JP2906252B2 - Automatic focusing device - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an automatic focusing device such as a camera driven to focus on a moving subject.

[Prior Art] Conventionally, a technique called tracking, moving object prediction or predictive driving has been known as a technique for driving and controlling a photographing lens without delay so as to focus on a moving subject.

For example, in the automatic focusing apparatus disclosed in Japanese Patent Application Laid-Open No. Sho 60-214325 by the present applicant, an image is obtained by predicting a future lens position from the latest defocus amount obtained from the focus detection means and the past several times of defocus amount. The drive of the lens is controlled.

[Problems to be Solved by the Invention] However, in an automatic focus adjustment device that performs drive control by predicting a lens position in the future from the latest defocus amount and a plurality of past defocus amounts, a moving object When the lens position changes almost linearly on the time axis, the prediction accuracy is high and accurate drive control can be performed. However, when the lens position does not change linearly, the prediction does not work well, and the lens is truly focused. However, there was a problem that overrun and underrun occurred.

For example, when an in-focus driving is performed on an automobile (constant speed moving body) traveling at a constant speed as shown in FIG. 15 with the closest distance of the camera being D, ideally, the solid line L in FIG.
Is obtained with respect to the time axis of the focusing lens position shown in FIG.

That is, in FIG. 13, the vertical axis represents the lens position Z (t), the horizontal axis represents time t, and the lens position is represented by a lens extension amount based on the position of ∞. Zd is the closest distance D
Is the lens position with respect to.

FIG. 14 shows a case in which a conventional apparatus for performing a straight line prediction on the subject shown in FIG. 15 which produces the lens movement pattern shown in FIG. 13 is applied.

In FIG. 14, a broken line M shows the relationship between the lens position and time when drive control is performed according to the conventional technique.
At the lens positions Z (t1) and Z (t2), the solid line L indicating the ideal pattern changes sharply. Therefore, if the prediction is performed by the conventional apparatus based on the previous data, an overrun occurs.

In particular, if the accumulation time of the sensor used in the focus detection means is low and the exposure time is between the focus detection operations, the time interval at which the defocus amount can be obtained is widened. , The prediction accuracy worsens,
The overrun becomes more remarkable, and in the worst case, the amount of movement of the subject during the time interval at which the defocus amount is obtained becomes large, so that focus detection becomes impossible and the lens cannot be driven.

The present invention has been made in view of such a conventional problem, and has as its object to provide an automatic focus detection device that can accurately follow a moving subject and focus.

Means for Solving the Problems In order to achieve this object, the present invention provides a focus detecting means for detecting a focus adjustment state of a photographic lens, output information obtained by detection of the focus detecting means, and a detection time. Or, based on an input from an external setting unit, preliminarily determines lens movement pattern information indicating a relationship of “the position of the imaging lens at each time”, and stores the imaging device with the imaging lens. The lens drive includes: a start unit that starts based on lens movement pattern information; and a lens drive control unit that can repetitively drive and control the photographing lens in accordance with the lens move pattern information based on a start command of the start unit. The control means controls the default value indicating the in-focus state of the photographic lens during drive control of the photographic lens based on a start command of the activating means. The lens movement pattern information is calibrated based on the focus amount.

[Operation] In the automatic focusing apparatus of the present invention having such a configuration, the drive control of the photographing lens is performed according to the lens movement pattern information stored in advance. Can accurately focus without performing complicated calculations such as tracking drive control every time as in the past.Furthermore, even if the movement of the subject is a non-linear movement, it follows the lens movement pattern information, so it is possible to focus accurately. You can get a suitable photo.

Further, when the subject temporarily deviates from the focus detection area of the camera, the focus is pulled to the background and the focus is largely deviated in the conventional tracking drive control, but according to the present invention, the lens stored in advance is used. Since driving is performed in accordance with the movement pattern information, accurate focus can be maintained without focusing on the background.

Even if the subject moves more complicatedly, it can be accurately followed by calibrating the lens movement pattern information.

FIG. 1 is a block diagram showing a first embodiment of the automatic focusing apparatus according to the present invention.

In FIG. 1, a light beam from a subject 1 is
A subject image is formed on the film surface through the, but is branched from the shooting optical axis by a sub-mirror of a known configuration,
The light is guided to focus detection means 3 arranged at the bottom of the camera or the like.
The focus detection means 3 is composed of a known focus detection optical system, an image sensor, and a focus detection calculation device. The focus detection means 3 receives a light beam from the subject 1 and detects the direction and amount of deviation between the subject image plane and the film plane (hereinafter referred to as "defocus"). Called "amount").
The defocus amount is generated at a certain time interval because the focus detection unit 3 needs a charge accumulation time by the sensor and a focus detection calculation time. Further, when a camera photographing operation is performed, the focus detection cannot be performed during that time, so that the generation interval becomes further longer.

The lens position detecting means 5 includes: an absolute position of the photographing lens 2;
For example, means for detecting a lens extension amount when the ∞ setting position of the photographing lens 2 is set to a reference 0 position, an encoder for generating a predetermined number of pulses corresponding to a predetermined movement amount of the photographing lens 2, and an encoder output It comprises a counter for adding and subtracting pulses according to the direction of movement, and reset means for initializing this counter. That is, when the power of the automatic focusing device is turned on, the counter of the photographing lens 2 is reset to ∞, the counter is reset, and the subsequent encoder output pulses are counted, so that the contents of the counter correspond to the absolute position of the photographing lens 2. Information.

The focusing lens position determining unit 4 converts the defocus amount generated by the focus detecting unit 3 into a moving amount of the photographing lens 2, and sets the lens position at the time when the defocus amount occurs to a moving amount based on the defocus amount. By adding
A lens position (hereinafter, referred to as a “focus lens position”) for focusing the photographing lens 2 on the subject at the present time is determined. If the photographing lens is moving during the focus detection operation time (sensor accumulation time, calculation time) of the focus detection means 3, the correction according to the movement amount during the focus detection operation time is performed by the defocus amount or the focusing lens. Add to position.

The lens movement pattern storage means 6 is a means for preliminarily storing lens movement pattern information indicating a relationship between a lens position and time corresponding to the movement of a specific moving subject.

For example, if the lens movement pattern information corresponding to the shooting situation of the moving subject is indicated by a solid line L in FIG.
The time t and the lens position Z (t) are stored in a memory such as a ROM or a RAM at addresses X, X + 1, X + 2,.
o, Z1, Z2,..., and the addresses X, X +
1, X + 2,... Are determined by values that are integral multiples of the time interval ΔT. In FIG. 2, X represents a reference address, ΔT represents a time interval, and Z represents a lens position.

Referring again to FIG. 1, the lens drive control means 7
Based on the lens movement pattern information, the focusing lens position information, and the lens position information, the relationship between the position and the time is controlled so that the taking lens 2 follows and focuses on a specific moving subject.

The control operation of the lens drive control means 7 will be described with reference to FIGS.

In FIG. 3A, a solid line L indicates a lens movement pattern which is a relationship between a lens position and time corresponding to the movement of a specific moving subject. This lens movement pattern L
When focused lens position information Z1 (initial position) is generated at a certain time t1, a point P1 where the lens movement pattern L is Z1 is searched. Here, the lens movement pattern L is a bivalent function having two values on the time axis with respect to the lens position Z.
P1 is uniquely determined.

When the first point P1 is determined, the lens drive control means 7 determines the relationship between the lens position of the taking lens 2 and time until the in-focus lens position information Z2 is generated at the next time t2.
Drive control is performed using the lens position information so as to be a part L1 of the lens movement pattern L starting from P1.

Accordingly, the lens position that is the actual control target is the solid line L1 in FIG. 3B.

Here, the broken line L 'in FIG. 3B shows the relationship between the lens position and the time when the movement exactly corresponds to the actual movement of the subject.
The broken line L 'varies from subject to subject, and does not exactly match the lens movement pattern L in FIG. 3A, but its error is expected to be small. Therefore, unless the time from time t1 to t2 is extremely long, the error at time t2 is small even if the lens drive control is performed with the solid line L1 as the target.

When focus lens position information Z2 is newly generated at time t2, a point P2 where the lens movement pattern L is Z2 is searched for on the condition that it is closest to P1 after the previous point P1 as in the previous time, and the next time Until the focusing lens position information Z3 is generated at t3, drive control is performed using the lens position information so that the relationship between the lens position of the photographing lens 2 and time becomes a part L2 of the lens movement pattern L starting from the point P2. Perform Therefore, the error between the solid line L 'indicating the lens position corresponding to the real subject and the control target L1 which has occurred at the time t2 is calibrated at this time.

Hereinafter, the same operation is repeated every time the latest focusing lens position information Z3, Z4, Z5,... Occurs at times t3, t4, t5,. L4, L5, ・ ・ ・
Is required.

Since the control targets L1, L2,... Become discontinuous at times t1, t2,.
・ Does not match at this time, but L at other times
The difference between the lens position and the time L ′ when the lens is actually controlled can be controlled by driving control so as to coincide with L1 and L2. Few.

In addition, the focusing lens position information generated at time intervals is not used to directly control the driving of the photographing lens 2 but is used only for calibration of the lens movement pattern. The drive control error does not increase.

In order to further reduce the error at times t1, t2,..., The next target is determined according to the difference ΔZ between the focus lens position and the actual lens position when new focus lens position information is generated, and the lens moving direction. The time axis up to the control can be enlarged or reduced. For example, the difference ΔZ between the focus lens position Z2 generated at time t2 in FIG. 3B and the control target L1 at time t2 is + in this case and the lens moving direction is +, so that the actual moving speed of the subject is faster than expected. That is, the actual lens movement pattern L 'is the stored lens movement pattern L
The error at the time t3 can be reduced by reducing the control target L2 from the time t2 according to the amount of the difference ΔZ (increase the time change rate), considering that the control target L2 is reduced in the time axis direction.

An embodiment in which the drive control of the taking lens 2 is performed based on the above concept will be described with reference to FIGS. 4A and 4B. In FIG. 4A, lens driving control is started in accordance with the lens movement pattern Z = F1 (t) stored from the time t1 and the lens position Z1, and the focused lens position information is obtained at the time t2.
When Z2 is obtained, a time S2 at which the lens position Z2 is reached when the lens movement is continued along the function Z = F1 (t) is obtained as shown in the figure. Lens position Z occurring at time t2
Can be considered to be due to the fact that the actual movement of the subject is performed according to a lens movement pattern Z = F2 (t) different from the lens movement pattern Z = F1 (t) stored in advance. The lens movement pattern Z = F2 (t) can be expressed as the following equation because it can be assumed that the lens movement pattern Z = F1 (t) is reduced / enlarged on the time axis and moved in parallel.

F2 (t) = F1 (a2 * t + b2) (A) The conditions at times t1, t2, S2, lens positions Z1, Z2 are as follows.

F2 (t1) = F1 (t1) = Z1 F2 (t2) = F1 (S2) = Z2 (B) From the equation (B), the coefficients a2 and b2 of the equation (A) can be obtained as follows.

a2 = (t2-t1) / (S2-t1) b2 = {(S2-t2) / (S2-t1)} * t1 (C) Therefore, from time t2, it is determined by (A) and (C). The drive control of the photographing lens may be performed according to the lens movement pattern Z = F2 (t).

Next, in FIG. 4B, lens drive control is started from the lens position Z2 at time t2 according to the lens movement pattern Z = F2 (t), and when the focus lens position information Z3 is obtained at time t3, the function Z = F2 ( When the lens movement is continued along t), a time S3 at which the lens position Z3 is reached is obtained as shown in the figure. In the same way as described above, it is assumed that the lens movement pattern Z = F3 (t) is performed in accordance with the lens movement pattern Z = F3 (t) between times t2 and t3. If it is considered that the image has been enlarged and translated, it can be expressed as in the following equation.

F3 (t) = F2 (a3 * t + b3) = F1 (a2 * a3 * t + b2 + a2 * b3) = F1 (A3 * t + B3) (D) Also, conditions at times t2, t3, S3, and lens positions Z2, Z3 Becomes as follows.

F3 (t2) = F2 (t2) = Z2 F3 (t3) = F2 (t3) = Z3 (E) From (E), the coefficients a3, b3, A3 and B3 of the formula (D) are as follows. You can ask.

a3 = (t3-t2) / (S3-t2) b3 = {(S3-t3) / (S3-t2)} * t2 A3 = a2 * a3 B3 = a2 * b2 + b2 (F) Therefore, time t3 Then, the drive control of the photographing lens may be performed in accordance with the lens movement pattern Z = F3 (t) determined from (D) and (F).

Generally, when the in-focus lens position information Zn is obtained at time tn, the lens movement pattern Z = Fn-1 (t) immediately before time tn
Then, the time Sn at which the lens position Zn is reached is obtained, and then the lens drive control is performed from time tn according to the lens movement pattern Z = Fn (t) expressed by the following equation.

Fn (t) = Fn-1 (an * t + bn) = F1 (An * t + Bn) where an = (tn-tn-1) / (Sn-tn-1)｝ bn = ｛(Sn-tn) / ( Sn−tn−1)｝ * tn−1 An = an × An−1 = a1 × a2 ×... × an Bn = Bn−1 + An−1 × bn (G) A1 = 1, B1 = 0 , a1 = 1, b1 = 0 Further, the time axis is fixed without being scaled. For example, the time change rate K of the error from the error ΔZ at the time t2 in FIG. 3B is expressed as K = ΔZ / (t2−t1). Then, the control target L2 for the next time t2 to t3 may be corrected as L2 = L2 + K (t−t2).

Generally, the lens movement pattern L is a multivalent function having a plurality of values on the time axis with respect to the lens position Z as shown in FIG. 5, but the search for the point corresponding to the focusing lens position Z starts from the smaller time axis. After that, it is only necessary to select the one whose movement direction is the closest after the previous point.

If the subject is shifted to a distance other than the expected distance and the point corresponding to the generated focusing lens position Z does not exist on the lens movement pattern L, the time axis from the previous point on the lens movement pattern L In the vicinity of the point where the elapsed time from the previous time to the current time is apart, the focusing lens position Z
And the point with the least difference is selected.

Further, when the focus cannot be detected at a certain time, or when the difference between the generated focus lens position Z and the current lens position is larger than a predetermined value, the drive control is continuously performed by continuing the previous control target. You may do so. This continuation control prevents the photographing lens from being focused on the background even when the moving object is removed from the distance measurement area and the focus is detected on the background.

6 and 7 are flowcharts when the first focusing lens position determining means 4 and the lens drive control means 7 are constituted by a microcomputer.

First, in FIG. 6, when the focus detecting means 3 obtains the defocus amount, an interrupt is generated in step S1.
In step S2, the in-focus lens position is obtained from the defocus amount and the lens position, and an interrupt is generated in step S3. The operation up to this point is the operation of the focusing lens position determining means 4.

Next, when the focusing lens position determining means 4 generates an interrupt in step S4 in FIG. 7, a corresponding point is obtained in step S5 based on the focusing lens position and the lens movement pattern information by the method shown in FIG. The photographing lens 2 according to the lens movement pattern L after the point
Is performed. The above is the operation of the lens drive control means 7.

FIG. 8 shows a second embodiment of the present invention. This embodiment is characterized in that the contents stored in the lens movement pattern storage means 6 can be set from the outside. The configuration is such that the lens movement pattern external setting means 9 and the storage control means 8 are added to the example.

The storage control means 8 is a control means for resetting and updating the contents of the memory of the lens movement pattern storage means 6. When updating the lens movement pattern, the memory control means 8 permits writing of the memory of the lens movement pattern storage means 6 and , The operation of the lens drive control means 7 is prohibited. When the writing of the memory is permitted, the lens movement pattern external setting means 9 writes the new lens movement pattern information into the memory of the lens movement pattern storage means 6. When the writing is completed, the storage control means 8 prohibits the writing of the memory of the lens movement pattern storage means 6 again and permits the operation of the lens drive control means 7.

The photographer may provide the desired lens movement pattern information directly to the lens movement pattern external setting means 9 or may select a desired pattern from a plurality of lens movement pattern information prepared in the lens movement pattern external setting means 9 in advance. You may make it select. Alternatively, the photographer may give the information (distance, speed, etc.) of the photographing condition, and calculate the corresponding lens movement pattern information inside the lens movement pattern external setting means 9. For example it is possible to calculate the lens movement pattern information by entering a short distance D and the initial distance D _o and vehicle speed V is in photographing conditions such as Figure 15.

First, the moving pattern d (t) of the subject distance is as follows.

Next, the lens movement pattern Z (t) is given by the following equation when the focal length of the taking lens is f and the distance from the film surface to the subject is d (t) in the above equation (1).

The second equation (2) can be treated as an ^{approximation, Z (t) = f 2} / (d (t) -2f) ··· (3).

FIG. 9 shows a third embodiment of the present invention, characterized in that the contents stored in the lens movement pattern storage means 6 can be set by the output of the focusing lens position determination means 4, The storage control means 8 is added to the embodiment of FIG.

The storage control means 8 is a control means for resetting and updating the contents of the memory of the lens movement pattern storage means 6. When updating the lens movement pattern, the memory control means 8 permits writing of the memory of the lens movement pattern storage means 6 and , The operation of the lens drive control means 7 is prohibited. When the writing of the memory is permitted, the focusing lens position determining means 4 writes the focusing lens position information into the memory of the lens moving pattern storage means 6 as new lens moving pattern information. When the writing is completed, the storage control means 8 prohibits the writing of the memory of the lens movement pattern storage means 6 again and permits the operation of the lens drive control means 7.

FIG. 10 shows how the focusing lens position determining means 4 writes lens movement pattern information to the lens movement pattern storage means 6. In other words, the focusing lens position determining means 4 sets the focusing lens position information Z (t) (shown by x) at the times t1, t2... Tn, tn + 1. ) Is written in the lens movement pattern storage means 6 as a pair with time t. Time interval Δt
When is large, the lens position information between them may be interpolated and written.

In the case of FIG. 10, the collection and writing of the lens movement pattern information are performed with the photographing lens 2 stopped in order to shorten the time interval Δt, but the driving operation of the photographing lens 2 and the operation of the focus detection means 3 are overrun. If the lap can be performed, the lapping may be performed while driving the photographing lens 2. Further, the lens movement pattern information may be obtained as an average of a plurality of times instead of performing only one detection. After writing the lens movement pattern information once, the lens drive control means 7 operates accordingly. The acquired focusing lens position information may be fed back to the lens movement pattern information.

Further, a lens having a long focal length generally has difficulty in capturing a moving subject and has a large amount of extension, so that focus detection is likely to be impossible, and it is difficult to collect lens movement pattern information. Therefore, when collecting the lens movement pattern information, a lens having a short focal length is used, the lens position information is once converted into the subject distance information by using the above equation (2), and then the focal length f of the lens to be used is used. May be converted again into lens position information.

FIG. 11 shows a fourth embodiment of the present invention. In this embodiment, the activation of the lens drive control means 7 is made controllable by the activation means 10, and the embodiment shown in FIG. It has a configuration to which means 10 is added.

By providing the activating means 10, the photographer can activate the operation of the lens drive control means 7 at a desired time or a desired subject distance.

For example, when it is desired to operate the lens drive control means 7 for a desired time, the activation means 10 may be a simple switch, and the operation of the lens drive control means 7 may be permitted / prohibited according to the state of the switch.

When starting from a desired distance, the photographer sets the distance in the starting means 10. For example, if you want the photographer so that activates the lens drive control unit 7 when the following and became a distance D _o settings such as FIG. 15, the distance D _o less information that the starting means
Set to 10. The activating means 10 sets the distance _Do to the (2)
It converts the lens position Z _o by formula, monitors the focus lens position information generated with a time interval. As a result, when the subject as shown in FIG. 12 is farther than the distance D _o is the focusing lens position activating means 10 because less Z _o does not start the lens drive control unit 7. Time tn + when the subject approaches
Focus lens position information Z generated in 1 (tn + 1) is the activation means 10 exceeds a Z _o activates the lens drive control unit 7.
For setting the distance _Do , the photographing lens 2 may be manually set to the distance _Do , and the set lens position may be taken into the starting means 10 via the lens position detecting means 5.

Before activating the lens drive control means 7 by the activating means 10, the taking lens 2 may be stopped,
The photographing lens may be driven according to the defocus amount as in conventional lens drive control.

In the above embodiment, the lens movement pattern storage means 6 stores the lens movement pattern information in the form of the time t and the lens position Z (t). However, the time t and the subject distance d ( Lens movement pattern information may be stored in the form of t). In this way, even a lens having a variable focal length, such as a zoom lens, can be converted into a lens position at each zoom position according to the formula (2) according to the focal length.

Further, the drive control of the photographing lens 2 is mainly performed based on the lens movement pattern information, and the focus lens position information or the defocus amount is used for calibration or correction. May be switched to the lens drive control according to the embodiment of the present invention when focus detection becomes impossible or when it is determined that the distance is short and the conventional predictive drive is difficult.

In each of the embodiments shown in FIGS. 1, 8, 9, and 11, the defocus amount of the photographing lens 2 is detected by the focus detection means 3 by TTL, but the subject distance is directly measured by a triangulation method such as active. You may do so.

[Effects of the Invention] As described above, according to the present invention, since the drive control of the photographing lens is performed in accordance with the lens movement pattern information stored in advance, the conventional method is applied to a moving subject that moves periodically. The focus can be adjusted accurately without performing complicated calculations such as tracking drive control every time as described above. Photos are obtained.

Further, when the subject temporarily deviates from the focus detection area of the camera, the focus is pulled to the background and the focus is largely deviated in the conventional tracking drive control, but according to the present invention, the lens stored in advance is used. Since driving is performed in accordance with the movement pattern information, accurate focus can be maintained without focusing on the background.

Even if the subject moves more complicatedly, it can be accurately followed by calibrating the lens movement pattern information.

[Brief description of the drawings]

FIG. 1 is a block diagram of an embodiment of an automatic focusing device according to the present invention; FIG. 2 is an explanatory diagram of a lens movement pattern storage method; FIG. 3A, 3B, 4A, 4B, and 5 are operations of lens movement control means. FIG. 6 is an operation flowchart of a focusing lens position determination unit; FIG. 7 is an operation flowchart of a lens drive control unit; FIG. 8 is a block diagram of another embodiment of the automatic focusing apparatus according to the present invention; 9 is a block diagram of another embodiment of the automatic focusing device according to the present invention; FIG. 10 is an explanatory diagram of information on a lens movement pattern; FIG. 11 is a block diagram of another embodiment of the automatic focusing device according to the present invention. FIG. 12 is an explanatory diagram of the operation of the activating means; FIG. 13 is an explanatory diagram of lens movement pattern information; FIG. 14 is an explanatory diagram of a conventional tracking operation; FIG. . [Description of Signs of Main Parts] 1: Subject 2: Photographing lens 3: Focus detection means 4: Focusing lens position determination means 5: Lens position detection means 6: Lens movement pattern storage means 7: Lens drive control means 8: Storage Control means 9: Lens movement pattern external setting means 10: Starting means

Claims (1)

(57) [Claims] A focus detection means for detecting a focus adjustment state of the photographing lens; and output information obtained by the detection of the focus detection means and a detection time thereof, or based on an input from an external setting means. Storage means for preliminarily determining and storing lens movement pattern information indicating the relationship of "the position of the photographic lens at each time"; activating means for activating the photographic lens based on the lens movement pattern information; Lens drive control means for repetitively driving and controlling the photographing lens in accordance with the lens movement pattern information based on a start command of the means, wherein the lens drive control means is configured to control the photographing lens based on a start command of the start means. Calibrating the lens movement pattern information based on the defocus amount indicating the in-focus state of the photographing lens during the drive control of An automatic focus adjustment device.