JPH04338707A - Focus detector - Google Patents

Abstract

PURPOSE:To reduce measuring errors by temporarily storing a defocusing quantity in the case that an object is low-contrasted, comparing it with a defocusing quantity measured again and selecting the smaller one. CONSTITUTION:A main CPU 35 calculates the defocusing quantity by specified arithmetic operation based on integration data outputted from a CCD sensor unit 21 for range-finding. Whether the specified object is low-contrasted is judged, and in the case of judging that it is not low-contrasted, the defocusing quantity in that case is selected. In the case of judging that it is low-contrasted, the defocusing quantity in that case is temporarily stored and the CCD sensor unit 21 is actuated again. The stored defocusing quantity and the defocusing quantity measured again this time are compared, and the smaller quantity is selected as the defocusing quantity.

Description

[Detailed description of the invention]

0001

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a focus detection device for optical equipment, and more specifically to a focus detection device capable of detecting focus with higher accuracy even when an object has a low contrast.

0002

[Prior art and its problems] Focus indication that photoelectrically determines the distance to the subject or the amount of defocus by the photographic lens and indicates whether it is in focus or out of focus, or auto that automatically adjusts the focus of the photographic lens. Optical devices such as cameras that have a focus function are known. In the focus detection device used in such an optical device, distance measurement processing may be continuously performed in order to reliably determine whether the object is in focus or out of focus.

However, especially in the case of a phase difference type focus detection device, when the object has a low contrast, a distance measurement error may occur even when measuring the same object under the same conditions. Therefore, even if the focus is once achieved, due to distance measurement errors, etc., it may be determined that the lens is out of focus even though it is in focus, and the lens may be driven, or the focus display may flicker (blink). There is.

0004

OBJECTS OF THE INVENTION It is an object of the present invention to prevent inconveniences caused by measurement errors in a focus detection device, especially when the contrast is low.

[0005]

SUMMARY OF THE INVENTION In order to achieve this object, the present invention includes: a defocus amount measuring means for measuring a defocus amount for a specific subject; a contrast determining means for determining whether the specific subject has a low contrast; When the contrast determining means determines that the contrast is not low, the defocus amount is selected, and when the contrast determining means determines that the contrast is low, the defocus amount at that time is temporarily stored and the defocus measuring means is re-operated. and a defocus amount selection means for comparing the memorized defocus amount and the currently re-measured defocus amount and selecting the smaller one as the defocus amount.

According to this configuration, when the object to be measured has a low contrast, the smaller defocus amount of the two previous defocus amounts is selected, so that malfunctions due to measurement errors are reduced.

[0007]

DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be explained below based on the illustrated embodiments. FIG. 1 is a block diagram showing the main configuration of an autofocus (AF) single-lens reflex camera to which the present invention is applied. This AF
The single-lens reflex camera includes a camera body 11 and a photographic lens 51 that is detachable from the camera body 11. Most of the subject light flux that enters the camera body 11 from the photographic lens 51 is reflected by the main mirror 13 toward the pentaprism 15 that constitutes the finder optical system, and a part of the reflected light is further reflected by the light receiving element of the photometry IC 17. (
(not shown). On the other hand, a part of the subject light beam incident on the half mirror section 14 of the main mirror 13 is transmitted through this, is reflected downward by the sub mirror 19, and is sent to the distance measuring CCD.
The light enters the sensor unit 21.

The photometric IC 17 is equipped with a light receiving element that receives the subject light flux, and this light receiving element logarithmically compresses the electric signal generated according to the amount of light received, and the peripheral area control circuit 23
The main CPU 35 performs A/D exchange via the main CPU 35. The main CPU 35 executes a predetermined exposure calculation based on the photometry signal and film sensitivity information, and calculates an appropriate shutter speed and aperture value for exposure. Then, based on these shutter speeds and aperture values, the exposure mechanism (shutter mechanism) 25 and the aperture mechanism 27 are driven to perform exposure. Further, the peripheral control circuit 23 drives the mirror motor 31 via the motor drive circuit 29 to perform up/down processing of the main mirror 13 at the time of release, and after exposure is completed, drives the winding motor 33 to film the film. wind up.

The distance measuring CCD sensor unit 21 is a so-called phase difference type distance measuring sensor, and although not shown,
It includes a splitting optical system that divides the object light beam into two, and a CCD line sensor that receives and integrates (photoelectrically converts and accumulates the electric charge) each of the two divided object light beams. The ranging CCD sensor unit 21 then outputs the integral data integrated by the CCD line sensor to the main CPU 35 as a control means. The distance measuring CCD sensor unit 21 is driven and controlled by the main CPU 35 and the peripheral control circuit 23. Note that the CCD sensor unit 21 is equipped with a monitor element, and the peripheral control circuit 2
3 detects the subject brightness through this monitor element and changes the integration time according to the detection result.

The main CPU 35 calculates the amount of defocus by a predetermined calculation (predictor calculation) based on the integral data output from the range-finding CCD sensor unit 21, and controls the AF motor 39 based on the defocus amount.
rotation direction and rotation speed (number of pulses of encoder 41)
Calculate. Then, the main CPU 35 controls the AF motor drive circuit 37 based on the rotation direction and the number of pulses.
The AF motor 39 is driven via the AF motor 39. Furthermore, the main CP
U35 controls the encoder 4 according to the rotation of the AF motor 39.
1 is detected and counted, and when the count value reaches the above-mentioned number of pulses, the AF motor 39 is stopped. During this drive, the main CPU 35 normally performs DC drive, and can control the AF motor 39 at a constant speed based on the interval of output pulses of the encoder 41 before stopping. Note that the rotation of the AF motor 39 is controlled by the rotation of the camera body 11.
The signal is transmitted to the photographing lens 51 side through a connection between a joint 47 provided on the mount portion of the photographing lens 51 and a joint 57 provided on the mount portion of the photographing lens 51 .

The main CPU 35 also includes a ROM 35a storing programs, a RAM 35b, and a reference timer 3.
5c and a counter 35d, and is connected to an E2PROM 43 as a memory means. This E2PRO
In addition to various constants specific to the camera body 11, the M43 stores various functions, constants, etc. necessary for the AF operation of the present invention.

Furthermore, the main CPU 35 includes a photometry switch SW that is turned on when the release button (not shown) is pressed halfway.
A release switch SWR, which is turned on by pressing S and all the way down, an automatic focus switch SWAF, and a main switch SWM, which turns on/off power to the main CPU 35, peripheral devices, etc., are connected. The main CPU 35 controls the set AF
, exposure, photographing mode, shutter speed, aperture value, etc. are displayed on the display device 45. The display device 45 is usually provided at two locations: on the outer surface of the camera body 11 and within the field of view of the finder.

The main CPU 35 functions as a control means for comprehensively controlling the entire camera system, and also functions as a focus judgment means and a moving object judgment means, and controls the CCD sensor unit 21, peripheral area control circuit 23, etc. The AF motor 39 and the like constitute a distance measuring means, and the AF motor 39 and the like constitute a lens driving means.

On the other hand, the photographing lens 51 includes a focus adjusting mechanism 55 that drives the focusing lens group 53 in the optical axis direction, and is provided at the mount of the photographing lens 51 and connected to the joint 47 of the camera body 11 to drive the AF motor. 39 to the focus adjustment mechanism 55, and a lens CPU 61 that calculates various data of the photographic lens 51. The lens CPU 61 has electrical contacts 59, 4
9, and executes predetermined data communication with the main CPU 35 via the peripheral control circuit 23. Lens CP
The data transmitted from U61 to the peripheral control circuit 23 includes the open aperture value Av, the maximum aperture value Av, the focal length,
There is K value information etc. Furthermore, what is K value information?
The image plane formed by the photographic lens is set at a unit distance (for example, 1
mm) This is data on the number of pulses of the encoder 41 (number of rotations of the AF motor 39) required for movement.

Next, the AF operation of this embodiment will be further explained with reference to FIGS. 2 to 9. This single-lens reflex camera starts AF processing when the photometry switch SWS is turned on. In the AF process, first, the distance measuring CCD sensor unit 21 starts integration. After the integration is completed, the main CPU 35 inputs the integral data, calculates the defocus amount and the number of drive pulses based on the data, and drives the AF motor 39 based on the number of drive pulses. Furthermore, this embodiment has an AF single mode in which the AF can be locked once the subject is in focus when the subject does not move, and a focus priority mode in which the camera cannot release the camera unless the subject is in focus.

The operation up to entering the moving object tracking AF process will be explained with reference to FIGS. 2 and 3. 2 and 3 are graphs showing the relationship between the subject image plane position and the film equivalent plane position (focus position) with reference to the position of the focusing lens 53.

When the photometry switch SWS is turned on and AF processing begins, the lens is driven (the AF motor 39 is driven) based on the drive amount (the number of drive pulses) determined by the first integration and calculation. In this embodiment, if the first integration and calculation after the photometry switch SWS is turned on is that the defocus amount (or the number of drive pulses) is larger than a certain amount, the integration and calculation are repeated even during the lens driving. When the defocus amount becomes smaller than a certain value during the repetition, subsequent integration and calculation are stopped, and the lens is driven based on the number of drive pulses determined by the immediately preceding (latest) integration and calculation. Note that the integration and arithmetic processing will also be referred to as distance measurement processing hereinafter.

When the first lens driving is completed, integration and calculation are performed again to check whether or not the lens is in focus. If the object is in focus, it is considered that the object is not moving, but since there is a possibility that the object is moving, the distance measurement process is performed again after a predetermined period of time. If an interrupt occurs due to turning on of the release switch SWR while waiting for this predetermined time, the release process is performed, and if no interruption occurs, the above-mentioned distance measurement process and lens driving process are repeated while the photometry switch SWS is turned on.

Furthermore, in the continuous processing of distance measurement and lens driving, if the object is not in focus several times in a row (three times in this example), it is assumed that the object is moving (moving object).
The camera then enters the moving object prediction AF mode. Note that while the photometry switch SWS is on, distance measurement and lens drive processing continue, and once the camera is in focus, it enters moving object predictive AF processing when out-of-focus conditions occur three times in a row. You can also do this.

"Moving Object Prediction AF Mode" Next, the moving object prediction AF mode of this embodiment will be outlined. Figures 4 and 5
is a graph showing an example of operation in moving object prediction AF mode processing. When the number of defocus pulses DP is obtained by calculation C based on the result of integral I when it is determined that the moving object is the subject, this number of defocus pulses DP
Lens drive M is performed based on.

When the lens driving M is completed, the integral I1 is again
The moving speed of the subject image (subject tracking speed) S is calculated in calculation C1 from the number of defocus pulses DP1 at that time and the time T1 from the midpoint of the previous integral I to the midpoint of the current integral I1. In order to match the subject image plane with the film plane in a short time, the subject tracking speed S is set at a predetermined times (twice in this embodiment) a time T1.
During this period, constant speed control (double speed lens drive) M1 is performed. After this double-speed lens driving M1 is completed, constant speed control (lens driving) M2 is performed at the object tracking speed S1 while integration I2 is being performed. Note that constant speed control refers to control in which the AF motor 39 is driven at a constant speed.

When the integration I2 is completed, the lens drive M2 is stopped and the calculation C2 is performed. In this calculation C2, the subject tracking speed S2 is calculated based on the current number of defocus pulses DP2, the calculated number of moving pulses of the subject MP1, and the time T2 between the integrals I1 and I2. Then, for the time t2 required for calculation C2, double-speed lens drive M3 is performed at twice the subject tracking speed S2. Then, when this double-speed lens drive M3 is completed, integration I3 is started, and this integration I
3, the tracking lens drive M4 is performed at the subject tracking speed S2.

Furthermore, the focusing lens 53 may move too far due to double-speed lens driving. For example, in FIG. 4, the double-speed lens drive M3 causes the defocus pulse number DP3 to go too far. This advance amount (defocus pulse number D
If P3) is less than a predetermined value, there is no problem in assuming that the subject continues to move, so after the calculation C3 is completed, double-speed lens drive M5 is performed at the moving object tracking speed S3.

[0024] However, if it goes too far beyond a predetermined amount,
Since there are cases where the moving speed of the object becomes slow or the object stops, a check process is performed to observe the situation without driving the lens, as shown in FIG. 5, for example.

In FIG. 5, when the number of overshooting defocus pulses DP4 equal to or greater than a predetermined amount is obtained by integration I4, double-speed lens driving is not performed and the process waits for the calculation time t4 required for calculation C4. Then, at the same time as the start of the next integration I5, the tracking lens drive M7 is performed at the subject tracking speed S3 obtained in the previous calculation C4. When the number of defocus pulses DP5 within a predetermined amount is obtained from the data of the integral I5, follow-up AF processing is performed to perform normal double-speed lens driving.

However, if the subject image does not fall within the in-focus range even after repeating this check twice to avoid double-speed lens drive, the subject may have slowed down, the subject has stopped, or the subject has moved. Since it is considered that the direction has changed, the moving object prediction AF mode is exited, the mode returns to the normal AF mode, and the processing shown in FIG. 2 is performed. 2
By checking twice, it is possible to reliably detect that the state of the moving object has changed. Note that in this embodiment, the moving object prediction mode is exited when the check becomes too far even after repeating the check twice, but the present invention is not limited to this number of times. Further, in the check process, a configuration may also be adopted in which the tracking lens is not driven.

"Calculation 1 of moving object following speed (subject image moving speed)" A specific method of calculating the moving object following speed S will be explained with reference to FIGS. 4 and 5. The initial moving object tracking speed S1 is determined by the number of defocus pulses DP1 obtained by the first integration I1 after entering the moving object prediction mode, and the previous integration I1.
The time T1 from the midpoint of the current integration I1 to the midpoint of the current integral I1 is determined by the following formula. Note that X1 is the output cycle (ms) of the pulses output by the encoder 41. X1=T1/DP1 (ms)
…■S1:1/X1 =DP1/T1
(puls/ms) …■

[0028] Moving object tracking speed S obtained from the above formula
By first driving the lens at double speed M1 for a time corresponding to the integration interval T1, the focusing lens 53
quickly move the camera to near focus. Thereafter, while the integration I2 is being performed, a tracking lens drive M2 is performed at a moving object tracking speed S1 to follow the movement of the object.

[Calculation 2 of moving object following speed Sn] During the moving object following operation, the moving object following speed Sn is calculated using the following formula. First, the lens drive corresponds to the amount of movement when it is assumed that the subject image was moving at a speed equivalent to the moving object tracking speed S1 within the time T2 from the midpoint of the previous integral I1 to the midpoint of the current integral I2. The number of pulses MP1 is determined by the following formula. MP1=T2×S1
…■

Next, the time T2 from the midpoint of the previous integral I1 to the midpoint of the current integral I2, and the time T2
From the number of pulses MP1 corresponding to the amount of movement within and the current number of defocus pulses DP2, find the period X2 of the AF pulse output that provides the moving object tracking speed S2, and calculate the moving object tracking speed S2 from the period X2 of the AF pulse output. demand. X2=T2/(MP1+DP2) (ms)…■
S2=1/X2=(MP1+DP2)/T2
(Pulse/ms)…■

[0031] With the moving object following speed S2 obtained from the above formula, during the time t2 required for calculation C2, 2 of the moving object following speed S2
Lens drive M2 is performed at double speed, and after completion of this double speed drive, while integration I3 is being performed, follow-up lens drive M3 is performed at moving object tracking speed S2 to follow the movement of the subject.

Note that in this calculation, the number of defocus pulses DP is a scalar quantity, so the sign changes depending on whether it is front focus or rear focus. Therefore, if the defocusing pulse has gone too far, the current number of defocus pulses DP2 is subtracted from the number of pulses MP1 corresponding to the amount of movement within time T2.

The calculation of tracking can be expressed as a general formula as follows. MPn-1=Tn×Sn-1
...■Xn=Tn/(MPn-1±DP
n) (ms)...■Sn=1/Xn
(Pulse/ms)...■ or more ■,
The moving object tracking shown in the figure becomes possible by repeating calculations using formulas (1) and (2), lens driving and integration based on the calculation results. In addition, in the calculation of this embodiment, the number of moving pulses MPn-1 and the number of defocus pulses DPn of the calculated subject image are taken as absolute values, so the number of defocus pulses ±DPn on the right side of equation (2) is It is a plus (+), and if it goes too far, it is a minus (-).

In the above lens driving, after the calculation is completed, the lens is driven at twice the moving object tracking speed S1 for the time required for the calculation, but this is from the time when lens driving is stopped until the next lens driving is completed. This is to move the image plane by an amount corresponding to the amount by which the subject image moves during that time. In other words,
This is to satisfy the relationship Sn×(lens stop time+lens drive time)=lens drive speed×lens drive time with a simpler calculation. Therefore, as long as the above formula is satisfied, the lens may be driven at other moving object tracking speeds.

``Processing when the release switch is turned on'' The operation when the release switch is turned on during the moving object predictive AF process will be described with reference to FIG.

Generally, in a single-lens reflex camera, the mirror is raised after the release switch is turned on, so it takes a certain amount of time from when the release switch is turned on until the film is actually exposed. This is called release time lag RTL. Therefore, when the subject is moving, the subject moves from the time the release switch SWR is turned on until the start of exposure (release time lag RTL
), it is desirable to continue driving the tracking lens.

Therefore, in this embodiment, it is checked whether the release switch SWR is turned on at the end of the calculation, and
When it is turned on, the lens drive M3 is performed at two speeds of the moving object tracking speed S3 for a time corresponding to the calculation time t3+release time lag RTL, and the focusing lens 53 is advanced by the release time lag RTL.

When the double speed tracking lens drive M3 is completed,
A photometric calculation and a mirror up process are performed, the aperture is narrowed down to the value obtained by the calculation, and the shutter mechanism 25 is driven at the shutter speed obtained by the calculation.

The above is the processing when the subject is approaching, but if you are too proactive when the subject is moving away, you will end up with so-called rear focus. The depth of field is shallower on the near side than the focal point, so it is better to focus on the front than on the back. Moreover, when the subject moves away, the image plane movement speed becomes slower for a while when the object movement speed is constant.

Therefore, in this embodiment, when the subject is moving away, the double-speed lens is driven for half the time as when the subject is approaching. In FIG. 7, operation C4
The lens is driven at a speed twice the subject tracking speed S3 for a period of time that is half of the calculation time t4 + the time corresponding to the release time lag RTL.

"Main Processing" Next, the operation of this embodiment will be explained in more detail with reference to the flowcharts shown in FIGS. 8 to 15. These processes are performed by the main CP
It is executed by the main CPU 35 based on a program stored in the internal ROM 35a of the U 35. Further, parameters necessary for the above calculation are stored in the E2PROM 47.

FIG. 8 shows a routine related to main processing by the main CPU 35. When the main switch SWM is turned on, this process begins, and step (hereinafter abbreviated as "S") 1
01, the system including each port and memory is initialized. Next, perform power down processing to eliminate unnecessary power consumption, check whether the photometry switch SWS is on, and repeat the power down processing and check processing until the photometry switch SWS is turned on (
S103, S105).

When the photometry switch SWS is turned on, the reference timer 35c is started, the switch status of the AF switch SWAF, etc. is checked, and lens communication is performed with the lens CPU 61 to determine the open aperture value, maximum aperture value, and focal length. and input K value data (S107,
S109, S111). Furthermore, what is K value data?
This is the number of pulses of the encoder 41 (amount of rotation of the AF motor 39) necessary to move the image plane of the subject imaged by the photographing lens 51 by a unit length.

Then, photometry data is input from the photometry IC 17, and a shutter speed and aperture value are calculated by a predetermined algorithm based on this data, film sensitivity, etc., and the calculated shutter speed and aperture value and other photographic data are calculated. is displayed on the display device 45 (S113, S115
).

Next, AF processing is performed, it is checked whether a predetermined loop time has elapsed, and the AF processing is repeated until the loop time has elapsed (S117, S119). When the loop time has elapsed, the process returns to S109, and S109-S
119 loop processing is repeated.

Further, while repeating this loop processing, the reference timer interrupt processing shown in FIG. 9 is performed every time the reference timer 35c times out. In this reference timer interrupt process, various processes are executed. In this interrupt processing, after counting the loop time, the photometry switch S
Input the status of WS and release switch SWR (
S121, S123). When the release switch SWR is off, or when the release switch SWR is on but release permission has not been issued, for example when the focus is not in focus, the process returns to the step before the interrupt (S12
5, S127).

When the release switch SWR is turned on and release permission is issued, release processing begins. In the release process, the mirror motor 31 is first activated to perform mirror-up processing, and the aperture mechanism 27 is activated to narrow down the aperture to the value calculated in S113 (S131
, S133).

Wait until the mirror up is completed, and when the mirror up is completed, the shutter mechanism 25 is driven at the shutter speed calculated in S113 to perform exposure (S135,
137). When the exposure is completed, the mirror motor 31 is started to perform mirror down processing and the film winding motor 33 is started.
is activated to wind the film by one frame and return to S107 (S139).

"AF Processing" Next, the AF processing of this embodiment will be explained with reference to FIGS. 10 to 13. First, check whether the focus mode is AF mode or MF (manual focus) mode, and if it is MF mode, select S2.
11 (S201). Note that the AF mode is a mode in which automatic focus adjustment is performed, and the MF mode is a mode in which the photographer performs focus adjustment.

When AF processing is first entered in AF mode, integration processing is not being performed, so the F-lock state is not present, the focus is out of focus (IN FOCUS flag is "0"), and the re-integration flag is also cleared. Therefore, the process advances to S211 (S205, S209). In the second and subsequent processing,
If the AF is locked, jump to AF lock processing and
If it is not in the AF lock state, if it is in focus, that is, after waiting a predetermined time after focusing, the AF lock flag is set and the process proceeds to reintegration processing (S203, S205, S207
). From the second time onwards, since the reintegration flag is set, the process proceeds to reintegration processing (S209).

When the AF process is performed for the first time or in manual focus mode, the process advances to S211 (S20
1, S203, S205, S209). In S211,
In step S213, it is checked whether AF is in operation, and in S213 it is checked whether MF is in operation. AF in operation (AF1
If it is the second time), proceed to S231, and if the AF is not in operation,
If the MF is in operation, distance measurement processing and focus/non-focus display processing are performed (S215 to S227).

In S215, integration and arithmetic processing are performed, and then it is checked whether the result of the arithmetic operation is valid (S217). If the calculation result is valid, in-focus display processing is performed, and the focus display LED of the display device 45 is activated.
(not shown) and further emit a sound from the electronic buzzer 46 to notify that the camera is in focus. However, if the camera is not in focus, the upper focus display process is not performed (S219).
). Furthermore, if the subject is low contrast and the process is out of focus for the first time, the process jumps to S229, and if not, it is checked whether the subject is in focus (S221, S223)
. Here, if in focus, set the AF release permission flag and proceed to S229 (S223, S225); if not in focus, set the secondary focus width, focus, release permission, and lock AF. Clear flag and display device 4
5, turn off the focus display and turn off the electronic buzzer 46.
The process advances to 229 (S223, S227).

Even if the calculation result is invalid, the secondary focus width is cleared, the focus, release permission and AF lock flags are cleared, and the focus display on the display device 45 and the electronic buzzer 46 are cleared.
is turned off and the process proceeds to S229 (S217, S227). S
In step 229, the loop time is checked, and if the loop time has not elapsed, the process returns to S211 to repeat the above processing, and if the loop time has elapsed, the process returns to S109 of the main routine.

[0054] If the AF is in operation at the time of the check in S211, the process advances to S231 to enter AF processing. In S231,
Performs integration and calculation processing. Then, it is checked whether the calculation result is valid, and if it is valid, the process jumps to focus check processing (S233). If a valid calculation result is not obtained, it is checked whether the supplementary rib light emission mode is in effect. Note that the auxiliary light projection mode is a mode in which auxiliary light (a striped pattern) is projected onto the subject from an auxiliary light projector (not shown) when the subject brightness is lower than a certain value. Since it is difficult to follow a moving object in the auxiliary light projection mode, a moving object prediction mode prohibition flag is set (S237). Then, the auxiliary light is emitted to perform integration and calculations, and the calculation results are checked. If the calculation results are valid, the process proceeds to focus check processing; if not, the process proceeds to S243 (S239
, S241). If it is not the auxiliary light projection mode, the process jumps from S237 to S241 and proceeds to S243.

Since no calculation result is obtained in S243, search integration and calculation are performed to detect the in-focus point while DC driving the AF motor 39 (S245).
, checks whether the calculation result is valid, and if it is not valid, executes the search integration and calculation again, and if it is valid, jumps to the driving direction check process (S245, S247
).

"Re-integration and focus check processing" Next, regarding the subroutine regarding reintegration and focus check processing,
This will be explained with reference to FIGS. 11A and 11B.

The reintegration process is the second and subsequent integration process. In the reintegration process, first, a reintegration flag is set, and then integration and defocus amount calculation processing are performed (S251
, S253). Then, when the calculation result of the defocus amount is invalid, the process jumps to AFNG processing, and when it is valid, the process goes to focus check (S255).

In the focus check process, it is checked whether or not the camera is in focus. If it is in focus, the focus display LED of the display device 45 is turned on and the electronic buzzer 46 is activated, but it is necessary that the camera is in focus. Otherwise, nothing is done (S261).

Next, it is checked whether this is the first out-of-focus time during low contrast, and if it is the first time, the process returns to S251 and the above-mentioned processing such as integration is repeated. and is not in moving object prediction mode, sets a release permission flag during AF (S26
3, S265, S267, S269). If the AF lock flag is not in the AF lock state (if the AF lock flag is not set), that is, the AF lock flag will be set during the first processing after focusing, but if it is already set, the AF lock flag will be set during the second processing after focusing. Therefore, the process jumps to AF lock processing (S283) (S271, S273).

When the AF lock flag is set in S273,
Provided that the subject has high contrast and the moving object prediction prohibition flag is down, wait for a release interrupt for a specified period of time, and if no release interrupt occurs within the specified period of time, clear the AF release permission flag and release the camera. and clears the AF lock flag to enable AF processing (S275, S277, S279, S281). Then, AF lock processing is started, and the loop time is waited for to elapse. When the loop time has elapsed, the process returns to S109 of the main routine (S283). Note that the AF lock process is a process in which the AF operation is not performed while the photometry switch SWS is on. If the subject is not of high contrast or if the moving object prediction prohibition flag is set, proceed to AF lock processing without performing wait processing (S275, S277, S
283).

Further, in the focus check process of S265, if the camera is out of focus, the process jumps to S285, and it is checked whether or not the moving object prediction mode is set. If the moving object prediction mode is not entered here, the process proceeds to the moving object check process in S301, and if the moving object prediction mode is entered, the process proceeds to S321 (
S285).

This moving object check process is a process performed to confirm whether the subject is a moving object (see FIG. 2).
, 3). In the moving object check process, first, the secondary focus width, focus, release permission, and AF lock flags are cleared, and the focus display LED of the display device 45 and the electronic buzzer 46 are turned off (S301).

Next, check the reintegration flag. If the reintegration flag is cleared, it is the first processing, so the process jumps to pulse calculation processing. If it is set, the previous lens driving direction and the current lens driving direction are Compare (S303, S3
05). If they are in the same direction, the moving object determination counter is decremented by 1 on the condition that the contrast is high and the moving object prohibition flag is cleared (S307, S309,
S311, S313). Note that the initial value of the moving object determination counter is set to "3". When the directions are not the same, when the contrast is not high, or when moving object prediction is prohibited, the process directly jumps to the pulse calculation process without performing the above counting process (S307, S309, S311).

When the moving object judgment counter is decremented, the counter value is checked, and if it is 0, the moving object prediction mode flag is set and the moving object judgment counter is reset (set to 3 in this embodiment). The process then proceeds to pulse calculation processing (S315, S317, S319). Counter value is 0
Otherwise, proceed directly to pulse calculation processing (S315)
.

If the moving object prediction mode is already in the moving object prediction mode when checking the moving object prediction mode in S285, the focus flag, release permission flag, and AF lock flag are cleared;
Focus display LED of display device 45 and electronic buzzer 46
is turned off (S321).

Next, the number of defocus pulses is calculated based on the integration result in S253 (S323). Note that this number of defocus pulses is a value for the subject at the midpoint of the integration time.

[0067] Then, the calculated number of movement pulses of the object,
The moving speed of the subject image (moving object tracking speed) is calculated based on the current number of defocus pulses and the time between the integration intermediate points, and when the calculation is completed, a moving object tracking flag is set to identify that the moving object is being tracked ( S325, S32
7). Then, the number of defocus pulses is counted by counter 35d.
, starts AF motor constant speed control, and jumps to constant speed control processing (S329, S331).

[0068] Also, S2 in the focus check process
In the moving object prediction mode check in step 67, if the moving object prediction mode is already in the focusing range, the above S323
to go into.

"Pulse Calculation and Backlash Driving" Next, pulse calculation and backlash driving will be explained with reference to FIGS. 12A and 12B. The pulse calculation process is a process of calculating a pulse amount based on the defocus amount and a pulse amount for eliminating backlash. Backlash drive refers to a drive for eliminating backlash in the built-in gears, joints 47 and 57, and lens drive mechanism 55 of the AF motor 39 when the drive direction of the AF motor 39 changes. In this example,
Backlash driving is performed separately before driving based on the defocus amount.

In the pulse calculation process, first the amount of defocus pulses is calculated, and further, when the driving direction changes from the previous time, the number of backlash pulses is calculated and a backlash drive flag is set (S351, S353). The amount of backlash pulses is the number of pulses and rotation direction necessary to eliminate backlash. The backlash amount of the camera body 11 is stored in advance in the E2PROM 43, and the backlash amount of the photographic lens 51 is stored in the ROM of the lens CPU 61.
35 RAM 35b.

Next, when the backlash drive flag is not set, the number of defocus pulses is set in the counter, and when it is set, the number of backlash pulses is set as the number of AF pulses in the counter, and DC drive of the AF motor 39 is started ( S355, S357). Note that the DC drive in this embodiment means that the AF motor 39 is driven by substantially direct current.

When not in backlash drive (when the backlash drive flag is cleared) and when not in moving object prediction mode (when the moving object prediction mode flag is cleared), the count value of the counter and the predetermined number of pulses When the count value is larger, the moving object prediction mode flag is cleared, overlap integration is started, and the defocus pulse amount is calculated (S369).
). Note that overlap integration refers to a process of performing integration while driving the lens.

Once the number of defocus pulses is determined, it is checked whether the number of defocus pulses is valid or not. If invalid, the process returns to S363, and if valid, the process proceeds to drive direction check processing (S373) (S371). .

In the drive direction check process, it is checked based on the defocus amount whether the focus position is in front of the subject image position and within a predetermined range, that is, whether it has gone beyond the predetermined range (S373, S375).
. If the defocus amount is within a predetermined range, the number of defocus pulses is calculated based on the defocus amount calculated in S369, and the number of defocus pulses is set in a counter.
363 (S375, S377). If it has gone too far, the AF motor 39 is braked to stop the lens drive, and the process returns to the reintegration process (S375, S379). Note that the brake of the AF motor 39 refers to the brake of the AF motor 39 in this embodiment.
This means short-circuiting the input terminals of the motor 39.

In addition, when backlash drive is necessary, when the moving object prediction mode is entered, or when the counter value is smaller than a predetermined pulse number, the AF pulse counter value and the predetermined number of pulses are compared and the AF Wait until the counter value becomes equal to or less than the predetermined number of deceleration pulses (S359 to S3
65, S381, S383).

When the AF pulse counter value becomes less than the predetermined number of deceleration pulses, the AF
After applying the brake to the motor 39, the process jumps to constant speed control processing (S385, S387).

During backlash drive, the number of AF pulses is compared with a predetermined number of pulses, and if the number of AF pulses is less than the predetermined pulse, the AF motor 39 is braked and the process proceeds to constant speed control processing (S389, S391, S38
7). Then, backlash elimination driving is performed by constant speed control.

When the number of AF pulses is equal to or greater than the predetermined number of pulses, the process waits until the DC drive for eliminating backlash is completed. When this DC drive is finished, the AF motor 39 is braked, the backlash drive flag is cleared, the backlash drive is finished, and the process returns to S355 (S391, S393, S395, S397). Next, the number of defocus pulses is set in the AF pulse counter, and lens driving processing for normal AF processing is performed.

With the above processing, backlash driving is executed before defocus driving. Moreover,
When the drive amount for backlash is larger than a predetermined value, DC driving is performed in a short time, and when it is smaller than a predetermined value, constant speed control is used to perform accurate driving without overshooting.

"Constant Speed Control" The constant speed control process will be explained based on the constant speed control process shown in FIGS. 13A and 13B. Constant speed control processing means that the AF motor 3
In this embodiment, the main CPU 35 drives the AF motor 39 at a predetermined constant speed based on the interval of pulses output from the encoder 41. In this embodiment, the AF motor 39 is driven at a constant speed at the subject image moving speed S or twice that speed.

In the constant speed control process, first, the constant speed control speed of the normal process is set (S401). Next, it is checked whether the moving object is being followed, and if the moving object is being followed, the constant speed control speed Sn is set to double speed 2Sn (S403, S405). If integration is in progress, constant speed control speed S is applied to follow the moving object at the following speed.
After setting n, the process proceeds to S411, but if integration is not in progress, the process directly proceeds to S411 (S407, S409).

[0082] In S411, it is checked whether integration is in progress and moving object tracking is in progress. If integration is in progress and moving object tracking is in progress, the process jumps to S459 to check whether the integration has ended; If not, S413
The process proceeds to (S411).

[0083] In S413, it is checked whether the constant speed control time has elapsed or not. If it has elapsed, the process proceeds to check whether the end point has been reached (S415), and if it has not elapsed, the process proceeds to S421. In S421, it is checked whether the AF pulse is being outputted from the encoder 41. If the AF pulse is not being outputted, the process returns to S411, and if it is being outputted, the process proceeds to S423.

[0084] In S423, it is checked whether the pulse count has ended or the driving time for moving object tracking has ended. Then, if any of them has been completed, the process proceeds to brake processing, and if none of them have been completed, the process proceeds to end point check processing (S423, S437, S425).

In S425, it is checked whether the focusing lens 53 has reached the end point, and the end point detection timer is reset. Then, for deceleration processing, it is checked whether the count value of the AF pulse counter has become smaller than the speed switching pulse number, and when it has become smaller, the drive speed is switched to a lower speed, and then the AF motor 39 is braked. Proceed to S433 (S427, S4
29, S431). In S433, it is checked whether the AF motor 39 is being driven, and if it is being driven, the AF motor 3
After applying the brake at step 9, the process returns to S401, and if the vehicle is not being driven, the process directly returns to S401.

[0086] In S423, if the pulse count has ended or the driving time for moving object tracking has ended, the process advances to S437 and brakes are applied to the AF motor 39 in order to stop the AF motor 39. Then, it is checked whether the moving object prediction mode and the AF lock flag are set, and if so, the AF release permission flag is set and the process proceeds to release processing (S439, S441,
443). If it is not in the moving object prediction mode or the AF lock flag is cleared, it is checked whether it is backlash drive, and if it is backlash drive, it proceeds to backlash drive end processing, and if it is not backlash drive, it clears the reintegration flag. and check the loop time, and when the loop time has elapsed, press A in S109.
Returning to the E process, if the loop time has not elapsed, the process proceeds to reintegration process (S445, S446, S447).

In the integration process, initial settings regarding integration and calculation are made, integration is started, and an integration-in-progress flag is set (
S451, S453, S455). If the moving object is being followed, the process jumps to constant speed control processing (S401), and if the moving object is not being followed, it is checked whether the integration has ended (S457, S459).

If the integration is not completed, if the integration time is longer than a predetermined time, it is considered that the subject brightness is low, so a moving object prediction mode prohibition flag is set and the process proceeds to S465; if it is short, the process directly proceeds to S465 (S459, S461, S4
63). Then, if the moving object is not followed, the process returns to S459.
Wait until the integration is completed (S459 to S465). If the moving object is being followed, constant speed control is being performed, and the process returns to S413. When the integration is completed, if the tracking drive is in progress, the AF motor 39 is braked, the integral data is input from the range sensor unit 21, the defocus amount is calculated, and the process returns to the step following the step where the integration started. (S46
7, S469, S471, S473).

"Defocus Calculation" The defocus calculation process will be explained with reference to the calculation process shown in FIG. 14. First, contrast calculation and checking processing is performed (S501). The contrast calculation is a calculation to find the sum of the differences between the integral data of each CCD of the CCD distance measurement sensor unit. Then, it is checked whether the contrast is sufficient for calculating the phase difference, and if it is insufficient, it returns, and if it is sufficient, it proceeds to the phase difference calculation process (S
503, S505).

[0090] In S505, first, a phase difference is calculated by a correlation method or the like. Then, if it is the 10-bit selection mode, the process advances to S551 of 10-bit selection processing, and the 10-bit
S509 of normal calculation processing if not in t-select mode
The process proceeds to (S507). 10 bits when first entered
Since the selection mode is not entered, the process advances to S509. Note that the 10-bit select mode means that when the degree of coincidence (correlation degree) between the integral data of the standard part and the reference part of the CCD line sensor is not very good, the 10-bit selection mode is This is a process of sequentially selecting regions and finding regions with good matching (high correlation).

[0091] In S509, it is checked whether the phase difference calculation result is valid, and if it is not valid, returns.
If it is valid, it is checked whether the degree of matching is good (S509, S511). If the degree of matching is good, defocus calculation is performed (S511, S531). Then, if the first out-of-focus flag at low contrast is set and the defocus amount is less than a predetermined amount, compare the previous and current defocus amounts and if the current one is smaller. The current defocus amount is set as the valid value, the defocus amount OK flag is set, and the process returns (S53).
5, S537, S539, S541, S543). If the low contrast first out-of-focus flag is not set, the current defocus amount is enabled, the defocus OK flag is set, and the process returns (S533, S541, S543).

Even if the first out-of-focus flag at low contrast is set, if the defocus amount is equal to or greater than a predetermined value, the previous defocus amount is valid, so the defocus OK flag is set and the process returns (S533, S535, S543
). Even if the first out-of-focus flag is set during low contrast and the defocus amount is less than the predetermined value, if the defocus amount this time is larger than the previous time, defocus is OK.
Set a flag, maintain the previous defocus amount, and return (S533, S535, S537, S539,
S543).

With the above processing, when the subject has a low contrast, the closer (smaller) of the two previous defocus amounts is used as the measurement value, which prevents malfunctions due to measurement errors, such as in-focus display. Flickering and erroneous activation of the AF motor 39 can be prevented.

If the degree of matching is not very good, the phase shift is compared with a predetermined value, and if the shift is smaller than the predetermined value, it is checked whether the contrast is high, and if the contrast is low, the process proceeds to S531 ( S52
1, S523). 10b for high contrast
It sets the it select mode flag and the moving object prediction mode prohibition flag, selects the first 10 bits, further sets the number of checks, and returns to S505 (
S523, S525, S527, S529).

When the 10-bit select mode flag is set, 10-bit select processing begins at S507. 10b
In the it selection process, first, the selected 10bit
Check whether the effective value of the defocus amount based on t has been determined at once, and if it has been determined, if the contrast is greater than the previous one and twice the previous value, then the previous DATA is invalidated. Clear the 10-bit select defocus OK flag and proceed to S559 (S551, S553, S555, S557). 10b
If the it select defocus amount has not been determined even once, proceed to S559, and if the current contrast is less than twice that of the previous one, 10 bit select defocus is OK.
Proceed to S559 without clearing the flag (S551 to S5
57). If the contrast is smaller than the previous one, the 10-bit data selected this time will not be used, so select S.
Jump to 571 (S551, S555).

In S559, it is checked whether the phase difference calculation result is valid, and if it is, defocus calculation is performed on the condition that the degree of coincidence is good (S55
9, S561, S563). If the 10bit select defocus OK flag is set, select the defocus amount that is closer (front) between the previous time and this time, and select 10bit select defocus OK flag.
It sets the it select OK flag and proceeds to S571, but 1
If the 0-bit select defocus OK flag is cleared, do nothing and set the 10-bit select OK flag and proceed to S571 (S565, S567, S569
). In addition, even when the phase difference calculation result is invalid or the degree of coincidence is poor, S5 can be performed without calculating the defocus amount.
The process advances to 71 (S559, S561).

[0097] In S571, the next 10 bits are selected. Then, the 10-bit select number counter is decremented by 1, and if the counter value is 0, S505
Return to S507, S551 to S57 until it becomes 0
5 is repeated (S573, S575). In addition, 1
The initial value of the 0 bit selection number counter is set to "4". If the value of the 10-bit select number counter is 0, flicker countermeasure processing at low contrast (S533) is performed on the condition that 10-bit select defocus is OK.
), and if it is not OK, clear the 10-bit select mode flag and return (S575, S577,
S579).

[0098] The above S505, S507, S551 to S
CCD with good phase difference and matching due to 575 processing
Integral data from a group of sensor bits can be selected. Note that in this embodiment, the number of bits to select is 10.
Although it is set to bit, the configuration is not limited to this and may be set to any number of bits.

"Moving Object Following Speed Calculation" Moving object following speed calculation will be explained with reference to the calculation process shown in FIG. 15. First, the time Tn between the previous integration midpoint and the current integration midpoint is calculated (S601). Note that the reason why the integration midpoint is used as a reference is that the integration time changes depending on the subject brightness.

Next, if the moving object is not being tracked, calculate the moving speed S1 of the subject image before tracking and return (S6
05). If the moving object is being followed, the number of movement pulses MPn-1 of the object image that would have moved within the time Tn is calculated (S607). Then, the previous and current driving directions of the lens are compared, and if the directions are the same, the moving object tracking overshooting flag is cleared, the current DPn is added to MPn-1, and the process proceeds to the tracking speed correction calculation process (S617). S60
9, S611, S613, S615). If the direction is different, it is considered that the moving speed of the subject has slowed down, that the subject has stopped and gone too far, or that the moving direction of the subject has changed, so proceed to S637 (S609, S611, S637)
. If neither the same direction nor the different direction occurs, that is, if some abnormality has occurred, the moving object prediction mode is cleared and the process proceeds to reintegration processing (S611, S613, S655).

"Following speed correction calculation 1" In the following speed correction calculation 1, first, the following formula is used to calculate the subject tracking speed S.
n is calculated (S617). Xn=Tn/(MPn-1+DPn) Sn=1/Xn=(MPn-1+DPn)/Tn

01
Next, a moving object tracking time Cn (sum of integral data input time and calculation time) is set (S621). Then, if the current calculation result indicates that the focus is in focus and the release switch SWR is turned on, the AF lock flag is set and the time lag correction time (integral data input time + calculation time Cn + release time lag equivalent time) is calculated. However, when the subject moves away, the constant speed control time is set to 1/2 of the time lag correction time, and when the subject approaches, the constant speed control time is set to the time lag correction time and the process returns (S623, S625, S627, S
629, S631, S633, S635).

If the camera is not in focus, or if the release switch SWR is not turned on even if it is in focus, the tracking operation will continue, so the process jumps to S635 and sets the constant speed control time to the moving object tracking time. Return (S623, S625).

[0104] When the driving direction of the lens is different from the previous driving direction and the current driving direction, if MPn-1-DPn≧0, then MPn-1-DP
n is calculated, and the moving object tracking speed Sn is corrected using the following formula: Xn=Tn/(MPn-1-DPn) Sn=1/Xn=(MPn-1-DPn)/Tn. Check (S639, S64
1).

[0105] If the focus is on, the moving object tracking overshooting flag is reset and the process returns (S643, S643).
645). If it is not in focus, check whether the moving object tracking overshoot flag has already been set, and if not, set the tracking overshoot flag, set the calculation time Cn, and wait for the calculation time Cn. Proceed to reintegration processing (S647, S649, S651, S65
3). If the moving object tracking overshooting flag has already been set, the moving object prediction mode is cleared and the process returns (S647, S655).

"Focus Check Processing" Referring to FIG. 16,
The focus check process will be explained. In this focus check process, the focus range is widened when a moving object is being followed, and the focus display is prevented from flickering when the contrast is low.

First, when in AF mode, leave the preset focusing width (predetermined defocus amount) as is;
If it is the moving object prediction mode, the focusing width is widened, and then it is checked whether the current defocus amount is smaller than a predetermined amount (S701, S703, S705).

First, if the defocus amount is larger than a predetermined amount (an amount larger than the in-focus width), the out-of-focus 1 at low contrast
The process resets the second flag and returns (S719).

If the defocus amount is within the predetermined amount and in focus (below the in-focus width), the in-focus flag is set, the low-contrast out-of-focus flag is cleared, and the display device 45 is displayed. Turn on the focus display LED and activate the electronic buzzer 46 (S705, S707, S709, S711
, S713).

[0110] Then, if the AF operation is in progress, the focus width corresponding to a predetermined image plane movement speed for checking a moving object is set and the process returns; if the AF operation is not in progress, the process returns as is (
S715, S717).

[0111] When the defocus amount is greater than or equal to the in-focus width and smaller than a predetermined amount, flicker countermeasure processing is performed during low contrast. When the low-contrast out-of-focus first flag has already been set, or even if it is not set, in the moving object prediction mode, clear the low-contrast out-of-focus flag and return (S721, S723, S719).

[0112] If the first low contrast flag is cleared and the moving object prediction mode is not in effect, the process returns directly when in the auxiliary light projection mode (S721,
S723, S725). When the mode is not the auxiliary flash mode and the mode is the 10-bit select mode, the first out-of-focus flag at low contrast is set and the process returns (S
725, S733). When not in 10-bit select mode, if the integration time is longer than the predetermined value, it will return as is, and even if it is shorter, if it is not low contrast, it will return as is, and if it is low contrast, it will set the first out-of-focus flag at low contrast. Set and return (S729, S731, S733).

As described above, if the process is out of focus when the process is started, if the contrast is low, by setting the low contrast first out of focus flag and returning, the process can be changed from in focus to out of focus. Even if the focus changes, the out-of-focus display processing (processing in S301) is not performed during the first out-of-focus state, so flickering of the in-focus display is prevented. Note that in this embodiment, the out-of-focus display is displayed when out-of-focus occurs twice in a row, but the number of times that out-of-focus changes to the out-of-focus display is arbitrary. For example, if out-of-focus continues three times in a row, The configuration may be such that an out-of-focus display is performed when the object is in focus.

"End point detection processing" Next, the focusing lens 53
Processing when the lens reaches the closest focus position or the infinite focus position will be described with reference to the end point detection process shown in FIG. 17. This process is also a process in which the AF motor is stopped by detecting that the end point has been reached or the end point has not been reached but the focusing lens group 53 cannot be driven due to some external force.

[0115] Check whether the AF motor is being driven,
If it is not being driven, the process proceeds to end point detection timer resetting processing (S791). When the AF motor 39 is being driven, when the pulse interrupt flag is set (when a pulse is output from the encoder 41), the process proceeds to end point check (S771), and when it is not set (no pulse is output), end point detection is performed. Proceed to timer count (S755).

[0116] The end point detection timer counting process is performed when the encoder 41 does not output pulses for a certain period of time or more even though the AF motor 39 is being driven, the focusing lens 53 has reached the end point or is in a state where it cannot be moved. Therefore, this is a process for stopping the AF motor 39. Here, first, when detecting an end point, the pulse presence flag is cleared, the end point detection counting flag is set, and the end point detection timer is counted down (S75).
5, S757, 759). Then, check whether the end point detection timer is 0, and if it becomes 0, proceed to end point processing,
If it is not 0, return (S801).

In the end point check process, whether or not the focusing lens 53 has reached the end point is detected by detecting changes in the pulses output from the encoder 41. Here, first,
Clear the pulse interrupt flag (S771).

[0118] Then, it is checked whether or not the end point detection timer is counting, and if it is not counting, or even if the end point detection timer is counting but the pulse presence flag is set when the end point is detected, the end point detection timer is set. The process proceeds to resetting processing (S775, S777). While the end point detection timer is counting, if the flag with pulse at end point is not set, the flag with pulse is set when the end point is detected, returns if constant speed control is in progress, and proceeds to S757 if constant speed control is not in progress (S777 , S77
9, S781).

The end point detection timer resetting process is a process that initializes the data related to end point detection, and clears the end point detection timer counting flag and the pulse presence flag when detecting an end point, and sets the end point detection timer data. Return (S791, S793, S795).

"Second Embodiment of Moving Object Prediction AF" Next, a second embodiment of the moving object prediction AF of the present invention will be described with reference to FIGS. 18 and 19. This second embodiment is characterized in that constant speed control is performed even during calculation. In other words, the method calculates the lens driving speed corresponding to the moving speed of the subject image, and performs integration while controlling the focusing lens at a constant speed at that speed to correct the tracking speed.The basic principle is the same as that of the first embodiment. The difference is that constant speed control is performed even during predetermined calculations. In order to execute calculations and constant speed control in parallel in this way, this embodiment includes a CPU 36 for controlling the motor drive IC 37.

First, the DC lens drive M is performed based on the defocus pulse DP when the moving object prediction mode is determined (when the moving object prediction mode is entered). When the DC lens drive M is completed, integration I1 is performed, and in calculation C1, the speed S1 at which the subject image moved during the time T1 between the midpoint of the integrations I and I1 is calculated using the same equations as the above-mentioned equations 1 and 2. Then, the lens drive time is determined using the following formula. (T1+(I1/2)+C1)/2...■

[
Then, constant speed control lens driving M1 is performed at three times the approaching speed S1 for the driving time calculated using the above formula (2). This constant speed control lens drive M1 allows the focusing lens 53 to be moved to near in-focus in a short time. Note that this triple speed drive is (T1+I1/2+C1
) substantially the same effect as driving the lens at twice the object image moving speed S1 for the time period S1.

When control lens driving M1 at triple speed is completed, integral I2 and calculation C2 are performed while performing constant speed control (lens driving M2) at follow-up speed S1. In this calculation C2, it corresponds to the amount of movement of the subject image when it is assumed that the subject image was moving at the tracking speed S1 within the time T2 from the midpoint of the previous integral I1 to the midpoint of the current integral I2. The number of pulses MP1 is calculated using the same formula as the above formula (
MP1=T2×S1). Furthermore, the velocity S2 at which the subject image moved during the time T2 is determined using the following equations similar to the above-mentioned equations (1) and (2). X2=T2/(MP1+DP2) S2=1/X2 Then, during integration I3 and calculation C3, constant speed control lens drive M3 is performed at follow-up speed S2.

[0124] Thereafter, similarly, the subject image becomes Sn within time Tn.
Find the tracking speed Sn using the number of pulses MPn-1 corresponding to the amount of movement of the subject (from the second calculation during tracking, MPn-1=Tn×Sn-2) assuming that the object was moving at a speed of −1. While performing the integration In+1 and the calculation Cn+1, the lens drive Mn+1 is performed using the follow-up speed Sn (constant speed control).

[0125] Through the above-described tracking control, the moving subject is almost always kept in focus. Further, when the release switch is turned on during the following operation, constant speed control is performed at the previously determined follow-up speed Sn even while the mirror is up (see FIG. 19). Due to this constant speed control during mirror up, the object is almost in focus when the front shutter curtain actually starts traveling (starts exposure).

Next, regarding the second moving object tracking operation, FIG.
0 and the flowchart shown in FIG. First, the time Tn between the previous integration midpoint and the current integration midpoint is calculated (S801).

[0127] Next, check whether the moving object is being followed,
If the moving object is not being tracked, calculate the subject image movement speed (following speed) Sn-1 before tracking and return (S803,
S805). If a moving object is being tracked and this is the first calculation during tracking, the number of pulses MP1 corresponding to the amount of movement of the subject image during time T2 is calculated, and from the second time onwards during tracking, the number of pulses MP1 is calculated during time Tn. Calculate the number of pulses MPn-1 corresponding to the amount of movement of the subject image (S803, S807,
S809, S811).

[0128] Then, the previous and current driving directions of the focusing lens 53 are compared, and if they are the same, the moving object tracking overshooting flag is cleared, and the calculated number of movement pulses M of the object image is set.
Add the current number of defocus pulses DPn to Pn-1 and proceed to S821 (following speed correction calculation 2 process) (S8
13, S815, S817, S819). If the driving directions are different, it is considered that the moving speed of the subject has slowed down, the subject has stopped, or the moving direction of the subject has changed and the subject has gone too far, and the process advances to S833 (S81
3, S817). neither in the same direction nor in different directions, i.e.
If any abnormality occurs, clear the moving object prediction mode and proceed to reintegration processing (S815, S817, S84
9).

"Following Speed Correction Calculation 2" In the following speed correction calculation 2, first, the moving object following speed Sn is calculated using the following formula (S821). Xn=Tn/(MPn-1+DPn) Sn=1/Xn

Next, if the focus is on and the release switch SWR is on, the AF lock flag and the AF release permission flag are set and the process returns (S825, S8
27, S829, S831). If the camera is not in focus or the release switch SWR is not turned on, the process returns directly (S825, S827).

[0131] Furthermore, when the driving directions of the lenses are different, if MPn-1-DPn≧0, the subject tracking speed Sn is determined by the following formula (S815, S833,
S835, S837). Xn=Tn/(MPn-1-DPn) Sn=1/Xn=(MPn-1-DPn
)/Tn

[0132] If the focus is on, the moving object tracking overshooting flag is cleared and the process returns (S8).
39, S841). If it is not in focus, check whether the moving object overshooting flag is set, and if it is already set, the moving object has gone too far and this is the second processing, so clear the moving object prediction mode and return (S843, S849).
. In other words, since there is a possibility of an error or the like if it goes too far once during tracking of a moving object, it is configured to exit when it goes too far twice. Note that the configuration may be such that it can be exited once, or it can be exited any number of times, including three or more times.

If the moving object tracking overshooting flag is not set yet, this is the first time the moving object tracking overshooting flag has been set, so set the moving object tracking overshooting flag, input the integral data, set the calculation time Cn, and wait for the calculation time Cn. Then, the process proceeds to reintegration processing (S843, S845, S847).

"Constant Speed Control 2" Next, another process of constant speed control 2 will be explained with reference to the subroutine shown in FIG. 21. First, the pulse interval (Xn) is set. If the moving object is being followed, the moving object following speed Sn (
1/Xn). Furthermore, if it is the first processing after entering moving object tracking, the triple speed of the moving object tracking speed Sn will be reset, but if it is the second or subsequent processing, it will not be changed (S8
53, S855, S857, S859). If the calculation has been completed (calculation end flag = 1), input and set data such as AF pulse and moving object tracking speed Sn, clear the calculation end flag, and return to S851 (S861, S
863, S865).

[0135] If the calculation has not been completed (calculation end flag = 0), it is checked whether the constant speed control time has elapsed, and if it has elapsed, the end point detection timer is counted, and the AF motor 39 is driven. If it is inside, the process returns to S861 (S867, S869, S871). AF motor 39
If it is not being driven, start driving the AF motor 39 and
The process returns to 851 (S871, S873).

[0136] If the constant speed control time has not elapsed,
Check whether the AF pulse is output (
S867, S875). S if there is no AF pulse output
Returning to step 861, if there is an output, proceeding to step S877.

[0137] In S877, it is checked whether the mirror up is completed, and if it is completed, the AF motor 39 is braked and stopped (S877, S879). If the mirror up is not completed, check whether the AF pulse count has finished, and if it has finished,
Stop the AF motor 39, check the loop time,
If the loop time has elapsed, AE processing is performed, but if it has not elapsed, the process proceeds to reintegration processing (S879, S881).
, S883).

[0138] If the AF pulse count has not ended, the end point check and end point detection timer are reset, and it is checked whether the AF pulse counter value has become smaller than the speed switching pulse (S879, S879).
885, S887). If it becomes smaller, switch the control speed to a low speed to decelerate, apply the brake to the AF motor 39, and
The process proceeds to S893, and if it is not smaller, the process directly proceeds to S893 (S887, S889, S891).

[0139] In S893, it is checked whether the AF motor is being driven. If not, the process returns to S851; if it is, the brake is applied and the process returns to S851 (S895).

"Integral processing" In the integral processing subroutine,
After initial settings are made, integration is started and an integration flag is set (S901, S903, S905). Then, the process waits while checking the integration time until the integration time ends (S907, S909). At this time, if the integration time becomes larger than a predetermined value, the subject is too dark, so a moving object prediction mode prohibition flag is set (S909, S9
11).

[0141] When the integration is completed, the integral data is input and the defocus amount is calculated (S907, S909, S9
11). If this calculation result is valid, if the moving object prediction mode is entered, the moving object tracking speed is calculated, and a calculation end flag is set, and the process proceeds to S925; if it is not valid, the process returns to S901, and if the calculation result is valid, If the moving object prediction mode is not entered, the process directly advances to S925 (S919, S
921, S923). In S925, it is checked whether the AF release permission flag is set. If it is not set, the process returns to S901, and if it is set, the process proceeds to release processing.

With the above processing, as shown in FIGS. 18 and 19, the focusing lens 53 is controlled at a constant speed even during integration and calculation, so that it can more faithfully follow the movement of the subject, that is, constantly This makes it possible to maintain focus.

As described above, according to this embodiment, even if the subject is in focus, it can be determined that the subject is moving when it is out of focus in the AF process after a predetermined period of time. It is possible to reliably determine that the subject is moving even when the focusing lens moves from the short-distance focusing position to the long-distance focusing position when the subject is approaching.

[0144] In this embodiment, in the AF process after focusing, it is determined that the bear is a moving object when the unfocused bear is detected three times, but the number of times is not limited to three. Further, in the AF processing after focusing, a configuration may be adopted in which it is determined that the object is a moving object when the object is out of focus a predetermined number of times, even intermittently, or when it is out of focus a predetermined number of times consecutively.

[0145]

As described above, according to the present invention, when the subject has a low contrast, the defocus amount at that time is temporarily memorized and the distance is remeasured, and the remeasured defocus amount is
Since the smaller defocus amount is selected as the defocus amount by comparing the memorized defocus amounts, distance measurement errors are reduced, and the lens is driven when it is determined that the focus is out of focus. The focus display no longer flickers.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing the configuration of an embodiment of a single-lens reflex camera to which an automatic focus device of the present invention is applied.

FIG. 2 is a graph showing the AF operation mode of the automatic focusing device before entering the moving object prediction AF mode operation.

FIG. 3 is a graph showing another AF operation mode of the automatic focusing device before entering the moving object prediction AF mode operation.

FIG. 4 is a graph showing an AF operation mode of the automatic focusing device in a moving object prediction AF mode.

FIG. 5 is a graph showing an AF operation mode of the automatic focusing device when the moving object prediction AF mode operation goes too far.

FIG. 6 is a graph showing the AF operation mode when the release switch is turned on while operating in the same moving object predictive AF mode while the subject approaches.

FIG. 7 is a graph showing the AF operation mode when the release switch is turned on while operating in the same moving object predictive AF mode while the subject is away from the subject.

FIG. 8 is a flowchart regarding the main operation of the automatic focusing device.

FIG. 9 is a flowchart regarding reference timer interrupt of the automatic focusing device.

10A and 10B are flowcharts regarding AF processing of the automatic focusing device.

11A, 11B, 11C are flowcharts regarding reintegration and moving object prediction AF in the same AF process.

12A and 12B are flowcharts regarding defocus and drive pulse calculation in the same AF process.

13A, 13B, and 13C are flowcharts regarding constant speed control processing in the same AF processing.

14A, 14B, and 14C are flowcharts regarding defocus calculation, 10-bit selection processing, and flicker prevention processing at low contrast in the same AF processing.

15A, 15B, and 15C are flowcharts related to moving object tracking speed and correction calculation in the same AF process.

FIG. 16 is a flowchart regarding focus check and low contrast flicker countermeasure processing in the same AF processing.

FIG. 17 is a flowchart regarding end point detection processing in the same AF processing.

FIG. 18: Second moving object prediction AF of the automatic focus device of the present invention
It is a graph showing processing operation.

FIG. 19 is a graph showing the operation when the release switch is turned on during the second moving object predictive AF processing operation.

20A, 20B, 20C are flowcharts related to the moving object tracking speed calculation process, etc. of the second moving object predictive AF process.

21A and 21B are flowcharts regarding constant speed control in the second moving object predictive AF process.

FIG. 22 is a flowchart regarding integration processing in the second moving object predictive AF processing.

[Explanation of symbols]

11 Camera body 12 Photography lens 13 Main mirror 14 Sub mirror 17 Photometry IC 21 CCD sensor unit for distance measurement 23 Peripheral control circuit 25 Shutter mechanism 27 Aperture mechanism 35 Main CPU 36 CPU 39 AF motor 41 Encoder 45 Display device 47 Joint 51 Photography lens 53 Focusing lens 55 Lens drive mechanism 61 Lens CPU

Claims (4)

[Claims] 1. Defocus amount measuring means for measuring a defocus amount for a specific subject; contrast determining means for determining whether the specific subject has low contrast; and the contrast determining means determines that the specific subject is not low contrast. When the contrast determining means determines that the contrast is low, the defocus amount at that time is temporarily memorized, the defocus measuring means is re-operated, and the memorized defocus amount is combined with the current defocus amount. A focus detection device comprising: defocus amount selection means for comparing the measured defocus amounts and selecting the smaller one as the defocus amount. 2. The focus detection device according to claim 1, further comprising focus determination means for determining whether or not the focus state is in focus based on the defocus amount measured by the defocus amount measuring means, A focus detection device characterized in that the defocus amount selection means selects the defocus amount regardless of contrast when the focus determination means determines that the focus is in focus. 3. The focus detection device according to claim 1 or 2, further comprising lens driving means for driving a focusing lens based on the defocus amount selected by the defocus amount selection means. A focus detection device characterized by: 4. The focus detection device according to claim 2, further comprising display means for displaying focus when said focus determination means determines that the object is in focus.