Exit to 060.

If the mirror was down last time, the mirror started to move up from down between the previous and current timer interrupts, so that a drive delay is started in synchronization with this as shown in FIG. In # 1025, the mirror-up flag is set (RLSFLG is turned on), and as a mirror-up, a test is performed in # 1030 as to whether the mirror is currently being driven (MOVFLG is on). If it is stopped, there is no need for a drive delay, so the flow goes to # 1060. When driving,
In # 1035, the delay flag is set (DLYFLG is turned on) to enter the drive delay state. Next, in # 1040, a test is performed to determine whether the mode is the tracking mode (PMDFLG is ON). In the case of the tracking mode, the delay time is set to a predetermined value T so that the mirror-up operation by the main CPU is completed at the time when the driving is completed and the exposure is started.
2 (DLY = T2), and proceeds to # 1060.

On the other hand, when the tracking mode is not set, the delay time is set to a predetermined value T0 (DLY = T0, T0) so that the driving ends before the mirror-up operation of the main CPU ends.
<T2), and the process proceeds to # 1060. Steps # 1060 to # 1070 are motor stop processing when the AF permission signal is disabled. In step # 1060, it is tested whether the AF permission signal (AF) is permitted (ON). If the signal is permitted, the process proceeds to step # 1075 without performing the stop process. If prohibited, #
At 1065, it is tested whether the AF motor is currently being driven,
If it is stopped, the process goes to # 1075 without performing the stop process. If the motor is being driven, the driving of the AF motor is stopped at # 1070 and the driving state flag is reset (MOV
FLG is turned off), and the process proceeds to # 1075. Steps # 1075 to # 1090 are blocks for timing of the drive delay time and delay end processing. In step # 1075, a test is performed to determine whether a drive delay is being performed (DLYFLG is ON). If the drive delay is not being performed, the process directly proceeds to step # 1095. If the delay is in progress, the drive delay time is reduced by 1 in # 1080 (DLY = DLY-1). 1ms timer interrupt
If the delay time is 55ms,
DLY = 55 is set at # 1045 and becomes 0 after 55 ms because it is decremented by 1 for each timer interrupt. In # 1085, whether the drive delay has ended (DL
Y = 0) Test, if not finished,
Exit to # 1095. When the process is completed, the drive of the AF motor is stopped in # 1090, the drive delay flag is reset (DLYFLG is turned off), the drive state flag is reset (MOVFLG is turned off), and # 1090 is executed.
Go to 95. Steps # 1095 to # 1125 are lens end processing blocks which stop the motor at the lens end in normal driving and drive reversal at the lens end in scan driving. In # 1095, is it the lens end (ECNT =
ELST) test. Since the timer interrupt is performed at predetermined time intervals, if no pulse is generated at the lens end, the contents of the pulse counter do not increase, and the contents of the pulse counter (ELST) such as the previous timer interrupt and the current timer are used. Contents of pulse counter at interrupt (ECNT)
Matches.

Accordingly, it can be determined whether or not the lens is at the lens end in accordance with the coincidence or disagreement of the contents of the pulse counter. If it is not the lens end in # 1095, the process goes to # 1115. If it is at the lens end, a test is performed at # 1100 to determine whether scanning is in progress (SCAFLG is ON). If the scanning is not being performed, the process proceeds to stop processing of # 1125. In the case of scanning, the number of times of reaching the lens end in # 1105 is a predetermined value LX (for example, 2) (LCOUNT = LX).
Testing. If the predetermined value LX has been reached, it is determined that the scan has been completed, and the process proceeds to # 1120, where the scan flag is reset (SCAFLG is turned off), the scan prohibition flag is set (NSCFLG is turned on), and the scan end processing is performed. And proceed to # 1125. # 1125
Is executed at the lens end in normal driving and at the end of scanning, driving of the AF motor is stopped, the driving state flag is reset (MOVFLG is turned off), and # 11 is executed.
Proceed to 30 and return to the main program. on the other hand,
If the lens end counter has not reached the predetermined value LX in # 1105, # 11 is performed to reverse the scan driving direction.
The process proceeds to step 10, where the content of the lens end counter is increased by 1 (LCOUNT = LCOUNT + 1), the driving direction of the AF motor is reversed, and the process proceeds to # 1115. In # 1115,
Update the contents of the last pulse counter (ELST = ECN
T) Then, the process returns to the main program in # 1130.

FIG. 27 is a flowchart of the pulse counter interrupt program. The pulse counter interrupt indicates that the cumulative number of pulses (ECNT) generated by the encoder 52 is the expected number of pulses (E) calculated by the drive control module (10).
TM), which is an interruption that is performed when the movement of the photographing lens to the in-focus position is completed. In step # 1140, the driving of the AF motor is stopped. In step # 1145, the drive state flag is reset (MOVFLG is turned off). Then, return to the main program.

The above is the description of each module and the interrupt program of the main program of the AFCPU.
The driving of the F motor is controlled. In the description of the present embodiment, in the (7) tracking algorithm of the AFCPU program, the absolute value (| PRED + PLST |) of the sum of the current tracking defocus amount and the previous tracking defocus amount is compared with a predetermined value δ. It has been determined whether or not to perform a tracking operation (branch of # 490). The reason for determining whether or not the subject is a moving subject and determining whether or not to perform a tracking operation in accordance with the determination as described above will be described in detail below. Conventionally, whether or not a subject is moving is determined based on the past and current defocus amounts by calculating a current tracking defocus amount as a defocus amount taking into account subject movement during a cycle of defocus detection. This is performed by comparing the current tracking defocus amount with a predetermined value. For example, the current tracking defocus amount PRED
(0) If | PRED (0) | ≧ δ, which is the predetermined value δ, it is determined that the object is a moving object. If | PRED (0) | <δ, it is determined that the object is not a moving object.

Referring to FIG. 28, a conventional tracking operation and determination of a moving subject will be described. In FIG. 28, the solid line shows the trajectory of the movement of the photographing lens for constantly matching the image of the moving subject to the film surface, the dashed line shows the trajectory of the actual movement of the photographing lens, and The charge accumulation and the focus detection calculation of the sensor are performed while the photographing lens is stopped, and the charge accumulation of the sensor (FIG. 28)
In (2), the accumulation time T (0) is set immediately after the driving is completed. In the tracking operation, the driving amount of the taking lens is the defocus amount obtained by focus detection,
The tracking correction amount is calculated by adding the tracking correction amount.
Is calculated as the sum of the immediately preceding defocus amount DEF (-1) (corresponding to the difference between the solid line and the dashed line in the figure) and the immediately preceding tracking correction amount COMP (-1). Also, the current defocus amount DEF (0) obtained at the end of driving.
The tracking defocus amount PRED (0), which is the sum of the previous tracking correction amount COMP (-1) and the previous tracking correction amount COMP (-1), is evident from FIG. It corresponds to the movement amount of the photographing lens indicated by the solid line, that is, the movement amount of the subject, and it can be determined that the subject is moving when the absolute value of the tracking defocus amount PRED (0) is equal to or more than a predetermined value. It is.

However, when the moving subject is determined by comparing the current tracking defocus amount alone with a predetermined value as described above, the defocus amount DE
In some cases, an erroneous determination was made due to an error included in F (0). In particular, when the movement of the subject is minute, the ratio of the error in the tracking defocus amount PRED (0) becomes relatively large. The movement of the photographic lens has become unstable repeatedly. Also, as described above, if the current tracking defocus amount is compared with a predetermined value alone, the cycle of detecting the amount of movement of the subject is as short as one cycle, which may affect the random noise movement of the subject. As a result, the determination of the movement of the subject is erroneous, and the movement of the minute subject cannot be detected. Further, even when the focus detection cycle fluctuates, there is a high possibility that an erroneous determination is made under the influence of the fluctuation. In this embodiment, in order to solve the above-described conventional problem, the sum of the current tracking defocus amount and the previous tracking defocus amount is calculated without comparing the tracking defocus amount alone with a predetermined value as described above. By comparing the absolute value with a predetermined value δ to determine a moving subject, the effects of errors included in the tracking defocus amount, random noise-like motion of the subject, and fluctuations in the focus detection cycle are statistically determined. This enables stable operation.

In general, a moving subject is determined by comparing a result obtained by performing a statistical averaging process on the tracking defocus amount instead of the process of # 490 in the embodiment with a predetermined value δ. Can be. For example, when the current tracking defocus amount is PRED (0) and the tracking defocus amount n times before this time is PRED (n) (n is a positive integer), a statistical process as shown in Expression (10) is performed. The determination made may be made in # 490. | K (0) × PRED (0) + k (1) × PRED (1) +... + K (n) × PRED (n) +... K (N) × PRED (N) |> δ (10) In the equation (10), k (n) is a predetermined weighting factor, and N is an arbitrary integer. In the embodiment, N = 1 in the equation (10),
This is the case where k (0) = k (1) = 1. A period (N × cycle time) for detecting the moving amount of the subject by appropriately selecting an arbitrary integer N in the equation (10).
Can be selected. Also, the weight coefficient k (n) may be set as in equation (11) in order to increase the responsiveness by weighting with the latest tracking defocus amount. k (0)> k (1)>... k (n) >>...> k (N) (11) Also, the weighting coefficient k (N) is determined by the tracking defocus amount PRED.
The parameter SLOP (n) and the contrast value CONT obtained by the defocus amount calculation at the time of calculating (n)
It may be set as in equation (12) in proportion to (n). k (n) = k × SLOP (n) or k × CONT (n) (12) In the equation (12), k is a predetermined constant. By setting as in equation (12), the tracking defocus amount is averaged with a weight proportional to the reliability, so that a moving object with higher accuracy can be determined. Also, in order to cancel the fluctuation of the focus detection cycle time (corresponding to the time when one drive ends in FIG. 28 from the time when the next drive ends) when the tracking defocus amount PRED (n) is obtained. Each cycle time T (n) can be measured and used to determine the weighting coefficient k (n) as in equation (13).

(Equation 7) In the equation (13), k is a predetermined constant. By setting as shown in equation (13) and performing the statistical averaging processing of equation (10), each weighting factor × tracking defocus amount becomes the tracking defocus amount per unit time, so there is no influence of cycle time fluctuation. It will be.

As described above, the statistical averaging process is performed on the tracking defocus amount to determine whether or not the subject is a moving subject. And the fluctuation of the focus detection cycle (especially when the shooting operation enters the middle during the tracking etc.) to reduce the influence of the error, and reduce the influence of the error. Therefore, δ can be set small, so that the movement of a finer subject can be determined as a moving subject and can be included in the tracking operation, so that improvement in tracking performance can be expected. One of the features of the present invention is that after the release is fully pressed, the mirror-up is started after a first predetermined time T1 from the start of the lens driving, and the cycle of the exposure, the cycle of the accumulation, etc. is independent of the lens driving time. There is a tracking technology that keeps the time constant and can always take a picture that is in focus at the moment of exposure. The tracking operation during continuous shooting has the following sequence. That is, mirror up, exposure, mirror down, charge accumulation, calculation, and lens driving are repeated. Mirror up and mirror down time 50 ~
100 ms is always constant and does not fluctuate in the same model. Also, the exposure time is substantially constant during continuous shooting, and does not contribute to variation since it is short in time and about 30 ms or less under bright conditions for tracking. The accumulation time also depends on the brightness of the subject, but is substantially constant during continuous shooting, and has a small contribution to variation because it is about 30 ms or less at normal brightness. The calculation time also varies slightly depending on the focus detection system, but 20.
The constant value is between ms and 100 ms, and the variation during continuous shooting is small.

On the other hand, the driving time of the lens fluctuates in the range of 0 ms to 100 ms or more according to the defocus amount driven for tracking. If one cycle is comprised of mirror up, exposure, mirror down, accumulation, calculation, and lens drive, and a sequence is formed as before, the mirror up for the next cycle starts after the previous lens drive ends. Assuming that one cycle time is about 300 ms (three frames per second), one cycle time is, for example, 250 ms to 3 depending on the length of driving time.
There is a possibility that it may fluctuate as much as 50 ms. In tracking, the following method can be used so as to be in focus at the moment of the next exposure. The past one accumulation cycle and the image plane movement amount due to the movement of the subject during that period (PRED)
And the time from the midpoint of the last accumulation period to the midpoint of the next scheduled exposure time, and the tracking correction amount (COM
P)

(Equation 8) And the value obtained by adding the above-mentioned tracking correction amount COMP to the final defocus amount DEF is driven as the tracking drive amount. If the mirror is raised at the end of the driving, the driving time will cause a variation of about 100 ms not only in the denominator of the above equation but also in the numerator, for example, the numerator will vary in the range of 150 ms to 250 ms. . Since the driving time cannot be known in advance, in the method of starting mirror-up by returning to the end of driving, the variation of the denominator and numerator becomes large and it is not possible to determine an appropriate α, so that effective tracking can be performed. Can not.

Therefore, in the present invention, regardless of the driving time, the mirror-up operation is started after a first predetermined time T1 from the start of driving, regardless of whether the driving is being performed or the driving is completed. As a result, both the denominator and the numerator of the above equation become substantially constant, and the tracking correction amount COMP is calculated using the value of α determined in this manner, so that the in-focus state is reliably realized at the moment of the next exposure. It is possible. Also, since the film is wound up after the mirror is lowered, the power supply capacity is insufficient, so that film winding and lens driving cannot be performed at the same time. Since the film winding time is substantially constant during this period, there is no problem even in this case because the cycle time does not change.

That is, when the essential points are put together in order, the focus detection means for repeatedly calculating the defocus amount of the photographing lens, and the subject during the defocus detection cycle based on the past and current defocus amounts A tracking correction amount calculating means for calculating a tracking defocus amount (PRED), which is a defocus amount accompanying the movement, and calculating a tracking correction amount (COMP) for tracking based on the calculated defocus amount, and a current defocus amount (DEF) Lens driving means for driving the photographing lens based on a tracking drive amount (DRIV) obtained by adding a tracking correction amount (COMP) to the camera, and means for controlling mirror-up, accumulation, and drive timing. Then, the control means performs the lens drive in the cycle of mirror up, exposure, mirror down, accumulation, calculation, and drive after the release is fully pressed. Control as is done with mirror-up after a first predetermined time T1 after the start.

Further, the control means limits the time during which the lens can be driven to a predetermined maximum time (T1 + T2). Therefore, when the lens driving is not completed even after the lapse of T2 from the mirror-up, the lens is forcibly stopped. . here,
T2 is approximately the same as the time from the start of the mirror up to the start of the exposure, and preferably, the time during which the driving is almost stopped immediately before the exposure. Predetermined maximum drive time (T1 + T2)
It is preferable to set a time for completely driving the defocus amount of 3 to 4 mm within that time, for example, about 100 ms. In this case, T2 which is almost equal to the mirror-up time is 50
If it is about msec, it will be about T1 to 50 msec. If the driving is not completed within the predetermined maximum time and the driving is stopped forcibly, the remaining amount of the driving is checked, and if it exceeds a predetermined value, the tracking correction should not be performed next time. By doing so, it is possible to avoid overrun that accompanies the tracking operation depending on the manner of movement of the moving object, and to speed up recovery from the overrun even when overrun occurs. Further, in the present invention, the configuration is such that one cycle of the focus detection operation is always included in the cycle of the mirror-up operation during continuous shooting.
This is to make the cycle of exposure and accumulation constant each time and to easily predict the lens position at the next exposure. Of course, the time from the end of exposure to the start of accumulation is controlled by the control means so as to be always constant.

In the present invention, if the maximum drive time is determined and the required drive amount is achieved within that time, focusing can be achieved at the moment of exposure regardless of the length of the drive time. Therefore, complicated time control and drive speed control are unnecessary, and it is easy to cope with the case where the load varies depending on the type of the interchangeable lens and the drive speed greatly varies.

[Table 2]

[Table 3]

[Table 4]

[Table 5] Table 5 <Conditions for performing tracking operation> (1) Tracking mode (2) Low contrast (3) Reliable (4) Not scanning (5) Driving last time (6) Driving direction not reversed (7) The defocus amount and the final tracking defocus amount have the same sign, or the defocus amount is small even if the sign is different. (8) The tracking defocus amount and the final tracking defocus amount have the same sign. (9) The tracking defocus amount and the final The absolute value of the sum of the tracking defocus amounts is equal to or more than a predetermined value δ. (10) The tracking defocus amount is equal to or less than a predetermined value r * the final tracking defocus amount. (11) The tracking defocus amount is equal to or more than the predetermined value k * the final tracking defocus amount. As described above, according to the first aspect, after the driving direction of the photographing lens is reversed, the tracking drive is prohibited in the predetermined number of times of driving, so that the tracking drive is unnecessarily entered near the focal point and hunting occurs. But Ku, high stability, moreover, it is possible to provide an automatic focusing device high followability to the object.

(Second Embodiment) The second embodiment of the present invention has almost the same configuration as the first embodiment, and is configured so that the tracking correction amount of the first embodiment can be obtained more accurately. is there. Hereinafter, description will be made with reference to FIGS. First, the points of identification and correction before and immediately after the release is fully pressed and after one mirror-up operation will be described. As shown in FIG. 20, means for discriminating the drive up to the full press of the release, the drive immediately after the full press, the second drive after the full press, the drive after the third press after the full press, and the timing of calculating the tracking correction document preceding them. , So that the above-mentioned identification results can be obtained by PCOUNT = 0, 1, 1 in FIG.
2, 3. Before the full pressing (PCOUNT0), the cycle time of accumulation, calculation, and driving is constant at FO. To calculate the tracking correction amount (COMP) from the tracking defocus amount (PRED), COMP = PRED × α Can be set to, for example, α = 1. In other words, if the aim is to focus each time at the middle point of the accumulation before full pressing, the cycle time between the previous accumulation = F0 and the interval until the next scheduled time are also F0, so α = F0 / F0 = 1. Because it becomes. After the full pressing, the exposure accompanying the mirror-up is inserted between the driving and the accumulation, so that the accumulation cycle time changes. After full pressing, the moment of optimization of focusing changes from the middle point of accumulation to the middle point of exposure (the middle point means the center time). Therefore, the tracking correction amount (COMP) is COMP = PRED × α in order to focus on exposure.

(Equation 9) Becomes Therefore, immediately after full pressing (PCOUNT1)

(Equation 10) Is determined by Mirror up after pressing all 2
At the time of calculation (PCOUNT2) prior to the second drive,

[Equation 11] It is. Further, at the time of calculation (PCOUNT3) prior to the third and subsequent drive after full pressing,

(Equation 12) Becomes

As described above, based on the above equation for calculating α, α
Is calculated and the tracking correction amount is obtained, accurate tracking is possible. Here, the denominator time is a past quantity, so the value is fixed, but the numerator is a future quantity, so the value is not fixed. Of these, the most uncertain is the lens drive time. However, if the approximate drive amount (DRIV) is calculated assuming α = 1, an approximate guide of the drive time can be obtained. When the time is determined, the time until the instant of the subsequent exposure can be determined, and therefore, the “time from the last accumulation to the next exposure”, which is the numerator of the above equation, is determined. In this case, the mirror-up is started by issuing a mirror-up instruction at the instant of the mirror-up time from the moment of the scheduled exposure to the morning. As a practical problem, as described in the embodiment, it is preferable to set the maximum value of the lens driving time constant after the release and to start the mirror-up after a fixed time (T1) after the start of the driving, which is preferable because the handling becomes simple. In this way, for PCOUNT3 and thereafter, α
Have the same value. In the case of PCOUNT2 and PCOUNT3, α may be the same or different depending on the case.
2 ′ = F3 ′ but the denominator is F1 = F2 and F1 ≠ F
Because there is a time of 2. If the lens drive starts immediately after the accumulation / calculation, F1 = F2.
≠ F2. Generally, the optimum value of α differs between PCOUNTs 0, 1, and 2, but the specific value differs due to the mirror-up time and other design fluctuation factors. The optimum value of α may be determined based on the equation for calculating α, or may be determined experimentally. In this embodiment, PCOUNTs 0, 1, 2, and 3 are identified using α thus determined.
The optimum value is determined with reference to the table of α (FIG. 17).

Next, a description will be given of a problem that occurs because the moving speed of the image plane is not constant with respect to a subject approaching at a constant speed. FIGS. 29A and 29B are more specific representations of FIG. 21B, and FIG.
In the case of a photographing lens of 0 mm, FIG.
3 shows the case of the taking lens. In each of the figures, the solid line shows the movement of the image plane when the subject approaches at 10 m / s, with the vertical axis representing a value obtained by subtracting a certain value from the distance between the lens and the image plane. The dotted line indicates the lens and the predetermined detection surface (conjugated to the film surface) when tracking is ideally performed.
The vertical axis represents the value obtained by subtracting the above-mentioned constant value from the distance from the distance, and at the instant of accumulation (black circle), the intersection with the solid line is achieved. As is clear from the figure, in the case of the long focal length lens in FIG. 29B, the speed of change of the image plane is almost constant,
The movement amount PRED of the subject image plane in one past cycle (from accumulation to accumulation) is used as is for the next estimated movement amount CO.
MP and therefore α = 1.

On the other hand, in the short focus lens (A), since the image plane moves away at a rapid acceleration, the value P
If the RED is used as the next expected amount COMP as it is, a tracking delay occurs as shown by a broken line a. Therefore, it is preferable to set α to be larger than 1 in COMP = PRED × α. In summary, when the predetermined focal length is FX and the focal length of the photographing lens is FL, for example, the following table is used.

[Table 6] If the tracking operation is slightly overridden for stabilization, the following table can be used as in the embodiment (the α table in FIG. 17).

[Table 7] As described above, the absolute value of the value of α differs depending on the value determination of the timing of focusing and the consideration of the stabilization, etc. By changing the value, an optimum tracking operation can be performed regardless of the focal length of the lens.

The effect of such correction is not conspicuous when the cycle time is as short as 100 ms or more, but the effect becomes large when the cycle time is 200 ms or more. In particular, the effect becomes extremely large when the cycle time reaches 300 ms due to mirror up. growing. Next, a case where the subject moves away will be described.
Whether the subject moves away or approaches can be easily determined by the PRED code. When the subject moves away, the speed is reduced as indicated by the dashed line in FIG. Therefore, if the tracking drive is performed as it is based on the past results, it tends to be overrun. Therefore, compared to when approaching, α
It is desirable to set the value of. Such acceleration / deceleration is more remarkable for a lens having a shorter focal length. In the embodiment, α in FIG.
Like a table, a small focal length (F
In the case of L <FX) and in the case of PCOUNT2 and PCOUNT3, which have a large deceleration effect even if the focal length is large, the value of α is set smaller when moving away from the case when approaching. There are various factors that require the change of α, such as the size of the focal length of the photographing lens, whether the subject approaches or moves away, before and after the release, and the size of the cycle time. Figure 1 shows the dependence on the time constant of the mechanical mechanism.
It is preferable to store an α table classified according to the case as shown in FIG. 7 and use a value of α according to the condition. The optimum value may be determined experimentally. If the body changes and the time constant of the mechanism changes, the optimum value of α also changes. However, in general, the optimal value range of α is

(Equation 13) Exists in the range.

(Third Embodiment) The third embodiment of the present invention has almost the same configuration as that of the first embodiment, and the focus display of the first embodiment is configured to be more easily understood. Hereinafter, FIG.
This will be described with reference to FIGS. In the AF CPU program focus determination / display module of FIG.
When tracking is in progress and the lens is not at the lens end as indicated by 590 to # 605, the display units 41 and 43 of the AF display unit 40 are both activated, and a display form different from the normal focus adjustment state display is given to the photographer. Letting you know you're tracking. The photographer knows that the subject is moving by this display, and responds to shooting of the moving subject, for example, adjusting the aperture,
You can select the shutter speed. Hereinafter, the display technique during tracking will be described using another embodiment. In # 605 of FIG. 22, the display during tracking is displayed by the AF display unit 4.
Although 0 is displayed, the tracking display may be displayed by a display unit other than the AF display unit 40 as another embodiment.

FIG. 30A shows the configuration of an embodiment in which the tracking display means 45 is separately provided. In FIG. 30A, the tracking display means 45 is controlled by the port P13 of the AF CPU 30 described with reference to FIG. The tracking display means 45 has a tracking display section 46. The tracking display section 46 is active only during tracking to notify the photographer that tracking is being performed. FIG. 31A shows a part of the program of the AFCPU in FIG. 7 of the above embodiment. FIG.
# 2010 in (A) is replaced with # 605 in FIG. Accordingly, when tracking is in progress and the lens is not at the lens end, # 20
By 10, the display section 46 of the tracking display means 45 is activated. On the other hand, the illustration in the program flowchart is omitted here, but in the above embodiment, if the camera is not being tracked or the lens is at the lens end, the display section 46 of the tracking display means 45 is turned off to indicate that the camera is not tracking. .
After # 2010, AF display by the AF display means 40 may be performed as shown in FIG. 22 according to the current defocus amount DEF, or all the display units of the AF display means 40 may be turned off.

FIG. 30B shows the configuration of an embodiment in which the tracking display means 45 displays a display during tracking and a display indicating the moving direction (approaching or moving away) of the subject. In FIG. 30B, the tracking display means 45 has display portions 47 and 48, and is controlled by the AFCPU 30 via the port P13. When the display unit 47 is active, it indicates that the subject is moving away, and when the display unit 48 is active, it indicates that the subject is approaching. When either of the display units 47 and 48 is active, it indicates that tracking is in progress. FIG. 31B shows a part of the AFCPU program of the above embodiment. # 2015 to # 2025 in FIG. 31B are replaced with # 605 in FIG. Accordingly, if tracking is being performed and the lens is not at the lens end, the sign of the tracking drive amount DRIV is tested in # 2015.
The display unit 48 is activated, the display unit 47 is turned off,
Indicates that the subject is approaching during tracking. # 2
If the sign is positive in 015, the process proceeds to # 2025 and the display unit 4
7 is activated, and the display unit 48 is turned off to display that the subject is moving away during tracking.

On the other hand, although illustration in the program flowchart is omitted here, in the above-described embodiment, if tracking is not being performed or the lens is at the lens end, both the display units 47 and 48 of the tracking display means 45 are turned off to perform tracking. Display that it is not inside. The processes after # 2020 and # 2025 are the same as in the previous embodiment. In the above embodiment, since the photographer can know the direction of tracking, the photographer himself cancels the tracking in the direction not intended by the photographer by operating another means (for example, half-pressing the release button or a dedicated button). It becomes possible. In FIG. 30B, the display units 47 and 48 for displaying the moving direction of the subject are shown as triangular display marks, but may be other display marks. For example

[Equation 14] Can be used as the display units 47 and 48.

FIG. 30C shows a configuration of an embodiment in which the tracking display means 45 displays the focus adjustment state during tracking.
In FIG. 30C, the tracking display means 45 is
3, 54, and 55, each having a port P of the AFCPU 30.
13 is controlled. The active states of the display units 53, 54, and 55 indicate the front focus, focus, and rear focus in the tracking state, respectively. FIG. 31C shows a part of the program of the AFCPU of the above embodiment. # 2030 in FIG. 31 (C)
Step # 2085 is replaced with step # 605 in FIG. Therefore, if tracking is in progress and the lens is not at the lens end, # 2030
Is tested first after the first or subsequent release, and if not, # 2035 to # 205
Step 5 is performed.

FIG. 32A shows an ideal trajectory (solid line) of the photographing lens and an actual driving trajectory (dashed line) of the photographing lens with respect to the moving object when the release is not performed. The photographing lens is driven to be tracked with a driving sequence as one cycle. When the tracking driving of the photographing lens is ideally performed, the solid line and the dashed line intersect at the middle point Im of the accumulation time of the image sensor. Therefore, in this case, the defocus amount DEF obtained from the accumulation during tracking should be zero. # 2035- #
At 2055, focusing, front focus, and rear focus are determined according to the value of the defocus amount based on the above concept. # 203
At 5, it is tested whether the absolute value of the defocus amount is larger than a predetermined value ZONEF. In general, the accuracy of the defocus amount obtained during tracking is lower than the accuracy of the defocus amount obtained during stillness. Therefore, the value of the predetermined value ZONEF is set to the predetermined values Z1, Z2, Z3 for focus determination shown in FIG. , Z4
The display is more stable if it is set larger than Z5. # 2
If 035 is smaller than the predetermined value ZONEF, # 20
At 40, the display unit 54 is activated to indicate that the focus is being tracked. If the sign is larger than the predetermined value ZONEF in # 2035, the flow advances to # 2045 to test the sign of the defocus amount DEF. If the sign is positive, the display unit 53 is activated to display the previous pin being tracked. # 204
If the sign is negative at 5, the display unit 55 is activated at # 2055 to display the rear focus pin during tracking. # 2
If it is determined in 030 that it is after the release, # 20
60 to # 2085 are performed.

FIG. 32B shows an ideal trajectory (solid line) of the photographing lens and an actual driving trajectory (dashed line) of the photographing lens with respect to the moving subject when the release is performed. The photographing lens is driven to be tracked with a sequence consisting of calculation and driving as one cycle. When the tracking drive of the photographing lens is ideally performed, the solid line and the dashed line intersect at the middle point of the photographing operation (exposure). Therefore, in this case, the defocus amount obtained from the accumulation during tracking, that is, the amount corresponding to the difference between the solid line and the one-dot chain line at the middle point Im of the accumulation time in the drawing does not become 0, and ideally during the shooting operation. The amount H that the taking lens has moved along the ideal locus (solid line) during the time between the point Em and the middle point Im of the accumulation time.
X.

The quantity HX is the midpoint of the current accumulation time (for example, I
_m1 ) and the middle point of the previous accumulation time (for example, I _mo ), the amount of actual movement of the taking lens is represented by DLST (this corresponds to the previous drive amount), and the middle point of the current accumulation time (I _m1) ) And the middle point of the previous accumulation time ( _Imo ) is TE, the middle point of the current accumulation time ( _Im1 ) and the middle point of the current shooting operation ( _Em1 ).
If the time between is TD, the following equation is obtained. HX = DLST × TD / TE (14) DLST can be obtained by storing the previous drive amount, and TD and TE can be obtained by measuring the time with a timer or the like built in the AFCPU. Therefore, # 2060- #
At 2085, focusing, front focus, and rear focus are determined according to a value obtained by subtracting a predetermined value HX from the defocus amount based on the above-described concept. In # 2060, H is calculated according to equation (14).
Calculate X. In # 2065, it is tested whether or not the absolute value obtained by subtracting HX from the current defocus amount is larger than a predetermined value ZONER. The predetermined value ZONER is the aforementioned predetermined value Z.
Set the same as or slightly larger than ONEF. If the value is equal to or less than the predetermined value ZONER in step # 2065, the flow advances to step # 2070 to activate the display unit 54 to indicate that the in-focus state is being tracked. If the value is equal to or more than the predetermined value ZONER in # 2065, the sign of the amount obtained by subtracting HX from the defocus amount is tested in # 2075, and if the sign is positive, # 208
When 0, the display unit 53 is activated to indicate that the front pin is being tracked. If the sign is negative in # 2075, the process proceeds to # 2085 to activate the display unit 55 to indicate that the rear pin is being tracked. # 2040, # 2
050, # 2055, # 2070, # 2080, # 20
The processes after 85 are the same as those in the previous embodiment, but it is desirable that all the display units of the AF display means 40 be turned off.

On the other hand, although illustration in the program flowchart is omitted here, in the above-described embodiment, if tracking is not being performed or the lens is at the lens end, the display units 53, 54 and 55 of the tracking display means 45 are all turned off and tracking is performed. Display that it is not inside. In the above embodiment, # 2035, # 2
At 065, the in-focus determination was made by comparing with the predetermined values ZONEF and ZONER. However, as shown in FIG. 21, the speed of the ideal lens generally decreases with time for a subject moving away at a constant speed. For a subject approaching at a constant speed, the speed of the ideal lens increases with time. Therefore, the focus determination zone of the sign-(front focus, approaching) is replaced with the focus determination zone of the sign + (back focus, moving away). If the value is set larger than this, the display becomes more stable. In the above embodiment, the photographer can recognize whether or not the subject is being tracked, and can know the focus adjustment state during the tracking. Therefore, the release operation according to the focus adjustment state, for example, the release operation if the subject is in focus during the tracking. Can be performed. In the above-described embodiment, the tracking display means 45 is provided separately from the AF display means 40. However, if it is not necessary to provide information as to whether or not tracking is being performed, the AF display means 40 also serves as the tracking display means 45. it can.

In the case described above, the focus adjustment state display is normal (even when tracking is not performed) even during tracking.
Because it is performed in the same form as
The display during tracking can be stabilized. FIG.
2. In the representation of the display unit in FIG. 31, the solid line is active,
The broken line indicates that it is OFF. FIG.
In the embodiment shown in FIG. 31, whether or not the vehicle is being tracked is displayed by the display means 45. However, the present invention is not limited to this. . In addition to displaying that tracking is being performed, other operating means of the camera can be automatically controlled when tracking is being performed. For example, when tracking is in progress, it is possible to automatically increase the shutter speed or reduce the aperture.

As described above, according to the present embodiment, the presence / absence of the movement of the subject is displayed, so that the photographer can recognize the movement of the subject and can appropriately cope with the moving subject, and can perform the conventional movement. Displays only the focus adjustment state of the shooting lens for the subject to prevent failure when the photographer cannot distinguish between a still subject and a moving subject, such as blurring when shooting with the same accuracy be able to. Also, if a member for displaying the focus adjustment state of the photographing lens and a member for displaying the moving state of the subject are used, there is an advantage that the cost is not increased and the photographer is not confused.

According to the present invention, the predetermined value for determining the movement of the subject is changed depending on the conditions (characteristics of the optical system, reliability, the result of the previous determination, the number of frames during continuous shooting, etc.). The transition from the normal drive state to the tracking drive state is performed stably, and the tracking drive state can be maintained stably.

[Brief description of the drawings]

FIG. 1 is a configuration diagram of a first embodiment of the present invention.

FIG. 2 is a configuration diagram of an AF module.

FIG. 3 is an operation time chart of the embodiment according to the present invention.

FIG. 4 is an operation time chart of the embodiment according to the present invention.

FIG. 5 is a program flowchart of a main CPU.

FIG. 6 is a program flowchart of a main CPU.

FIG. 7A is a program flowchart of an AFCPU. (B) The program flowchart figure of AFCPU.

FIG. 8 is a program flowchart of an AFCPU.

FIG. 9 is a program flowchart of an AFCPU.

FIG. 10 is an explanatory diagram for a program of an AFCPU.

FIG. 11 is a program flowchart of an AFCPU.

FIG. 12 is a program flowchart of an AFCPU.

FIG. 13 is a program flowchart of an AFCPU.

FIG. 14 is an explanatory diagram of a focus detection calculation.

FIG. 15 is an explanatory diagram of a focus detection calculation.

FIG. 16 is a program flowchart of an AFCPU.

FIG. 17 is a program flowchart of an AFCPU.

FIG. 18A is an explanatory diagram for a program of an AFCPU. (B) Explanatory drawing for the program of AFCPU.

FIG. 19 is an explanatory diagram for a program of an AFCPU.

FIG. 20 is an explanatory diagram for a program of an AFCPU.

FIG. 21A is an explanatory diagram for a program of an AFCPU. (B) Explanatory drawing for the program of AFCPU.

FIG. 22 is a program flowchart of an AFCPU.

FIG. 23 is a program flowchart of an AFCPU.

FIG. 24 is a program flowchart of an AFCPU.

FIG. 25 is a program flowchart of the AFCPU.

FIG. 26 is a program flowchart of the AFCPU.

FIG. 27 is a program flowchart of an AFCPU.

FIG. 28 is an explanatory diagram of a conventional tracking operation.

FIG. 29A is a diagram illustrating a locus of the photographing lens when the focal point of the photographing lens is different in the second embodiment. FIG. 6B is a diagram illustrating a locus of the photographing lens when the focal point of the photographing lens is different in the second embodiment.

FIG. 30A is a diagram for explaining the configuration of the third embodiment. (B) The figure for explaining the composition of the 3rd example. (C) The figure for explaining the composition of the 3rd example.

FIG. 31A is a diagram for explaining the configuration of the third embodiment. (B) The figure for explaining the composition of the 3rd example. (C) The figure for explaining the composition of the 3rd example.

FIG. 32A is a diagram showing a locus of a photographing lens with respect to a moving subject when a release is performed and when a release is not performed; FIG. 5B is a diagram illustrating a locus of the photographing lens with respect to the moving subject when the release is performed and when the release is not performed.

[Explanation of symbols]

 Reference Signs List 10 lens 11 photographing lens 12 lens-transmission system 13 lens-CPU 20 body 23 AF module 24 focus detection optical system 25 CCD 26 sensor control means 30 AFCPU 40 AF display means 50 AF motor 51 body transmission system 52 encoder 60 release button 61 frames Speed mode selecting means 62 focus mode selecting means 70 main CPU

──────────────────────────────────────────────────続 き Continued on the front page (72) Inventor Shozo Yamano 1-6-3 Nishioi, Shinagawa-ku, Tokyo Nikon Oi Works Co., Ltd. (56) References JP-A-62-253107 (JP, A) ( 58) Field surveyed (Int. Cl. ⁶ , DB name) G02B 7/11

Claims (5)

(57) [Claims] A focus detection unit configured to repeatedly detect a focus adjustment state of a subject image formed by a main optical system, and to generate a focus detection signal in a time series according to the detected focus adjustment state; Movement information detecting means for detecting movement information relating to a time change of a subject image plane due to movement of a subject based on a plurality of focus detection signals which are generated periodically, and determining whether the subject has moved by comparing the movement information with a predetermined value Subject movement determining means, and driving means for driving the main optical system according to a drive control method according to a result of the determination by the subject movement determining means, wherein the subject movement determining means changes the predetermined value. An automatic focus adjustment device. 2. The automatic focusing apparatus according to claim 1, wherein said predetermined value is changed according to an optical characteristic of said main optical system. 3. The automatic focusing apparatus according to claim 1, wherein the predetermined value is changed in accordance with a result of the previous determination of the movement of the subject. 4. The image processing apparatus according to claim 1, further comprising: reliability calculating means for calculating reliability of the focus detection signal generated by the focus detecting means, wherein the predetermined value is changed according to the reliability. Automatic focus adjustment device. 5. Apply the camera equipped with imaging means for continuously shooting a subject image formed by the main optical system, and an operation means for instructing continuous shooting operation by the shooting means
2. The automatic focus adjustment device according to claim 1, wherein the subject movement determining unit changes the predetermined value according to the number of times of photographing by the photographing unit after a continuous photographing operation is instructed by the operation unit. .