JPH0215215A - Automatic focusing camera - Google Patents

Abstract

PURPOSE:To accurately follow and correct a lens driving amount with respect to a dynamic object by correcting a focusing position according to a value obtained by multiplying the changing speed of defocus quantity due to the movement of the object by a prescribed constant corresponding to a prescribed time including a mirror-up time and to the direction of the changing speed. CONSTITUTION:A focus detecting means 1 detects the focusing state of the photographing lens and outputs the defocus quantity DFO. A changing speed detecting means 2 detects the changing speed VHO of the defocus quantity due to the movement of the object based on an output from the focus detecting means 1. A correcting amount determining means 3 determines the correcting amount DELTADF of the defocus quantity by multiplying the changing speed VHO by the prescribed time including the mirror-up time and the prescribed constant corresponding to the direction of the changing speed VHO. Based on the defocus quantity detected by the focus detecting means 1 and the correcting amount DELTADF determined by the correcting amount determining means 3, a lens driving means 4 drives a focusing lens toward the focus position.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention relates to an automatic focusing camera that drives a focusing lens toward an in-focus position according to a focus detection result. It is particularly suitable for reflex cameras.

[Prior Art] In conventional automatic focusing cameras, the focusing accuracy when photographing a dynamic subject is lower than the focusing accuracy when photographing a still subject. This is because if the subject is moving, even if the lens is driven to the in-focus position at the time of current focus detection and released, the subject will deviate from the in-focus position until the lens is released. Therefore, JP-A-60-
As disclosed in Japanese Patent No. 214325, it has been proposed to perform tracking correction in which the lens is driven in accordance with the defocusing speed of the subject when the subject is moving. In this conventional example, the amount of tracking correction is determined by multiplying the defocusing speed of the subject by the time for performing tracking correction. Further, although the defocus speed of the subject is determined by the weighted average method, it is stated that it may also be determined by the quadratic curve approximation method.

[Problems to be Solved by the Invention] As mentioned above, if the defocus speed is determined using the weighted average method, it should be possible to eliminate the variation in the detected defocus speed and accurately determine the defocus speed. However, this means that the past defocusing speed of the subject is accurately determined, and it does not mean that the defocusing speed of the subject when performing tracking correction is accurately determined. For example, when a subject approaches a photographer at a constant speed, the defocusing speed of the subject increases approximately in proportion to the square of time. For this reason, if the tracking correction amount is determined based on the past defocusing speed of the subject determined by the weighted average method, there is a problem that a tracking delay occurs. On the other hand, when the subject moves away from the photographer at a constant speed, there is a problem that the tracking correction amount becomes excessive.

Therefore, as suggested in the above-mentioned prior art, it is possible to calculate the defocusing speed of the subject when performing tracking correction using the quadratic curve approximation method, but this The calculation requires focus detection three times, which complicates the calculation and causes a tracking delay.

The present invention has been made in view of the above points, and its purpose is to provide an autofocus camera with a simple configuration that can accurately follow and correct the amount of lens drive for a dynamic subject. be.

[Means for Solving the Problems] In order to solve the above-mentioned problem =3, in the autofocus camera according to the present invention, as shown in FIG. In a camera, there is a focus detection means (1) that detects the amount of defocus (defocus DFO) of a photographic lens with respect to the subject to be focused on, and a change rate V of the amount of defocus due to movement of the subject.
A speed-of-change detection means (2) for detecting HO, and the speed-of-change detected by the speed-of-change detection means (2) are multiplied by a predetermined time including the mirror-up time and a predetermined constant depending on the direction of the speed of change. Correction amount ΔDF of focus shift amount due to subject
a correction amount determining means (3) for determining the amount of correction; and a focus detecting means (1).
) Defocus amount and correction amount determination means (3) detected by
The present invention is characterized by comprising a lens driving means (4) for driving a focus adjusting lens toward a focusing position based on the correction amount determined by.

Note that a determining means (5) for determining whether or not the subject is a dynamic subject based on the rate of change detected by the rate of change detecting means (2) is provided, so that it is determined that the subject is a dynamic subject. It is desirable that the correction amount determining means (3) be put into operation only when this occurs.

Further, the predetermined constant is set to be greater than 1 when the rate of change in the amount of defocus due to the movement of the subject is in the direction in which the subject approaches the photographer, and is set to be greater than 1 when the rate of change is in the direction in which the subject moves away from the photographer. It is desirable to set it small.

However, FIG. 1 is an explanatory diagram showing the configuration of the present invention in functional blocks, and in the embodiments described later, "1"
The main parts of the configuration described above are realized by a microcomputer program. Specifically, the focus detection means (1) is placed at 1107, #], 13 in FIG. 7, the change rate detection means (2) is placed at #313 in FIG. 10, and the correction amount determination means (3 ) are #401 to #409 in FIG. 12, the lens drive means (4) are #35 in FIG. 11 and #410 to #416 in FIG. 12, and the determination means (5) is #409 in FIG.
329 respectively.

[Effect] Hereinafter, the operation of the present invention will be explained with reference to FIG.

The focus detection means) detects the focus state of the photographing lens with respect to the subject to be focused on, and outputs the amount of defocus (defocus DFO>).
is a variable whose sign indicates the direction of defocus and whose absolute value indicates the magnitude of defocus.

The change speed detection means (2) detects the change speed V of the amount of defocus due to the movement of the subject based on the output of the focus detection means (1).
I-(0 is detected.The correction amount determining means (3) determines the correction amount DF for the defocus mouse by multiplying this change rate V HO by a predetermined time T including the mirror up time and a predetermined constant. In a reflex camera where the mirror up time is approximately constant, even if there is some variation in the time required for focus detection, the delay time T from the center time of focus detection to the release timing is approximately constant. Therefore, even if the rate of change VIIO changes nonlinearly over time, the correction amount ΔDF can be adjusted by multiplying it by a predetermined constant.
can be corrected with high precision. The predetermined constant varies depending on the direction of the change rate VI (0, and is set larger than 1 when the subject approaches. This prevents insufficient tracking. Also, when the subject moves away In this case, the predetermined constant is set smaller than ].This prevents excessive tracking.The lens drive means (4) uses the defocus DFO detected by the focus detection means (1) and the correction amount determination means. In (3), the focus adjusting lens is driven toward the in-focus position based on the determined correction amount ΔDF. The camera is activated only when it is determined that the subject is a dynamic subject.

[Example] Hereinafter, embodiments of the present invention σ) will be described based on the drawings.

FIG. 2 shows a single-lens reflex camera with interchangeable lenses, where 10] is a camera body, and 102 is a zoom lens, which is an example of an interchangeable lens (taking lens). Reference numeral 103 denotes a main mirror, which has a reflecting section and a transmitting section. The light that has passed through the photographic lens is reflected by the reflecting portion of the main mirror 103 and guided to a finder optical system (not shown), while a portion of the light is transmitted and guided to the submirror 104. Submirror 104 is main mirror 1
03 is reflected to the focus detection module 105. FIG. 3 shows the camera body 10 viewed from the front. As mentioned above, ]0] is the main mirror of the camera body 103, 105 is the focus detection module, and 106 is the mechanical unit 1 configured to automatically perform mirror up, exposure operation, film winding, and rewinding. .

Since these are not directly related to the present invention, their explanation will be omitted.

FIG. 4 shows a circuit diagram of a camera to which the present invention is applied. Reference numeral 201 is a microcomputer for controlling the camera, which performs functions such as sequence control of the entire camera, calculation control of exposure, and calculation control of AF 1 to focus (hereinafter abbreviated as AF). It is equipped with various input/output terminals P1 to P21, etc. Reference numeral 202 is an AF distance measuring unit that measures the amount of defocus of a subject image, and includes a one-dimensional self-scanning image sensor (hereinafter abbreviated as CCD), a CCD drive unit, an A/D conversion unit, and a reference power source for A/D conversion. Consists of sources etc. The image information obtained by this CCD is A,
The data is taken into the CPU 20 via the F data bus 201a. 203 is a liquid crystal display (1, , , C
D) Or, there is a display section consisting of a light emitting diode (LED), and the shutter speed Tv and O are the calculation results of automatic exposure (hereinafter abbreviated as AE) sent from the CPU 20.
- Information such as aperture value A, v, in-focus/out-of-focus, or shooting mode 1~ is displayed by this display section 203. 20/l is a lens data circuit that is provided inside each interchangeable lens 102 and stores the maximum aperture value, minimum aperture value, focal length, extension amount conversion coefficient necessary for focus adjustment, etc. When mounted on the camera body 101, the data is transmitted to the camera body 101 via electrical contacts provided near the mounting portion. 205
is a photometry unit that measures the brightness Bv of the subject, and includes a photoelectric conversion element for light reception, an A/D conversion unit, a reference voltage source for A/D conversion,
It consists of a data exchange unit etc. with the CPU20], and
20], the light passing through the photographic lens is photometered. A film sensitivity reading section 206 automatically reads the sensitivity of the loaded film, and the film sensitivity on the film cartridge is read through an electrical contact provided in the cartridge chamber of the camera.

The display section 203, lens data circuit 204, photometry section 2
05, each piece of information from the film sensitivity reading section 206 is input as a serial signal to the serial input/output section 201c (abbreviated as serial I10 in the figure) via the serial data bus 201+1. Reference numeral 207 is a sequence motor M8 for winding and rewinding the film, an AF momo-M2 for driving the lens for AF, and a driver control unit for exciting various magnets required during exposure operation, and is an output terminal of the CPU 201. It is controlled by control output lines CMDO to CMD8 from P8 to P16. SW1～
SW3, SW5~5WIO each has a switch,
One end of these switches is grounded, and the other end is connected to input terminals P1 to P7, P2O, and P21, respectively. SW
l is a switch that turns ON when film charging starts and turns OFF when film charging is completed; SW2 is a switch that turns ON while the mirror is up, and turns OFF when mechanical charging is completed; SW3 is a switch that turns ON multiple times while the film is running.
This is a switch that repeats F. SW5 is a photometry switch that is turned on at the first step of pressing down a jack button (not shown), and the CPU 201 outputs a signal to start photometry and distance measurement. While this switch SW5 is ON, if the lens is in an out-of-focus position due to distance measurement, the lens continues to be driven, and when it reaches the in-focus position, the lens stops driving, but while the lens is being driven, the shutter When the button is released and the switch SW5 is turned off, the driving of the lens is stopped.

SW6 is a release switch that is turned ON in the second step of pressing down the shutter button, and when the release is possible, if this switch SW6 is turned ON, the CPU 20
1 commands a release operation. Note that when the release switch SW6 is turned on, the photometry switch SW5 is configured to be kept in the on state. SW7 is a film detection switch provided in the film running path,
When there is a film at the film detection switch SW7, the switch SW7 is OFF, and when the film runs out, it is turned ON. When rewinding, if this switch SW7 changes from OFF to ON, the film is
This is used as a switch to determine the end of rewinding, indicating that the rewind is slightly out of the roll.

SW8 is a cartridge detection switch provided near the electrical contact of the film sensitivity reading unit 206 provided in the cartridge chamber of the camera, and is turned ON when a cartridge is in the cartridge chamber and the back cover is closed. If there is no patrone, it will be in OFF state. SW9 is the back cover opening/closing switch, and it turns O when the back cover is completely closed.
It becomes N. 5WIO is a multiple exposure mode changeover switch, and when it is turned on, the multiple exposure mode is set.

RESET is set to control power supply voltage +Voo by resistor R1.
After the power is turned on, capacitor C1 is charged via resistor R1, and its voltage changes from l L oIII+ level to "High".
``When the level changes, the CPU 201 is reset.X is a crystal oscillator for giving a clock signal to the CPU 201.

Next, the driver control section 207 and each control section will be explained. ICMg is a magnet for holding the shutter 1 curtain, and when the control output line ICMGO reaches the "LOL11" level, the output from Macnet 1 to ICMI? is energized, and the first shutter curtain is held. 2CMg has a magnet for holding the second shutter curtain, and the control output line 2CMGO is “
When it reaches I L oulI+ level, Macunet 1~
2CMg is energized, the second shutter curtain is held, and the time from when the first curtain shutter is released to when the second curtain shutter is released corresponds to the shutter speed. FMg is a micnet for locking the diaphragm of the photographic lens. When the control output line FMGO goes to "Low" level, energization is applied to macnet 1 to FMg to hold the diaphragm locking member and release the holding. The aperture locking member then operates to lock the aperture in a predetermined position. RMg is a magnet for release, and the control output line RMGO is connected to "Lou" for a certain period of time.
+”When the release is reached, the release member is unlocked,
The aperture is stopped down and the mirror is raised.

Q1~QIO is Z-Kensmoke M1 and AF Momo-M
This is the second driving transistor. This Z-Kensmoke M1 has two types of coils inside, and can obtain the characteristics of high torque low speed rotation and low 1 Herc high speed rotation, and it is possible to switch between both characteristics, and the forward and reverse rotation of each is possible. Transistors Q1-Q6 are connected as possible. Zunawaji, high speed side terminal 1-1 of sequence motor M1
is connected to the common connection point of transistors Q] and Q2, the low-speed side terminal is connected to the common connection point of transistors Q3 and Q4, and the remaining common terminal C is connected to the common connection point of transistors Q5 and Q6. Ru. Table 1 shows I-transistor Q1~
This figure shows how the rotational state of the Z-ken smoke M1 changes depending on the on/off state of Q6.

(Margin below) Table 1 Note that in non-implemented examples, the high-speed brake is not used, and only the low-speed brake is used. Therefore, in the following explanation, the term "brake" refers to the low-speed brake SBR. Q7 to Q1.0 are one-transistor resistors for driving the AF motor M2, and are connected in a bridge configuration so that the A and F motors M2 can be rotated in forward and reverse directions. AF Momo-M
Step 2, rotate forward to extend the lens I~, then reverse to retract the lens. 0M1~OM], 0 is each 1~ransistor Q1~
This is a control output line for switching of Q1.0.

211,. 212 is an aperture encoder and an AP encoder made of photocouplers, and input signal lines PT1.
It is connected to the driver control unit 207 by PT2. Aperture encoder 2]] is the aperture preset 1 at the time of release.
. . . which monitors the stroke of the lever. At the time of release, the light emitted by the light emitting diode 2 ] 1 a is detected by the photo 1 - transistors 2 ], ], +1, and is input to the driver control unit 207 via the input signal line PT1. After being waveform-shaped into a pulse by this driver control unit 207, it is sent to the CPU via an output signal line FP.
201 is sent to input terminal P18. AP encoder 2] 2 is the AF momo-M2 for driving the lens that goes down during AF
The light emission from the light emitting diode 212a is detected by the photodiode 212a to the transistor 212b, and is input to the driver control unit 207 via the input signal line PT2. After being waveform-shaped into a pulse by this driver control unit 207, it is sent to the CPU via an output signal line AFP.
201 is sent to the input terminal P], 9. These output signal lines A and FP are connected to the internal counter 20 of CPtJ201.
], d are also connected, and are used to monitor the extended position of the photographic lens. In other words, the counter 201d is cleared to 0 at the ω end of the lens, and by setting it to up run 1~ when driving in the near force direction and to down run 1~ when driving in the ■ direction, the counter 201d can be cleared from the ■ end of the lens at any time. The number of delivery pulses can be obtained. This AF
The P signal is also connected to an interrupt terminal (not shown) of the CPU 20, and an interrupt is generated when the A, FP signal falls. Further, the CPU 201 has a built-in timer 201e, and is configured to be able to read the time by counting an internal clock. Furthermore, CPU2
01 is a so-called E2PROM that can be written and read electrically and retains memory contents even when the power is turned off.
20] Built-in Go. In addition, CPU 201. is equipped with an interrupt timer (not shown) that generates a timer interrupt when a set time has elapsed.

Table 2 In the table, H is 'High' level L means "Low'" level.

No. 3 tables In the table, H is 'High' level L means 'Lou+''' level.

CMDO to CMD8 are control output lines output from output terminals P8 to P16 of the CPU 201 to control the driver control unit 207, and control output lines RM for controlling magnets RMg and FMg, respectively, are provided by CMDOCMDI.
Control output lines ICMGO and 2CMGO for controlling magnets ICMg and 2CMg are controlled by CMD2 and CMD3, respectively. Also, CM
Control output lines ○M1 to oM6 for driving the sequence motor M1 are controlled by D4 to CMD6, and CMD7. CMD8
Control output line OM7 to 0M for driving AF Momo-M2
Control l0. Table 2 shows the control of the sequence motor M, and Table 3 shows the control of the AF motor M2. In the table, H means II Hi gl, I+ level, and L means "Lou+ level."

AMg is a lock release magnet that holds the film still, and is a magnet for releasing the film. CPU2 via resistor R2
0] is connected to the output terminal P17. ■・Langister Q
The connection point between the base of ll and resistor R2 is grounded via resistor R3. The output terminal P17 of the CPU 201 is normally "'
Since the transistor Qll is in the OFF state, the magnet AM& is not energized and holds the attraction piece by attraction. In order to release the engagement between the winding stopper and the winding stopper lever, the CPU 201 Output terminal P17
When it reaches the 'High' level, magnets 1 to AMH are energized and the attraction force disappears.

Next, a series of release operations in this embodiment will be explained with reference to FIG. As shown in the figure, the release operation is roughly divided into four sequences: mirror up, exposure, mechanical charge, and film charge.

In the mirror-up sequence, the main mirror and submirror are retracted, and the diaphragm of the photographing lens is unlocked to narrow the diaphragm. In the exposure sequence, the exposure time (shutter speed) is controlled by controlling the first and second curtains of the focal plane shutter. In the mechanical charge sequence, for the next release, the main mirror, sub-mirror, aperture of the photographing lens, and the first and second curtains of the shutter are activated by (-1).

In the film charge sequence, the film is advanced.

A more detailed explanation will be given below using time chart 1. As already explained with reference to FIG. 4, the switch S W 6 is turned on by pressing down the shutter button two steps and starts the release operation. When the release operation is started, first, the control output line RM GO is set to "LouI" to energize the release Macnet RMB, and the main mirror, which is biased by the spring, is locked. unlock. As a result, the main mirror is retracted toward the finder' side, and the submirror is also retracted in conjunction with the main mirror. Next, control output line FMGO
By setting the aperture to Low level, the aperture locking magnet l□FMg is energized, and the aperture of the photographing lens, which is biased by the splash, is released from the lock. When the lock is released, the aperture starts to be narrowed down, but the state of the aperture at this time is inputted to the dryer control unit 207 from the monitor photo 1, Heransister 2, ], and b, as explained in FIG. C as a waveform-shaped FP Shinkyu
It is manually operated by PU201. The CPU 201 calculates a number l? corresponding to the aperture value based on a predetermined exposure calculation. Click the P signal 1~
Then, by setting the control output line FMGO to the 'Hi Hb' level, the aperture is stopped and the photographing lens is set to the desired aperture value.Next, in order to perform the exposure operation, the 'LouI' level is set at the start of the release. The control output line I becomes
One of the control output lines 1c of CMGO2CMGO
MGO is set to 1iHh" to I\. As a result, the first curtain of the focal plane shutter runs. After the exposure time according to the predetermined exposure calculation has elapsed, the other control output line 2CMGO is set to the 'I(igl+'" level). By doing this, the second curtain of the focal plane shutter runs,
Exposure control is performed. After exposure, a mechanical charge sequence begins. In the mechanical charging sequence, first of all, high torque is required when starting the sequence motor M, so it is driven in low speed mode F (L). Switch to high speed mode F(H). This allows for efficient Z-ken smoke M.
1, and can achieve high-speed mechanical charging and fill 18 charging.

As the sequence motor M1 rotates, the main mirror and the submirror are activated by the town and splash, and at the same time, the aperture of the photographing lens and the first and second curtains of the focal plane shutter are also activated by the splash. When this mechanical charging is completed, as already explained, the switch SW2 is turned off as a mechanical charging completion signal. C.P.
U 20 ] C When the OFF of this switch SW2 is detected, the process shifts to the film charging sequence. This releases the lock that fixes the film and starts winding the film. At this point, the switch SWI that monitors the film charge is turned on, and the
Film winding is performed by Kensmoke M.

When film winding for one frame is completed, switch S
WI is turned off and the CPU 201 is notified. C.P.
When U201 detects 0FFp of switch SWI, it applies a flake (not shown) to stop sequence motor M1. This completes the release operation for one frame.

Next, the sequence from continuous shooting mode 1 to continuous shooting mode in which the release is performed continuously while the switch SW 6 is ON will be explained, including the focus detection operation, with reference to the time chart of FIG. 6. Section I in FIG. 6 indicates the first frame of continuous shooting, and section H indicates the second frame of continuous shooting. Section I of the frame of the snapshot is as explained in FIG. By the way, to perform focus detection,
It is necessary that the main mirror and submirror are lowered and stable. Therefore, the switch SW2 is turned OFF, and after the mechanical charging is completed and a period of time for mirror stabilization is waited, CCD integration is started. In this embodiment, this waiting time is set to 30 m5ec. In the diagram■
1 The part not shown is the CCD integration time, and the part shown Dl is C.
It shows the data dump time for A/D converting the pixel data of the CD and importing it into the memory of the CPU 201. CCD
Once the pixel data is taken into the CPU 20, the reliability of defocus, defocus direction, and focus detection is determined by predetermined calculations. Since this focus detection calculation is not directly related to the present invention, a description thereof will be omitted. Now, when the first film charge is finished, switch S
W1 is turned off, and low speed brake SBR is applied to sequence motor M.

At this point, the CPU 201 determines whether the switch SW6 is ON. If the switch SW6 is ON and the camera is in the snapshot mode, then the release of the second frame, which is indicated by section 3, is started. In the case of this second frame, since the release is immediately after the film is charged, the low-speed play-'esB is performed using the Z-Kensmoke M1 to ensure that the film stops.
Wait for a while with R applied, and energize the release magnet RMg. Now, if the result obtained in calculation 1 is that the camera is in focus, the next time you release the camera with the photographic lens stopped, you will be able to take a photograph that is in focus. If you do, you'll end up with out-of-focus photos. In particular, when shooting in continuous shooting mode, there are many cases where the subject is moving and the defocus changes over time, so in calculation 1, the defocus due to the movement of the subject is Detect minutes. By performing this defocus drive while the mirror is up, it is possible to obtain an in-focus photograph even at the next release. The AF Momo-M2 column in FIG.
This indicates that the FM section is being driven in either direction. To compensate for the defocus amount obtained as a result of calculation 1 or for the movement of the subject, the photographing lens is driven during mirror up in the next section H. Furthermore, when the control output line RMGO is at the "Lou+" level, that is, when the release magnet "RM g" is energized, the AF MOMO-M is de-energized. The current flowing through the magnet RMg for release is very large, and if AF MOMO-M2 is driven during this time, the driving accuracy of AF MOMO-M2 may drop, or the current flowing through the release magnet RMg may decrease. This is to avoid the problem that the mirror may not be able to be raised due to the mirror not being locked.

Thereafter, similarly to the first frame, the release is continued while the switch SW6 is ON.

Next, the operation of this embodiment will be explained using the flowcharts from FIG. 7 onwards. FIG. 7 is a flowchart when the photometry switch SW5 mentioned above is turned on. When the switch SW5 is OFF, the camera is in a low power consumption mode, a so-called sleeve mode. When the switch SW5 is turned on, clock oscillation starts and is activated. C
When the PU 201 is activated, it sends an activation signal and a clock to the peripheral IC in #101, and performs activation processing such as initializing the boat. Next, in #102, flags, constants, etc. used in the program are initialized. continue,#
At 103, switches in each part of the camera body are checked, and serial communication is performed with the flash, lens, display element, etc. Furthermore, in order to discharge unnecessary charges from the CCD, which is a focus detection element, initialization is performed at #104.

Proceed to the next step, focus detection processing CD I NTA (#105 onwards). First, prior to CCD integration, #10
In step 6, the integration start time is input to the memory TMI of the CPU 201 from the timer and saved.

Similarly, the counter indicating the lens position mentioned above is read and saved in the memory T1. Thereafter, in step #1.7, CCD integration is performed to obtain a signal level suitable for focus detection. When the CCD integration is completed, the timer value is saved in the memory TM2 and the counter value is saved in the memory T2 in #108. Next, #1094 Trowel Memory TML
Save the value of memory TM in memory TM, and save the value in memory TM (TM
2-TMl)/2. TMI and TM2 indicate the integration start and end times, respectively, and (TM27MI
)/2 means the time of the center of integration. That is, #10
9, the previous integration center time is saved in the memory TML, and the current integration center time is saved in the memory TM. Similarly, regarding the counter value, the memory MIL is set at #110.
Memory M1 (+lf wo (knee 7, meter t=
Safe Lee M I ni (T 2 - T 1 )/2. As mentioned above, the counter value corresponds to the lens position, so Ti and T2 indicate the lens position at the start and end of integration, respectively, and (T2-Tl)/2 does not indicate the lens position at the center of integration. . That is, in #110, the lens position of the previous integration center is safed to M I L, and the current lens position of the integration center is safed to Ml.

Next, in #111, each pixel data of the CCD is sent to the CPU2.
0]. Using this CCD pixel data, before starting the focus detection calculation, the value of the memory DFO is saved in the memory LDF in #112. At this point, focus detection by CCD integration in #]07 is not performed, so in #112, the previously detected tefocus DF○ is stored in memory L.
This means that it is saved in DF. In #1]3, a focus detection calculation is performed based on the CCD pixel data integrated in #107, and the tefocus DFO and defocus direction of the photographic lens as well as the reliability of focus detection are calculated. Subsequently, in #1]4, serial communication and exposure calculation are performed, and photometric values are displayed, each switch is sensed, etc. For example, switch SW between #]07 and #]14
5 is turned off, it is detected at #1]4, processing such as turning off the display and turning off the motor is performed, and the state returns to the sleeve state. When serial communication and exposure calculation are completed, #1
Proceed to step 15 and use flag V to determine whether or not you are currently taking a snapshot.
It will be held at L Y F. This continuous shooting flag LYF is set to 1 to 2 when the photographer has selected the quick shooting mode during operation, as will be explained later. When the continuous shooting flag VL Y F is determined in step #1]5, the process branches to the flow of continuous shooting AF (#30). The flowchart of the quick-shooting AF will be explained in detail later using FIG. 10.

Now, if the continuous shooting flag V L Y F is 0, #11
Proceed to 6 and focus! A determination is made as to whether or not it is true. This focus judgment is made when the reliability of the focus detection in #1]3 is higher than a predetermined value and is smaller than a predetermined amount of [def]-cus DPO (in-focus is determined. In this case, the predetermined amount is , 100μIn. Also, this value is set to 60μm-1-8X (2(log2F
It may also be set as No+1>. Here, FNO is the F number of 1st aperture.

This predetermined amount is the E2 of the CPU 201 explained earlier.
It is written in PROM20 if. This makes it possible to write small values for users who want high precision, and large values for users who want feel and focusing time, allowing us to respond to users in detail.

If h black is determined in #1]6, proceed to #]17,
Focus is displayed. After in-focus display, exposure recalculation and serial communication are performed. This is done in order to reflect the focus detection result in the exposure calculation. Thereafter, it waits while performing serial communication in step #120 until it is detected in step #1]9 whether the switch SW6 is turned on.

If it is determined that focus cannot be achieved with #1,
Proceed to the subsequent out-of-focus processing ○UTFS.

First, in #122, the focus display is turned off, and then in #123, as a result of the focus detection calculation, if it is determined that the focus detection result is not used due to low reliability, that is, it is determined that the focus is low, the AF mode is turned off.
To perform the next focus detection without driving M2,
The process branches to focus detection processing CD INTA after #105. If there is no low contrast, proceed to #C24 and A
F motor M.

After waiting until the driving is completed in step #125, the next focus detection is performed, and the process proceeds to focus detection processing CD INTA starting from #105. In #124, the conversion coefficient K I- obtained from the photographing lens is applied to the TeFocus DFO.
The number of driving pulses ER of the AF Momo-M2 described above is multiplied by
Calculate RCNT.

In other words, the number of drive pulses ERR, CNT is ERRC
Determined by NT=DFOXIgL. #] 25 is A
The ARP signal for monitoring the rotation of F momo-M2 is detected for the number of drive pulses ERR, CN T.
F motor M2 is driven. Note that this conversion coefficient K
L varies depending on the focal length of each photographic lens, and is sent from the photographic lens through serial communication. Thereafter, the same process is repeated again from #105 to #1]6 until focus is determined.

Next, a description will be given of the processing after the ON of the switch SW6 is detected in #119. In #126, it is determined whether the drive pulse number ERRCNT is smaller than the constant NPI written in the E2PROM 201f (#126). The number of driving pulses ERRCNT is a constant NP
If the number is greater than I, a constant N is added to the drive pulse number ERRCNT.
Please reset the PI #127), drive pulse number ERRCN
If T is smaller than the constant NPI, leave it as is #]2
Proceed to 8. In #128, it is determined whether the defocus DFO determined to be in focus in #116 is smaller than the constant DFC1. If DFO<DFCl, the process branches to release (#131) to immediately perform a release operation. If DFO≧DFC1, proceed to #129,
It is determined whether the number of drive pulses ERRCNT is smaller than 4 pulses. If the number of driving pulses ERR, CNT is smaller than 4 nozzles, the process immediately branches to release (#131). #129 drive pulse number ERRCNT is 4
If it is equal to or greater than the pulse, driving of the AF momo-M2 is started in #130. In the processes from #126 to #1-31, it is determined whether or not to drive the photographing lens during mirror up.

As already explained, in #116, focus determination is performed based on whether the detected defocus is within a predetermined defocus range. If this focusing range is made very small, accuracy will increase, but it will take time to focus due to variations in focus detection, camera shake by the photographer, etc. Alternatively, problems such as small hunting of the photographic lens may occur. Furthermore, when the subject is a moving object, the detected defocus changes moment by moment, so no focus determination is made. For this reason, this focusing range is set to be wide enough to absorb variations in focus detection or changes in defocus due to changes in the subject. However, in this case, the defocus corresponding to the focusing range remains as an error. For this reason, this remaining defocus is used during mirror up to create an even more accurate autofocus camera. Also, in order to shorten the time until release, the number of pulses that can be driven during mirror up is limited to a constant NPI #126. #127>, if sufficient accuracy is ensured, that is, #128 and DFO<D
If it is FCI, or if #129 drive pulse number ERRCNT is smaller than 4, AF Momo-M2
The drive accuracy of the camera is also taken into consideration, and the shutter is released immediately.

Next, a series of release controls (from #201 onwards) including mirror up, exposure time control, mechanical charging, and film winding will be explained using FIG. First, #2
02, set the flag RMGONF to 1, which indicates that the release magnet 1-RMg is energized, and #20
Step 3 energizes the AF MOMO-M2 to 0FFL, and step #204 energizes the release magnet RMg, shutter curtain holding magnets) ICMg and 2CMg. Then #2
After waiting for a time in 05, the power to the release magnet RMg is set to 0FFL in #206, and the flag R is set in #207.
Clear MGONF to 0. #20 from this #204
By the process 6, the main mirror submirror is unlocked and the mirror-up is started. Also, #202, #20
3. Through the process of #207, energization to AF Momo-M2 is prohibited while power is being energized to Macnet 1 to RM g. Specifically, AF Momo-M is controlled by turning on AF Momo-M2 by a timer interrupt that occurs every predetermined time.
This is done by braking by interrupting the FP signal, and when the flag RMGONF is 1, turning on AF momo-M2 and braking are both prohibited, and AF momo-M2 is disabled.
M2 is turned off.

When the flag RMGONF is reset, driving of the AF momo-M2 is automatically restarted by a timer interrupt. The above process requires a large current to energize the RMg for the release, and by energizing the AF MOMO-M2, the current flowing to the RM8 will not be released, and the mirror will not be released. This is done to prevent inconveniences such as not being able to upload.

When the mirror-up is started, then in #209, the aperture locking pins 1 to FMg are energized.

As a result, the diaphragm of the photographic lens is unlocked and the aperture of the photographic lens is started. At #210, the pulses of the aperture encoder 21 mentioned above are counted and the pulses set by the exposure calculation are divided, and at #2]1 the power to the lens 1 to FMg is turned off to fix the aperture.

After this, use Macnet RM g for release at #212.
Wait for a predetermined time t1 to elapse from energization to #21
Proceed to step 3. In #213, in order to prevent the photographing lens from being driven during exposure, the power to AF Momo-M is turned off to 0FFL, and then in #214, Z-Kensmoke M is turned off.
Make it F. As already explained, the Z-Kensmoke M1 has a flake applied that must be turned off in step #214 for the second and subsequent frames of quick shooting.

#215 to magnet 1 for holding shutter 1 curtain lc
Turn on electricity to Mg? By doing li', one curtain of the focal brain shutter runs, and when the shutter speed is reached at #2]6, IH? Wait, #2]7
Then, the magnets 1 to 2 CM for holding the second shutter curtain are energized ('11 FF), and the second curtain of the focal brain shutter is run. With this, the exposure is completed.

#218 Save the timer value to memory] "IME". #2] 9 Save the timer value to memory]
In order to wait for the second act or the run to be completed, the Ll
Wait for B to pass and wind up as routine (#220)
Move to.

In the winding routine shown in Figure 9, first #22]
energize Siken Smoke M1 at low speed mode 1-,
After waiting for a predetermined time until the rotational speed of the motor M1 increases in #222, the high speed mode is energized in #223. As a result, the mechanical charging progresses, and the process waits until the switch SW2, which is the mechanical charging completion signal, is turned OFF in #224.

When the switch SW2 is turned OFF and the mechanical charge sequence is completed, the main mirror and sub mirror are darkened, so T T +- photometry is possible, and #22
Start photometry at step 5. Before transitioning to the film charge sequence, as explained earlier, the AMH is set for a predetermined time t6 to release the stop that makes the film static.
energization is performed (#226).

Thereafter, the process waits for a predetermined time 11o (#227), and then moves to #228. This predetermined time L+o is set to 30 InSec in this embodiment, and is a waiting time for stabilizing the →nof mirror. In #228, it is determined whether the quick shooting mode is selected by the photographer, and if continuous shooting mode 1 is selected, the continuous shooting flag V is set in #229 without indicating that continuous shooting is in progress. After setting L Y F to 1, perform focus detection next time, #105
Proceed to the subsequent focus detection process CDINTA (#230>
. On the other hand, if the continuous shooting mode is not 1~, switch SW6
After waiting at #231 until the switch turns OFF, #105
The process proceeds to the subsequent focus detection process CD INTA.

The automatic focus adjustment operation during quick shooting will be described using flowchart 1 in FIG. From #230 in Figure 9 #
After proceeding to focus detection processing CDINTA after step 105,
As explained in FIG. 7, focus detection calculation and exposure calculation are performed, and the process branches to continuous shooting AF (#301) in #115. First, at #302, exposure calculation and serial communication based on the photometric values started at #225 in FIG. 9 are performed.

This makes it possible to always obtain appropriate exposure even if the subject brightness changes during continuous shooting. Next, at #303, the process waits until the switch SWI is turned off. Zunawara,
Film chassis medium focus detection is limited to one time. Since this is continuous shooting mode 1~, when the 6 switch SWI OFF is detected to give priority to the next release, the film charge sequence ends.
Immediately apply 11 flakes to the sequence motor M1 in #304. #306 Then, it is determined whether the time required to wind up the film 71 is longer than a predetermined time. The film winding time varies greatly depending on the tension of the film, power supply conditions, number of frames, etc. #306 film winding” - (If it is determined that the film winding is slow,
7, the follow-up mote flag TF, the follow-up first flag T1. S
Reset each TF to 1. Follow-up motor flux T
As for F, which will be explained later, it is set in the tracking mode to correct the movement of the subject when the subject is a moving object. When it is determined in step #306 that the film winding is slow, the interval between snapshots is long and an error occurs in the correction for the movement of the subject, so the tracking mode flag TF is reset.

In #308, it is determined whether the focus detection result is low contrast 1 or not. If it is not low contrast 1~last 1~, the process branches to #309 and the previous ignore flag LIF is reset to 1~. This previous ignore flag LIF is set to 1 or more when the next release is performed while ignoring low contrast during continuous shooting. Next, VHO to VHI at #310
is stored as the previous defocus speed (velocity of the subject on the image plane), and then the current defocus speed VHO is determined in #313.

The calculation of the defocus speed VHO in #313 is performed as follows. At the time of #313, the current focus detection result is stored as a defocus DFO. The previous focus detection result is saved in the memory LDF by the process #112 in FIG. Also, the lens position at the center of integration of the CCD that obtained the defocused DFO is:
In the process of #1]0 in FIG. 7, the lens position at the center of integration of the CCD at which the previous defocus L DF was calculated is saved in the memory MIL. As a result, the defocus change δDF caused by the movement of the subject is: δDF=DFO-LDF-(MI-MIL)/KL.
・It is calculated as ■. DFO-LDF is the change in defocus, and (M I - M I L)/K I- is the defocus due to the movement of the lens during that time. On the other hand, the time Δt required during that time can be determined by subtracting the previous integration center time from the current integration center time.

From the process #109 in FIG. 7, ΔL=TM-TML...■ is calculated. From equations (1) and (2), the subject speed VHO is determined as VHO-δDF/Δt.

After determining the subject velocity VHO in #313, the process proceeds to #31. On the other hand, if it is determined in #308 that the low contrast is 1 to last, the process advances to #311. In #311, it is determined whether the previous ignore flag LIF is 1 or 0,
If it has been cleared to 0, the process advances to #312, sets the previous ignore flag LIF, sets 0 to the current detection defocus DFO, and 0 to the number of drive pulses ERRCNT. In #311, if the previous ignore flag LIF was set to 1, the release routine in #314 ○UTR■2
After branching to #315, the tracking mode flag TF and the tracking initial flag Tl5TF, which will be explained later, are reset, and after clearing the continuous shooting flag VLYF in #317, ○UTFS in #121 in FIG. Jump to (#
318). As described above, when low contrast 1 Hellas I is detected twice in a row during focus detection during quick shooting by the processing from #308 to #318, the quick shooting mode is canceled,
The release operation is prohibited, and again according to the flowchart explained in FIG. 7, the release is prohibited until focus is achieved.

As a result, in the case of a moving subject, even if the subject moves out of the focus frame during quick shooting, or if a subject with no contrast enters the picture, it is possible to take the picture at least once, so you do not miss the so-called dramatic moment. The probability is greatly reduced. Furthermore, the probability of significant defocusing is low. In particular, in the snapshot mode, the photographer often wants to give priority to the release, and this is particularly effective. Moreover,
If low contrast is detected twice in a row,
Since continuous shooting is interrupted until the camera is in focus again, the release will not continue with a large amount of defocus, and if the photographer is careless and takes continuous shots of an unintended subject, or if the photographer puts his/her hand in front of the shooting lens. Even in cases where the film is covered with dirt, the film is not wasted. #312 Since the contrast is low this time, defocus DFO1
The number of drive pulses ERRCNT is undefined. Therefore, each of these is set to 0. Furthermore, since the defocus speed VHO cannot be calculated, it is not updated.

That is, the previously detected telefocus speed VHO is used. Subsequently, in #319, it is determined whether the current tracking mode is 1 or more. If the follower mote flag TF is set to 1 and the follower mote is not set, #320 is set.
Proceed to #3] Defocus speed V found in 3
Compare HO and constant V 1, and if VHO>VEI, set constant F Z r (E L 1 to INFZ, which is the focusing range, in #321; if V HO≦VEI, set #321) At step 322, the constants FZREL2 are set to 1.FZR,ELL are set narrower than FZREL2.In other words, the defocus speed V
If HO is small, the subject is stationary, and in this case the defocus change due to subject movement is small, so a wide focusing range is set, and if the defocus speed V HO is large, the defocus change is is large, so set a narrow focusing range. As a result, when the subject is still and low, it is possible to take stable and fast snapshots with less driving of the photographing lens, and when the defocus speed V HO is larger than the predetermined speed VEI, achieved high-precision quick shooting with little tracking delay by using a narrow focusing range.6 Also, these constants VE],
, FZREL1, FZREL2 are CPU20]
(7) It is written in the E2PROM 201f and can be rewritten according to the photographer's preference. In #323, the focus range I NF set in #321 and #322
Z and the currently detected defocus DFO will be compared. If it is smaller than the defocus DFO or the focus range INFZ,
It is determined that the accuracy is sufficiently high, and the process proceeds to #370, where it is determined whether the switch SW6 is ON. Switch SW6 is O
If N, the next release has been requested by the photographer, and the process jumps to the next release (#37).

In other words, since the accuracy is ensured, the next release is performed without performing mirror-up driving. If switch SW6 is OFF, the release process from #372 onwards is 0U.
After branching to TRV and resetting the continuous shooting flag VL)'F and the tracking initial flag Tl5TF to 1 in #373 and #374, in order to perform the next focus detection, focus detection processing CD from #]05 onwards. Jump to I NTA (
#375).

On the other hand, in the judgment of #323, the currently detected defocus DFO
If the value is greater than or equal to the focusing range NFZ, the process advances to #324 to determine whether the photographic lens is driven or in tracking mode. In #324, it is determined whether the subject brightness is bright or dark. Specifically, the determination is made based on the integration time of the CCD and the gain multiplied by the output data, and if it is dark, the process branches to #333. If it is bright, the process proceeds to #325 and calculates the image magnification β of the subject. In #326, it is determined whether the image magnification β is larger than the constants BETA, LOCK. β> B E T A L,
In the case of OCK, the process branches to #333. β≦BE
In the case of TA LOCK, the process proceeds to #327, and it is determined whether the differential focus speed V HO is greater than the constant RVMIN. If VHO≦RVMIN, the process branches to #333, and if VHO>RVMIN, the process branches to #328. In #328, defocus speed V
Compare HO with constant RVMAX.

If V HO≧RVMAX, branch to #333,
V) If I O < RV M A X, #329
Proceed to. In #329, the current and previous differential focus speeds VHO, V obtained in #310 and #313 are
It is determined whether the HI direction is the same direction or the opposite direction, and if the direction is the opposite direction, the process branches to #336, and if the direction is the same, the process proceeds to #330/\. In #330, it is determined whether the follow-up initial flag Tl5TF is set to 1 or not. If it is cleared to 0, the process branches to #335 and is set to 1. If it is already set to 1, the flag Tl5TF is set to 1. For #3
At step 31, the tracking mode flag TF is set to 1, and the process jumps to the tracking processing routine RNAFTI.

Through the processes from #323 to #332 in the second section, a tracking mode for correcting the change in defocus due to the movement of the subject is determined. In other words, if it is determined that the subject is dark in #32/I, the CCD integration will take time and the noise component will be large, so the defocus speed VH will not be accurate.
Since I can't ask for O, I can't go into follow mode. Further, if the image magnification is determined to be large in the determination in #326, the tracking mode cannot be entered in the same way because the influence of camera shake of the photographer is large. #327 Defocus speed■If HO is less than the constant RVMIN, it cannot be determined whether the defocus change is caused by variations in focus detection or by the movement of the subject, and in order to avoid erroneous correction, the camera is set to tracking mode. Can't enter. Even if a defocus change occurs due to movement of the subject, the speed is slow, so the defocus change is small and can be ignored even without correction.

If it is determined in #328 that VHO≧R, VMAX, the defocus change is abnormally large and cannot be considered to be a movement of the subject, but it is determined that the subject has changed, that is, the camera has been shaken, and the following mode is not set. does not enter. If the directions of the previous and current defocus speeds VHI and VHO are reversed in #329, it is thought that the focus detection is unstable or the subject is moving irregularly, and there is a high possibility that incorrect correction will be made. I can't enter it. Furthermore, #330, #331, #3
By performing the process in step 35, the following mode is entered when the determination conditions in steps #323 to #329 are passed twice in succession. This makes it possible to reliably determine whether or not the subject is a moving object, and there is no risk of erroneous correction. Also, the constant BETA
LOCK, RVMIN, and RVMAX are E2PROM201. It is written to f. #32 is the defocus speed V If the direction of HO is reversed, particularly unstable focus detection or subject movement is expected, so cancel processing 0UTR after #336
Proceed to V3, reset the tracking mode flag TF, initial tracking flag Tl5TF, and continuous shooting flag VLYF in #337, and jump to focus detection processing CDINTA from #105 onwards in order to perform the next focus detection (#338 ).

As a result, the next release is prohibited, and as explained in FIG. 7, the lens is driven until the lens is brought into focus again, so there is no need to worry about out-of-focus photography being performed.

If it is determined in #324, #326, #327, #328, and #330 that the process branches to #333, the currently detected defocus DPO is compared with the constant INFZEl. When DFO<fNFZEl, the defocus is not very large and the reliability of focus detection is high, and sufficient accuracy is ensured even if the taking lens is driven by this amount of defocus while the mirror is up and the next release is performed. Therefore, the process branches to the drive routine RNMTR during mirror up. D
In the case of FO≧INFZE1, the release process 0UTRV2 is performed (#334), since there is a possibility that accuracy cannot be ensured if the release is performed next time because the teff] - dregs are large. By performing the processing in #333 and #334, when the defocus is small, high-precision automatic focusing and high-speed continuous shooting are achieved by driving the mirror up, and when the defocus is large, it is possible to perform a transformation inspection. Highly accurate automatic focusing is achieved because the lens is brought out and focused.

In addition, if the defocusing process 0UTFS is entered after the cancellation process 0UTRV2 from #334, as explained in FIG. The detection is fast and accurate. Also, the constant INFZE1 is the E2PR of the CPU 201.
, 0M201f, and can be changed according to the user's preference.

Now, the follow-up process RNAFTI executed by branching based on the determination in #319 or jumping from #332 will be explained with reference to FIG. First, in #340,
It is determined whether the directions of the current and previous defocusing speeds VHO and VHI are the same. If the directions are different, it is possible that the subject suddenly stopped, changed direction, or shook the camera. In this case, #3
41, jumps to cancellation processing 0UTRV3, cancels the tracking mode, and performs automatic focusing operation until focus is achieved again. As a result, even if the subject suddenly stops, changes direction, or shakes the camera, highly accurate focusing can be achieved without erroneous correction.

If they are in the same direction, proceed to #342 and (VH
Compare O+VH1)/2 with the constant AVESH. (VH
O+VH1)/2 is +1 1. from the process of #310. It shows Vl-1 and is a weighted average value. In Equation 1, 11 is the number of loops, and V Hi does not indicate the previous speed.

That is, in #342, the weighted average value and the constant AVES H
Compare with. The weighted average value is a constant A VE S I-
(In the following cases, set the defocus speed ■HO in #343.
Reset the weighted average value to the constant A V E S I-1
If it is larger, the process directly proceeds to #344. In other words, in the case of slow speeds, stable correction is achieved by absorbing variations in focus detection by performing weighted averaging, and in the case of a subject approaching at a constant speed, the change in defocus is approximately the same. Since it increases in inverse proportion to the square of the distance, this achieves correction with good responsiveness and little follow-up delay at high speeds. In addition, the constant AVESH is CPU20
It is written in E2PROM 201f built in 1.

#344 Calculate the image magnification β and use the constant B E TAL
Compare with OCK 2. As the image magnification increases, the influence of camera shake increases as described above, so in #347, the canceling process OUT RV 2 is performed, and the tracking mode is also exited. Note that the constant B ET AI-OCN3 is cPU201. (7) It is written in E2PR, 0M20 If, and is set larger than the constant B E T A L OCK. In #345, compare the defocus speed V HO and the constant R ○ UT, and if the defocus speed is within V I-10 or R, VOUT, the defocus speed is sufficiently slow and there is no error due to variations in focus detection. Branch to #347 so that no correction is required. #346
Then, the defocus speed VHO and the constant RVMA
Compare with X 2. If the defocus speed V I-10 is R, V M A Branch out.

At #347, the process jumps to the cancellation process 0UTRV2, cancels the tracking mode, prohibits the next release, and performs focus detection again. As a result, in the case of a very high-speed subject, release is prohibited, and it is possible to prevent a photograph of a subject being delayed. The processes of #344 to #346 prevent erroneous correction and achieve highly accurate correction.

Subsequently, follow-up correction calculation 1 is performed in #348, and the number of drive pulses E RRCN T is calculated. This calculation will be explained in detail later using FIG. 12. In #349, the defocus MDF after tracking correction is compared with the constant INFZE2. If the defocus after tracking correction is large, the process branches to #350 and jumps to #316 and later (FIG. 10) cancellation processing UTRV2] in order to improve accuracy. As a result, the continuous shooting flag V L Y l? clear only, keep tracking mode 1, defocus processing 0UT
Jump to FS. Constant I NFZE2 is CPU20
1 E2PROM 201f. Also,
This constant NFZE2 is the constant I explained in #333.
NFZE]. This is because (,, l tracking correction is performed, so if the correction amount is not large, it is not possible to proceed to #351.

The steps after #351 are mirror-up drive processing, and #33
This is the branch from #3 or the flow from #349. First, in #352, it is determined whether the switch SW6 is ON. If switch SW6 is ○FF, the next release is not requested, so the process branches to #363 and the release process ○
Jump to UTRV. Subsequently, in #353, the number of driving pulses ERRCNT and the constant NPI are compared. As explained in FIG. 7, the constant NP is the number of pulses that can be driven during mirror up. If the number of drive pulses ERRCNT is less than or equal to the constant NP1, it is possible to drive during mirror up, and the process branches to #359.

If the number of drive pulses ERRCNT exceeds the constant NPI, the mirror cannot be driven for long while the mirror is up, so a drive time is required before the next release starts. Also, this A
The time required to drive the F Momo-M2 varies depending on the power supply conditions, characteristics of the interchangeable lens, etc. Therefore, if the time until the next release start is fixed at 40+n5ec, and the number of driving pulses ERR, CNT is within 40m5ec and the number of pulses that can be driven during the next mirror up (constant NP2),
Drive AF momo-M2 and set the number of drive pulses ERRCN
If T exceeds the constant NP2, refocus detection is performed. As a result, tracking correction can be performed accurately for 4Qmsec even in the moving object mode. Furthermore, even if the camera waits for 4 Q m5ec, the speed of snapshots will only drop from 3 frames per second to 27 frames per second, and the deterioration of the feel of snapshots will be minimal. Furthermore, when the number of driving pulses ERRCNT exceeds the constant NP2, refocusing is detected, which solves the problem that the error associated with lens driving increases indefinitely.

If ERRCNT≦NPI in #353, the process advances to #354 and the tracking flag TF is determined. If #354 is tracking mode (TF=1), tracking correction calculation 2 for 40m5ec is performed in #355, if TF=0, #355 is skipped, and in both cases, drive pulse number ERRCN is set in #356.
Compare T with constant NP2. Drive pulse number ERRCNT
If exceeds the constant NP2, the process branches to #364 and jumps to release processing 0UTRV2. #349, #35
If the next release is to be performed by driving during mirror up without detecting the focus again in the determination in step 6, it is determined that the camera is in focus, and the focus display is maintained. Next, start driving AF Momo-M2 at #357, and set 40 +n5ec at #358.
Wait for an amount of time. In #359, the drive of AF Momo-M2 is started in case it branches from #353, and in #36
The process advances to the quick-shot release process RNRELESE after 0.

Since #361 is a quick shot, in order to completely stop the film, it waits for a predetermined time tll and then jumps to the next release.As is clear from the explanation above, in tracking mode, the defocus changes depending on the subject. Since the amount must be corrected, the shooting lens remains stopped and the next release is not performed.

Next, the follow-up correction calculations after #401 will be explained with reference to FIG. First, in #402, it is determined whether or not the current focus detection was performed with low contrast. If low contrast I/last is not achieved, the time T for correction is determined in #403.

The time of the current CCD integration center is saved in the memory TM, and if you subtract the value in the memory TM from the current timer value TC and add the mirror-up time of 7Q m5ec, from the current integration center to the next exposure. Find the time. In the case of low contrast 1-, proceed to #404,
The time T from the previous exposure time to the next exposure time is determined. As explained in FIG. 8, the previous exposure time is saved in the memory TIME1. Therefore, it is sufficient to subtract the value of the memory TIME 1 from the current timer value TC and add 70 m5ec. That is, #40
In steps 2 to #404, when the contrast is not low, the calculation is performed based on the current defocus DFO, and when the contrast is low, the calculation is performed assuming that the defocus was 0 during the previous exposure.

Next, in #405, the defocus speed VHO is set to the above #
The correction amount Δ is obtained by subtracting the time T obtained in 403 or #404.
I'm looking for a DF. Next, in #406, the defocus speed VHO and the constant VVH are compared. If VHO>VVH and the defocus speed is fast, it is determined in #407 whether the subject is approaching or going away, and if the subject is approaching, the correction amount ΔDF is multiplied by 1.25 (#408)
. As mentioned above, if the subject approaches in the optical axis direction at a constant speed, the defocusing speed will increase in inverse proportion to the square of the subject distance.
Alternatively, the defocus speed also increases during the time T determined in step #404. In order to correct this error, the correction amount Δ
DF is multiplied by 1.25. When the subject is far away, the defocus speed becomes slower, so the correction amount ΔD
Multiply F by 0.75 (#409>. Next, in #410, set the currently detected defocus DFO and defocus speed V H
Check the direction of O, and if the direction is the same, the corrected defocus MDF will be DFO + ΔDF (#41.1
). If the directions are different, the currently detected defocus DFO and the correction amount ΔDF are compared in #412, and DFO≦ΔD.
If F, ΔDF-DFO is applied to the corrected defocus MDF.
(#413). If DFO>ΔDF, the photographing lens must be driven in the direction opposite to the defocus speed direction, and a large defocus is detected in the direction opposite to the defocus speed direction. For this reason,
Assuming no error or abnormal movement of the subject due to reversal of shooting sense, perform stack initialization in #414 and release process 0UT in #415.
By proceeding to RV21, the next release is prohibited and refocusing detection is performed. Correct Tef with #・111L#413] -
When the waste MDF is determined, in #416, the coefficient 1 to convert the defocus into the number of pulses is set. Multiply (discard), set drive pulse number E R, R, CNT, return (#/
1.1.7).

As a result, highly accurate correction is possible even when the subject is at high speed, and it can also be used for both close (q<subject) and distant subjects.Furthermore, as is clear from the explanation of Fig. 10, #308 Even when the low contrast 1 helast is ignored once in ~#312, the moving bones of the subject are corrected correctly.

Finally, timer interrupts and A and FP interrupts will be explained.

FIG. 13 shows a timer interrupt processing routine for driving AF Momo-M2. The CPU 20 has a built-in interrupt timer (not shown) that generates a timer interrupt when a set time has elapsed. When a timer interrupt occurs, #
At 502, the interrupt timer IT is reset. As a result, the interrupt timer IT will
If the set time elapses, a timer interrupt is automatically generated.

Next, in #503, the flag RM G ON F is determined, and if it is set, the release magnet 1~
Since power is being applied to R and Mg, the AF motor M is turned off as described above (#505). Flag RMG O
NF If the number is set to 1~, #504 is set to A.
Energize the F motor M7 and return (#506).

Figure 14 shows A, which occurs when the F P signal falls.
, FP interrupt processing routine. A, FP When a P interrupt occurs, first, in #602, the number of drive pulses ERR is
, CN Decrease T by one. In #603, the driving pulse number ERRCNT becomes O, and it is determined whether or not the driving of AF Momo-M2 has ended. The driving pulse number ERRCNT
If it is not 0 and driving of AF Momo-M2 is not completed, proceed to #604 and reset the interrupt timer IT.
to do. In #605, the flag RM_ONF is checked. If the flag RM G ON F is set and the release Macnets R and M g are energized, the AF momo-M2 is turned OFF in #608. If flag RMGONF has been reset from 1 to #6
At 06, apply the brakes to AF Momo-M2 and return (#607). On the other hand, if it is determined in #603 that the driving of the A and F motors M2 has been completed, the process branches to #609, and in #609, the energization of the A and F motors M2 is turned off. Subsequently, in #610 and #611, timer interrupts and ARP interrupts are inhibited, respectively, and the process returns (#607). As described above, the AF momo-M2 is driven by the timer interrupt and the ARP interrupt, and the AF momo-M2 for release is driven by the timer interrupt and the ARP interrupt.
While the motor is energized, the A and F motors M2 are controlled to be OFF.

[Effects of the Invention] In the automatic focusing camera of the present invention, in a reflex camera in which the mirror assembly time is approximately constant, the rate of change in the amount of defocus due to the movement of the subject is detected, and the mirror is adjusted to this rate of change. The amount of correction for the amount of defocus caused by a dynamic subject is determined by multiplying the amount of correction for the amount of defocus caused by a dynamic subject by multiplying the predetermined time including the up time and the direction of the speed of change by a predetermined constant. This has the effect that it is possible to correct the problem with high precision.

Note that if the correction amount determining means is operated only when it is determined that the subject is a dynamic subject based on the rate of change detected by the rate of change detection means, unnecessary following may occur in line with a stationary subject. This will prevent corrections from being made.

Further, the predetermined constant is set to be greater than 1 when the rate of change in the amount of defocus due to the movement of the subject is in a direction in which the subject approaches the photographer, and when the rate of change is in a direction in which the subject moves away from the photographer. A setting deviation smaller than 1 can prevent insufficient tracking or excessive tracking.

[Brief explanation of the drawing]

Fig. 1 is a block diagram showing the basic configuration of the present invention, Fig. 2 is a side view of a camera as an embodiment of the invention, Fig. 3 is a front view of the same, and Fig. 4 is a flock circuit of the same. 5 and 6 are operational waveform diagrams, and FIGS. 7 to 14 are flowcharts showing the same operation. (1) is focus detection means, (2) is change rate detection means, (
3) is a correction amount determining means, (4) is a lens driving means, and (5) is a correction amount determining means.
) is the determination means.

Claims (3)

[Claims] (1) In a reflex camera in which the mirror up time is approximately constant, there is a focus detection means that detects the amount of defocus of the photographing lens with respect to the subject to be focused on, and a change that detects the speed of change in the amount of defocus due to movement of the subject. A speed detection means, and a correction amount for an amount of defocus due to a dynamic subject is determined by multiplying the speed of change detected by the speed of change detection means by a predetermined time including a mirror-up time and a predetermined constant according to the direction of the speed of change. Comprising a correction amount determining means, and a lens driving means for driving a focusing lens toward a focus position based on the amount of defocus detected by the focus detecting means and the correction amount determined by the correction amount determining means. An automatic focusing camera characterized by: (2) Further comprising determining means for determining whether or not the subject is a dynamic subject based on the rate of change detected by the rate of change detection means, and the correction amount determining means determines whether the subject is a dynamic subject or not. Claim 1 characterized in that the means is activated only when it is determined that
Autofocus camera as described. (3) The predetermined constant is set to be greater than 1 when the rate of change in the amount of defocus due to the movement of the subject is in the direction in which the subject approaches the photographer, and is set to be greater than 1 when the rate of change is in the direction in which the subject moves away from the photographer. The autofocus camera according to claim 1 or 2, wherein the autofocus camera is set smaller.