JPS5924811A - Automatic focusing device - Google Patents

Abstract

PURPOSE:To reduce the rate of a change in the driving extent of a lens and to move stably a lens by controlling the driving extent of the lens in accordance with the information from a means for discriminating the focusing state. CONSTITUTION:The light past the image pickup lens is bisected and is imaged on a pair of sensors, the outputs whereof are binary encoded and are inputted to the peak position detector 9-11 in a correlator. The signal dx indicating the focusing state is outputted and is stored in a memory 13-1; at the same time, a focusing lens is driven, and a fresh correction output dx is stored in the memory 13-1. The previous content is transferred into a memory 13-2. The absolute value of the difference between the two contents is determined with a subtractor 13-3, and is compared with a set value E in a comparator 13-6. If the value is smaller than E, a counter 13-7 is counted up and when the count value attains a set number, the signal for increasing the driving extent of the lens is outputted to a control device 10, by which the focusing state is attained in a short time. Thus the lens operates stably.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to an automatic focusing device, that is, a device that automatically focuses an imaging optical system without relying on visual inspection.

FIG. 1 shows an example of an automatic focusing device based on the prior art. The two light rays from the subject passing through one end IA of the lens 1 pass through a small reflecting mirror 4C.

An image is formed on one end 3A of a one-dimensional solid-state image sensor (hereinafter simply referred to as a sensor) 3 made of a CCD or the like. This image will be called the A image. Here, the sensor is a photoelectric element that converts a spatial image into an electrical video signal. The light ray 2B passing through the other end IB of the lens 1 forms an image on the other end 3B of the sensor 3 via a small reflecting mirror 4D. This image will be called Bfl.

The sensor 3 is arranged at a position optically equivalent to the imaging surface 12. When the lens 1 is extended forward from the correct focusing position, the optical path changes as shown by the dotted line in Figure 1.
Therefore, the two images on parts 3A and 3B of sensor 3 also move.

The moving direction is such that when the lens 1 is extended forward, the two images approach each other, and when the lens 1 is retracted backward, the two images are moved away from each other.

Therefore, the distance d between the two images (image A and image B) is detected.

By controlling the lens position so that the interval matches the value at the time of correct focusing, the lens 1 can be maintained at the correct focusing position. When the sensor 3 is driven by the drive circuit 7, the optical images (A image, B image) on the sensor 3 are converted into time-series electric video signals. The generated video signal is converted into two binarized video signals by a binarizer 51 and a distributor 6. One of them is obtained from an image formed on a portion 3A of the sensor 3 by the light beam 2A, that is, an image A, and this is called a binary image signal A. The other one is also:
This is obtained from the B image formed on 3B, and is called a binary video signal B. Both signals are input to the correlator 9 via signal lines 8A and 8B, respectively.

A signal related to the distance between the A image and the B image is obtained from the correlator 9. This signal is input manually to the lens control device 10. The lens control device 10 causes the signal of the distance between the two images to match the value at the time of correct focusing.

Controls the lens drive section 11.

The correlator 9 will be explained using FIG. Two binary video signals A and B are first manually input to shift registers 9-1a and 9-1b. Here, a logic circuit is provided attached to each shift register 9-1a, 9-1b,
After the first application of the binary video signals A and B to the shift registers 9-1a and 9-1b, the shift registers 9-1a and 9-1b are put into circulation mode.

These logic circuits are AND gates 9-2a, 9-
2b and 9-3a, 9-3b, inverter 9-4
a, 9-4b.

Includes ORGATE 9-5a and 9-5b. Two binary video signals A and B are Antogame) 9-3a,
Receive at 9-3b. The control signal from the control signal generation circuit 9-7 becomes the human power for one of the ant gates 9-3a and 9-3b.

In addition, it is reversed by inverters 9-4a and 9-4b, and becomes the human power for one of the computer games 9-2a and 9-2b.

Further, the outputs of the shift registers 9-1a and 9-1b serve as the other inputs of the AND gates 9-2a and 9-2b, respectively. The output of Antogame) 9-2a and 9-2b is
Or game) 9-5a respectively.

It will be the human power for one side of 9-5b.

Let's talk about the operation. The logic signal "l" from the control signal generation circuit 9-7 is applied to AND gates 9-3a and 9-3b.
The new binary video signals A and B can be manually input to the shift registers 9-1a and 9-1b. This logic signal °゛1°゛ is the AND gate 9-
2a.

As a result of disabling shift registers 9-2b, the information previously stored in shift registers 9-1a and 9-1b is not reloaded. New binary video signals A and B are transferred to the shift register 9--in synchronization with clock signals CLKA and CLKB.
1a and 9-1b, the control signal generation circuit 9
The control signal from -7 changes to a logic signal °゛0°°.

This makes AND gates 9-2a and 9-2b usable, and AND gates 9-3a and 9-3b unusable. Therefore, shift registers 9-1a and 9-1b
is in circular mode. Shift registers 9-1a and 9-1
The contents of b are shifted in response to two clock signals CLKA and CLKB applied from control signal generating circuit 9-7. /Every time the logic signals II, Il+ and the logic signal "1" occur at the output of the shift registers 9-1a and 9-1b, and the logic signal II O++ and the logic signal II O++ occur, that is, the outputs of the two shift registers are equal. When a logic signal occurs, exclusive negative OR (hereinafter referred to as EX-NOR)
Gate 9-6 outputs a logic signal ++ 1 ++.

Then, from this output pulse train, the control signal generating circuit 9-
An output synchronized with the strobe pulse STR generated by 7 is taken out by an AND gate 9-9.

The output pulse of the AND gate 9-9 is output from the binary counter 9-9.
It is counted by 10.

Here, the respective clock pulses CLKA and CLKB exactly match the number of bits of the shift registers 9-1a and 9-1b. - when the circular shift is completely completed.

The count value accumulated in the counter 9-10 represents the degree of similarity between the binary video signals A and B during the cyclic shift (comparison cycle) period at the current state of the address counter 9-8.

Upon completion of the first cycle/shift in the contents of shift registers 9-1a and 9-1b at a given time (on completion of the first comparison cycle), - special lock pulses are generated in control signal generating circuit 9-. 7 to the clock input of shift register 9-1b. this is.

/Shift the contents of shift register 9-1b one bit at a time relative to the contents of shift register 9-1a. At the same time, the count value of address counter 9-8 increases by one. The counter 9-10 is then cleared, a second comparison cycle begins, and the output pulses of the AND gate 9-9 are counted again by the counter 9-10. This sequence continues until the number of cyclic shifts is equal to the maximum value of °n°° to be considered.

Thereafter, each time each comparison cycle ends, the contents of the counter 9-10 are changed by increasing the count value of the address counter 9-8 by l each time each comparison cycle ends.
8, and the similarity between the binary video signals A and B, that is, the correlation output.

Output sequentially. The peak position detector 9-11 detects the address when the degree of similarity is highest.

That is, it outputs a signal dX related to the interval d between two values. A technique for detecting the interval between binary values using a correlator is known, for example, from Japanese Patent Laid-Open No. 101111/1983.

Next, the lens control device 1o will be explained. Calculator 10A that calculates the lens drive direction and drive amount using the third force
and a motor drive circuit 10B for driving the motor of the lens drive unit 11 based on the calculation result. Since the motor drive circuit 10B can be easily configured using well-known conventional techniques, the explanation will be omitted here, and the arithmetic unit 10B will be omitted here.
A will be explained.

Each time the correlator 9 completes detection, it sends a detection completion pulse to the up/down counter 10-14. Output dx of correlator 9
is supplied to the seven-node terminal S of the up/down counter 10-14 (Finally. 08CIO-11 is a low frequency oscillator, and the up/down counter 1 is output through the ant gate 10-12.
Supply clock pulses to 0-14. The up/down counter 10-14 performs an amplifier count when a logic level ``1'' is input to the up/down terminal U/D, a down count when a logic level ``0'' is input, and a clock input terminal C. This is done in synchronization with a clock pulse applied to the Further, when a logic level °°1°° is input to the pulse input terminal P from the correlator 9.

The states of counters 10-14 are set to the contents of the binary signal applied to terminal S. Matrix 10-15 is
When the up/down counter 10-14 reaches a preset value "F" (corresponding to the value of dx at the time of focusing).

It outputs logic level II OI+. AND gate 10-12 is matrix 10-15 and 08CI
The output of O-11 is logically ANDed, and when the up/down counter 10-14 reaches the set value ++F", the clock pulse input to terminal C is prohibited.Inverter 10-14
17 inverts the outputs of matrices 10-15. Further, the OR gate 10-16 outputs the determination signal "equal" (logic level "l") generated from the comparator 10-9.
) and the signals from matrices 10-15. Inverter 10-18. and gate 1
0-19 is an or-game) The output of 10-16 is ANDed with the judgment signals of the comparator 10-9, ``thick'' and ``small.''

Next, the operation will be explained. Detection output content of correlator 9 d
When x is smaller than the set value F (dxO value at the time of focus), the output terminal of the judgment signal "small" of the comparator 10-9 is sent to the terminal U/D of the up/down counter 10-14 at the logic level "1°". is output and the up/down counter 10-1
4 is in an up-count state. At the same time, the dxO value is increased by the up/down counter 10 due to the detection completion pulse.
-14 terminal S is set.

Then, O outputs a clock pulse with a frequency f set as described later according to the moving speed of the main lens 1 (Fig. 1).
8C10-11 performs up-counting by the value of 1dx-Fl, and during the time until the output of the counter 10-14 reaches F, the judgment signal of the comparator 10-9 is
The motor 1 drive circuit 10B is operated by "Small" to drive the main lens 1 by 1. Then, when the output of the counter 10-14 becomes F, the output of the matrix 10-15 becomes the logic level °°0'', so the clock pulse that was manually input to the terminal C through the antagonal gates 0-12 stops, and At the same time, the output of inverters 10-17 reaches the logic level °
°1", the logic level °°1" is output through the OR gate 10-16, while the inharter 10-18
, comparator 10 by and gate 10-19
-9 judgment signal “small” becomes logic level °°0”.

This generates an instruction signal to the motor drive circuit 10B to stop driving the lens, and the main lens 1 stops.

Here, the amount of lens movement is expressed by lens movement speed (constant) x lens drive time (ldx-F I X 1/f). Therefore, by setting the clock frequency f to an appropriate frequency value in consideration of the lens movement speed,
Adjust the amount of lens movement.

For example, if the main lens 1 is located far from the in-focus position,
If the detection output II dXu at a certain point in time is 1 and F is 32, the up/down counter 10-14 is set to 1 from the terminal S, and the absolute value of dx-F is 31.
The count will only increase. Expressing this in terms of lens driving time, the main lens 1 is driven for 1/r x 31 seconds, then stops, and the main lens 1 is brought into focus at the in-focus position.

Next, if the output content dx from the correlator 9 is larger than the set value F, the output terminal of the comparator 1o-9's judgment Shinshu Elementary becomes logic level '0', so the counter 10-14
Logic level IT OI+ is manually input to terminal U/D.

The counters 10-14 enter a down-counting state.

The value of dx at this time is then set in the counter 10-14.

At a certain point in time. When the value of dx is 62, counter 1
0-14 is set to 62, and a down count is started toward the set value F (=32). And 1dx-F
If you count down l = 62-32 = 30,
The output of the counter 10-14 becomes the set value F-32, and the output of the matrix 10-15 becomes the logic level ゛°0'', so the clock pulse that was manually inputted through the controller 1012 stops, and at the same time, the inverter 10 -17 output becomes logic level ``1°''.

Logic level ``1°'' through OR gates 10-16
is output, and becomes a lens drive stop instruction signal to the motor drive circuit 10B to stop the main lens 1. That is, 1/f
×30 seconds Move main lens 1.

The above-mentioned conventional technology has the following drawbacks. In the above-mentioned conventional technology, the amount of drive W of the lens 10 to the in-focus position is determined by the distance d between the two images on the portions 3A and 3B of the sensor 3.
It is proportional to the deviation x-dx-F from a certain reference value. That is, by the proportionality constant a.

W”ax ・・・・・・・・・・・・・・・・・・
It can be expressed as (1).

By the way, generally speaking, if the subject is at an infinite distance, when the subject approaches a distance L from the front focal point of the lens, the image point will be at a distance 2 given by the following equation (Euton's equation). move only.

Z = f2/L where f is the focal length of the lens. As is clear from this equation, the smaller the focal length f of the lens, the less movement of the image point with respect to movement of the subject over a certain distance. If the movement of the image point is small, the amount of change X in the distance d between the two images on the sensor 3 will also be small, and therefore the drive amount W given by the above equation (1) will also be small.

This causes the following problems. That is, when a conventional automatic focusing device is applied to a variable focal length lens such as a zoom lens, the amount of lens drive changes as the focal length of the lens changes.

The lens is driven at high speed on the long focal length side, and at low speed on the short focal length side. Therefore, with zoom lenses, where either the operation becomes unstable at high speeds or the response speed is slow at low speeds, focusing is usually done by moving the focus lens. When the lens is driven by a certain amount, the amount of image point movement is large on the long focal length side and small on the short focal length side.

The present invention is intended to solve the above-mentioned problems. In other words, the focal length information of the imaging device is obtained from the sensor output,
Information corresponding to iris information, etc. is obtained, and the lens and drive amount during focusing are controlled according to this information. For example, when a conventional TTL automatic focusing device is applied to a variable focal length lens such as a zoom lens, information corresponding to the focal length is obtained from the sensor output, and this information is used to drive the lens. It is an object of the present invention to reduce the difference in focusing response speed between the lens on the long focal length side and the short focal length side by adjusting the focal length.

FIG. 4 shows a configuration diagram of an embodiment of the present invention. 13 is a focal length detector, which detects the focal length of the variable magnification lens system (not shown), that is, the position of the zoom ring, based on a signal from the correlator 9 that corresponds to the distance between the two images on the sensor 3. is calculated and the information is sent to the lens control device 10. Lens control device 10
controls the lens drive amount based on this information.

That is, the value of the proportionality constant α in the above-mentioned equation ( ) is changed depending on the position of the zoom ring.

The rest of the configuration and operation shown in FIG. 4 are the same as those of the prior art shown in FIG. 1, so their explanation will be omitted.

FIG. 5 shows a specific configuration diagram of an embodiment of the focal length detector 13 of the present invention. When the correlation by the correlator 9 in FIG. The resulting binary output dx is manually inputted in the memo January 3-1 of FIG.

Also, at this time, the contents stored in Memo January 3-1 are moved to Memo January 3-2. The output signal X of memo January 3-1 and the output signal Y of memory 13-2 are
3 will be done manually. The subtracter 13-3 calculates the two human input signals X
and an output c=1x-y1 representing the absolute value of the difference between and Y.

It generates an output when X)Y.

Output C from subtractor 13-3 is input to comparator 13-6. The comparator 13-6 compares the reference value set in advance with the human power signal C, and when C(E), 1
Outputs two clock pulses. This clock pulse is input to a binary counter 13-7.

On the other hand, the output of the subtractor 13-3 is manually input to the memo January 3-4. At the same time, the contents previously stored in memo January 3-4 are moved to memory 13-5. Memo 1
Month 3-4. The outputs of notes January 3-5 are the outputs of exclusive OR (EX-OR) gates 13-8, respectively. EX-O
The output of the R gate 13-8 is the logical sum (OR) gate 3-9.
It becomes a reset terminal for the counter 13-7 through one of the manual inputs. Further, a lens drive stop signal in the lens control device 10, that is, a signal indicating that the subject is in focus, is output from the OR
The counter 13-7 becomes seven knots of human power through the other human power of game 3-9.

The operation will be explained.

When the correlation in the correlator 9 is completed, the peak position detector 9
The signal dx indicating the degree of focus on the subject from -11 is stored in memo January 3-1. At this time, the focus lens system of lens l is driven by 1 depending on the degree of focus. Next, the correlation output dx after the focus lens system is driven by a certain amount is the same as last time.
Stored at 1. The contents previously stored in memo January 3-1 are moved to memory 13-2. This 2
The absolute value of the difference between the contents of the memo January 3-1.13-2 is obtained by the subtractor 13-3, and the value is 11: comparator 13-5
When the value is smaller than the set value "E", the counter 13-7 is counted up by one. That is, when the focus lens system moves by a certain distance.

When the corresponding movement of the image point is small (subtractor 13
-3 output IX-Yl is small), the counter 13-
7 is counting. When counter 13-7 counts a set number 9, for example 10, counter 13-7
A signal is output from -7, indicating that the focal length of the variable magnification lens system of lens 1 is in a small state. Furthermore, even when the degree of focus approaches the in-focus state, the movement of the image point with respect to movement of the focus lens system at a constant distance is reduced. A counter 13-7 is provided to distinguish between this state and a state where the movement of the image point is small due to a small focal length.

That is, 9. When a state in which the movement of the image point is small continues for a long time, it is determined that the focal length is small, and a signal for increasing the amount of lens drive is output from the counter 13-7 to the lens control device 10. If the amount of movement of the image is small because it is close to the in-focus state, the amount of lens movement up to the 9-in-focus position is also small, so the in-focus state is reached in a short time. Therefore, there is no need to change the amount of lens drive.

At this time, the signal indicating the 1-focus state is OR game H as described later.
3-9 to the reset terminal of the counter 13-7. As a result, the counter 13-7 is set and no signal for increasing the one-drive balance is output from the counter 13-7.

As a result of the above sequence, the driving amount of the lens 1 is increased in the lens control device 10, and when the lens 1 reaches the in-focus state, the lens 1 is stopped by the control according to the prior art described above, and at the same time, the counter be done.

Furthermore, because the amount of lens driving has been increased, the lens 1 may pass through without stopping at the focal point.

In this case, the D output from the subtracter 13-3 shows a value opposite to that before passing through the focused point. If the D output before passing through the in-focus point is at logic level ``1'', then after passing through the in-focus point it will be ``IQI+1''. If the D output before passing through the in-focus point is at logic level ``0'', after passing through the in-focus point it will be at ``1''. Become.

Immediately after passing the in-focus point, the value of D immediately after passing the in-focus point is stored in memo January 3-4, and memo IJ 13-5
is the value of D in the previous state, that is, D before passing the 9 in-focus point.
The value of is stored. As a result, the output of the EX-OR gero 3-8 becomes the logic level ゛°1'°.

Counter 13-7 is reset via OR gate 13-9. Therefore, once the lens I passes through the in-focus point, the amount of lens drive becomes small, and the in-focus state can be reached without causing hunting.

The output of the force counter 13-7 in the focal length detector 13 is:

The signal is sent to the lens control device 10 to control the amount of lens drive. A specific example thereof is shown in FIG. In this embodiment, a case will be described in which the output from the focal length detector 13 is used to control the amount of current supplied from the power source to the lens drive motor.

Connections are made between the emitter of the transistor Tr2 and the emitter of the transistor Tr3, the base of the transistor Tr2, the base of the transistor Tr3, the emitter of the transistor Tr4 and the emitter of the transistor Tr5, and the emitter of the transistor Tr5.
The base of the transistor Tr5 is connected to the base of the transistor Tr5, the base of the transistor Tr5 is connected to the base of the transistor Tr5, the collector of the transistor Tr2 is connected to the collector of the transistor Tr4, and the collector of the transistor Tr3 is connected to the collector of the transistor Tr5. Further, the collectors of the transistor Tr2 and the transistor Tr4 are connected to the power supply 10-1. The collectors of the transistors Tr3 and Tr5 are grounded via a resistor R.

Further, it is connected to the collector of the transistor Tri, ffl. The emitter of the transistor Trl is grounded, and the output of the focal length detector 13 is connected to the base. The output from the motor 6 drive amount calculator 10A described in FIG. 3 is input to the bases of the transistors Tr2 and Tr3 and the bases of the transistors Tr4 and Tr5. When the lens is extended, the output becomes "1°" on one side and "0°" on the other.

When the lens is retracted, the focal length detector 13 has a terminal output of 1'' and 0°.
At this time, the transistor Tri is cut off and the current for driving the motor flows through the resistor R. However, when the signal from the focal length detector 13 is 1'', the transistor Tri becomes conductive.

The motor and drive current is passed through the transistor T without passing through the resistor R.
Flows to ground via rl. Therefore, when the focal length detector 13 outputs a signal of 1°'.

That is, when a signal indicating that the amount of image movement is small because the focal length is small is output, the motor drive current increases and the amount of drive of the lens drive unit 11 increases.

Note that the transistors Tr2, Tr3, Tr4,
Since the circuit operation of Tr5 is a general well-known circuit, its explanation will be omitted.

Further, since the motor drive amount calculator 10A is the same as that of the prior art, a description thereof will be omitted.

As explained above (when the focal length f is short, the amount of change in r(-dxF) in equation (1) is small as described above.However, in the present invention, as described above, the motor, drive current increases, that is, the value of α in equation (1) above increases by 1 (becomes a
The value of the lens drive amount W, which is the product of Naruga (many)
Motor 5 drive current decreases.

The value of a in the above-mentioned equation (1) becomes smaller. As a result, lens 1
The value of drive iW is smaller than in the prior art case.

That is, according to one aspect of the present invention, the range of change in lens and drive amount W due to change in focal length f is smaller than in the case of the prior art.

In the above embodiment, as means for determining the in-focus state on the imaging plane of the imaging lens, means for detecting the separation distance between two images using a correlator was used. In addition to this method, a contrast method is also known as a method for determining the in-focus state.

In the contrast method.

Image sensors are placed optically in front and behind the correct focal plane, and the sharpness (contrast) of the image at these two positions is compared, and the lens position is controlled so that the two are equal. In this method.

Contrast of output video signals of two sensors) KA and K
The difference KA-KB of B is 1dx-Fl in the above example.
corresponds to Therefore, the present invention can also be applied to contrast methods.

To simplify this operation, a case where the subject is a point light source will be described. Contrast depends on the diameter of the circle of confusion of an image produced by a single point light source. That is, as the diameter of the circle of confusion increases, the contrast decreases.

When a point light source object is imaged, if the distance Z from the image side focal point to the image point is constant, then the focal length of the lens is f.

Namely.

(image) diameter of circle of confusion)” (D/, f)-z=(p/, f)
-(fA)=("-9t., (L is the distance from the focal point on the point light source side to the point light source)3, In other words, the diameter of the circle of confusion of the image depends on the diameter of the exit pupil and the focal length of the lens. Therefore, when the focal length is constant in an automatic focusing device using the contrast method, when the diameter of the exit pupil changes, a change corresponding to the change in the driving amount due to the change in the focal length in the above embodiment occurs. When the diameter of the exit pupil is large, the diameter of the circle of confusion of the image becomes large for each unit distance of the lens, and as a result, the lens is driven at high speed.When the diameter of the exit pupil is small, the diameter of the circle of confusion of the image becomes large for each unit distance of the lens, and as a result, the lens is driven at high speed. As the distance moves, the diameter of the circle of confusion of the image becomes smaller, and the lens is driven at a slower speed.

Even in such a case, if the present invention is used, changes in the lens and drive amount due to the size of the exit pupil can be reduced.

As explained above, according to the present invention, the amount of lens drive is controlled based on 1% of the various information of the imaging device from the sensor output and jiK9.
Even when the present invention is used in a conventional automatic focusing device, the amount of change in lens driving amount due to change in focal length of the lens is reduced compared to the case of the conventional technology. As a result, even on the long focal length side, the lens operates stably without being driven at high speed. Also,
Even on the short focus side, the lens is slow and not driven (.

The response time is also slow.

[Brief explanation of drawings]

FIG. 1 is a block diagram of a conventional automatic focusing device, FIG. 2 is a block diagram of a correlator 9, FIG. 3 is a block diagram of a lens control device 10, and FIG. 4 is an example of an automatic focusing device of the present invention. FIG. 5 is a block diagram showing an example of the focal length detector 13 of the present invention, and FIG. 6 is a block diagram showing an example of the lens control device of the present invention. 1: Imaging lens, 3: Image sensor, 5: Binarizer,
6: distributor, 7: drive circuit, 9: correlator. 10: Lens control device, 11: Lens drive unit, 13: Focal length detector. District 1, Figure 3, Figure 4

Claims (1)

[Scope of Claims] An imaging lens, means for determining a focusing state on an imaging plane of the imaging lens, and means for controlling a focusing state on the imaging plane of the imaging lens based on the determination result. In an automatic focusing device consisting of. An automatic focusing device comprising means for controlling a lens drive amount based on information from the focus state determining means.