JPH05346536A - Focal point detecting device - Google Patents

Abstract

PURPOSE:To detect a focal point state with high precision by making the reliability criterion for a detected image stricter when it is decided that defocusing is large than when small. CONSTITUTION:In a step 505, weighting for determining condition setting for deciding the reliability of the detected image is performed on the basis of reliability data on the detected image. By this condition setting, the weight coefficient A of a reliability threshold value FNC-NOR is set to '1' when a flag LRGPRDF is '0' (namely, when a detected defocusing quantity is small) or to 2/3 when the flag LRGPRDF is '1' (namely, when the defocusing quantity is large); and the value is saved. In steps 506 and 508-511, a reliability comparison is made to use 2/3 as the coefficient A used for the calculation in reliability decision making when the defocusing quantity is large or '1' when the defocusing quantity is small.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a camera provided with a focus detecting device, and more particularly to an automatic focus detecting device for detecting a focus state from a relative positional relationship between two images representing a subject.

[0002]

2. Description of the Related Art Generally, an in-focus or out-of-focus state is detected using a storage sensor such as a CCD, and a lens is driven based on the result of the focus detection to automatically adjust the focus. Cameras equipped are known.

The principle of focus detection conventionally performed in such a camera is that a CC formed as a pair of line sensors by light fluxes passing through different regions of a photographing lens.
Images are formed on D respectively, and the relative position difference between these formed images (hereinafter referred to as "defocus amount") is determined by the focus position and the detected focus of the photographing lens. By taking advantage of the fact that there is a proportional relationship with the amount of deviation from the state, an appropriate calculation is performed on the image signal representing the spatial phase difference between the two images, and the focus state at that point, that is, the focused state. The moving distance of the taking lens necessary for the direction of deviation from and the in-focus position is calculated.

Further, in the conventional focus detection method as described above, the image formed on the pair of sensors is the basic information for focus detection, and therefore this image performs the detection method. The question is whether there is sufficient reliability. Therefore, the reliability of the detected defocus amount is determined, and when the reliability is poor, the driven shooting lens does not reach the in-focus position and is useless even if the focus is adjusted based on the uncertain image signal. The AF control is not performed because the operation is performed.

The reliability of an image in the conventional focus detection device is determined by using the coincidence of images formed on a pair of sensors as a guide. That is, if the signal patterns of a pair of images formed on a sensor array such as a CCD match, the defocus amount is obtained based on the relative positional relationship between these images to perform an appropriate focus adjustment operation. Can be done, but on the other hand, if these images do not match well,
This is because the information on the defocus amount obtained based on this becomes uncertain, and it is not possible to perform appropriate focus adjustment. Further, in Japanese Patent Application No. 60-247210 and Japanese Patent Application No. 0-247211, it is appropriate when the threshold value for judging the above reliability is not suitable by measuring the object brightness or the object contrast. It is determined that it is impossible to perform accurate focus detection, and the threshold value for determining the reliability is set to be strict so that more accurate focus detection can be performed. The determination of the reliability, that is, the determination of the coincidence between a pair of images is usually performed by performing an arithmetic process for superposing image signals showing these images and comparing the coincidence with a predetermined threshold value. ing.

[0006]

By the way, in the conventional method for determining the reliability of the image formed by the focus detecting optical system as described above, the determination condition is changed depending on the size of the defocus amount. Has not been done.

However, according to the study by the present inventor,
In the large defocus state, the deviation from the in-focus state is large (the focus deviation is large), so that the common part of the object that is guided and imaged on the pair of sensors used for reliability calculation is the small defocus state. There are few things compared to the focus state,
Since the rise and fall times of these images used for calculation become long, there is a tendency that the reliability is apparently increased when the reliability determination is performed using the image signal obtained from such images. Therefore, if the reliability of the image is determined without distinction from the case of the small defocus state and the lens is driven for focusing, as a result, the focus adjustment is performed with uncertain accuracy in the large defocus state. It will be decided.

The present invention has been made to solve such a problem, and it is said that there is a difference in the result of conventional image reliability judgment between the large defocus state and the small defocus state. An object of the present invention is to provide a focus detection device that eliminates defects and performs stable and accurate reliability determination without being affected by the degree of defocus amount, and that is capable of avoiding lens driving that is accurate and wasteful. And

[0009]

In order to achieve the above object, the inventor of the present invention has proposed a first image in which the relative positional relationship between the images changes in accordance with the focus state of an optical system for forming an image of a subject. And a focus detection optical system for forming a second image, image signal output means for outputting respective image signals photoelectrically converted corresponding to these images, and the above-mentioned first based on these image signals. Calculation means for obtaining the amount of deviation between the image and the second image and for calculating the defocus amount of the focus detection optical system, the first image and the second image
Detection by means of image reliability determination means for determining the reliability of the image formed by the focus detection optical system by comparing the mutual agreement of the image signals representing the images of 1) with a predetermined threshold value. In the apparatus, means for varying the threshold value in the image reliability determining means according to the magnitude of the defocus amount, for example, when the defocus amount is larger than a predetermined constant value, the threshold value is set to a small value and A focus detection device having a structure characterized in that weighting means for increasing the threshold value when the amount is less than the above-mentioned fixed value is provided is completed.

As the weighting means, for example, a predetermined threshold value is multiplied by a coefficient of 1 in the large defocus state depending on the size of the defocus amount, and a smaller coefficient (eg, 2 in the small defocus state). It is possible to use a coefficient means which multiplies a value such as / 3), but is not particularly limited to this.

According to the focus detecting apparatus having the image reliability determining function as described above, it is possible to perform a stable and accurate focus adjusting operation without waste even in a large defocus state.

[0012]

EXAMPLES Examples of the present invention will be described.

FIG. 1 is a circuit diagram showing an embodiment of a camera provided with an automatic focusing device according to the present invention.

In the figure, PRS is a camera controller,
For example, a CPU (central processing unit), RO (not shown)
It is a one-chip microcomputer (hereinafter referred to as a microcomputer) having M, RAM, and A / D conversion functions. PRS
Performs a series of camera operations such as an automatic exposure control function, an automatic focus adjustment function, and film winding / rewinding in accordance with the camera sequence program stored in the ROM. Therefore, PRS is a signal for communication SO, S
I, SCLK, communication selection signal CLCM, CSSDR, CD
The DR is used to communicate with the peripheral circuits in the camera body and the control device in the lens to control the operation of each circuit and the lens.

SO is a data signal output from PRS,
SI is a data signal input to the PRS, and SCLK is a synchronous clock of the signals SO and SI.

LCM is a lens communication buffer circuit,
When the camera is operating, power is supplied to the lens power supply terminal VL, and the selection signal CLCM from the PRS is at a high potential level (hereinafter abbreviated as “H” and a low potential level as “L”). Sometimes it serves as a communication buffer between the camera and the lens.

PRS sets CLCM to "H", and SCL
When certain data is sent from SO in synchronization with K, LC
M is SCLK, S via the camera / lens communication contact
The respective buffer signals LCK and DCL of O are output to the lens. At the same time, the buffer signal of the signal DLC from the lens is output to SI, and PRS synchronizes with SCLK and SI.
Enter the lens data from.

SDR is a drive circuit of a line sensor device SNS for focus detection which is composed of a CCD or the like, and has a signal C
Selected when SDR is "H", SO, SI, SCL
Controlled from PRS with K. The signal CK is a clock for generating CCD driving clocks φ1 and φ2, and the signal INTEND indicates that the accumulation operation is finished.
This signal informs S.

The output signals OS of the SNS are clocks φ1 and φ.
These are time-series image signals synchronized with 2, and are amplified by the amplifier circuit in the SDR and then output to the PRS as AOS.
The PRS inputs AOS from an analog input terminal and sequentially stores it as a digital signal by a built-in A / D conversion function at a predetermined address of a RAM (not shown) in synchronization with CK.

SAGC which is also an output signal of SNS
Is an AGC (automatic gain control: Au) not shown in the SNS.
The output of the to gain control sensor is input to the SDR and used for the SNS storage control.

The SPC is an exposure control (AE) photometric sensor that receives light from a subject through a taking lens.
The output SSPC is input to the analog input terminal of the PRS, A / D converted, and then used for automatic exposure control according to a predetermined program.

DDR is a switch detection and display circuit, which is selected when the signal CDRD is "H", and SO,
It is controlled from PRS using SI and SCLK. That is,
The display of the display member DSP of the camera is switched based on the data sent from the PRS, and the PRS is informed of the on / off state of various operation members of the camera by communication.

SW1 and SW2 are switches interlocked with a release button (not shown), and SW1 is turned on when the release button is pressed in the first step, and SW2 is turned on when the release button is pressed in the second step. The PRS performs photometry and automatic focus adjustment when SW1 is on, and exposure control and film winding are performed when SW2 is on.

MTR1 is a film feeding motor, MTR2 is a motor for mirror up / down and shutter spring charging, and forward / reverse control is performed by respective drive circuits MDR1 and MDR2. PRS to MDR1,
Signals MIF, MIR, M2 input to MDR2
F and M2R are signals for motor control.

MG1 and MG2 are shutter front and rear curtain running start magnets, and signals SMG1, SMG2, respectively.
The amplification transistors TR1 and TR2 are energized, and shutter control is performed by PRS.

The ALED is an auxiliary light source for illuminating the subject when it is determined that the luminance is low and the contrast of the subject is low, and the signal at the output terminal SAL of the PRS is "H".
Driving the drive transistor ATR by
Auxiliary light is emitted through the optical system ALNS. It should be noted that the light projection is pattern light projection. The ALED is composed of, for example, a light emitting diode.

The switch detection and display circuit DD
Since R, motor drive circuits MDR1, MDR2, and shutter control are not directly related to the present invention, detailed description thereof will be omitted. A signal DCL input to the in-lens control circuit LPRS in synchronization with LCK is data of a command from the camera to the lens LNS, and the operation of the lens in response to the command is predetermined. The LPRS analyzes the command according to a predetermined procedure, and performs focus adjustment and aperture control operations and output D.
Output from LC the operating status of each part of the lens (driving status of focus adjusting optical system, driving status of diaphragm, etc.) and various parameters (open F number, focal length, defocus amount vs. coefficient of moving amount of focus adjusting optical system, etc.) I do.

In the embodiment, an example of a single lens for the entire extension is shown. When a focus adjustment command is sent from the camera, the focus adjustment motor LMTR is driven in accordance with the drive amount and direction sent at the same time. Driven by the signals LMF and LMR, the optical system is moved in the optical axis direction to perform focus adjustment. The movement amount of the optical system is monitored by the pulse signal SENC of the encoder circuit ENC and is counted by the counter in the LPRS. When the predetermined movement is completed, the LPRS itself sets the signals LMF and LMR to “L” and the motor. Brakes the LMTR.

Therefore, once the focus adjustment command is sent from the camera, the camera control device PRS does not need to be involved in lens driving at all until the lens driving is completed. In addition, the contents of the counter can be sent to the camera when requested by the camera.

When an aperture control command is sent from the camera, a stepping motor DMTR known for driving the aperture is driven according to the number of aperture steps sent at the same time. Since the stepping motor can be open-controlled, it does not need an encoder for monitoring the operation.

Next, the operation performed by the circuit of FIG. 1 will be described with reference to FIGS. First, when a power switch (not shown) is operated, power is supplied to the microcomputer PRS, and the microcomputer PRS executes the storage program.

FIG. 2 is a flow chart showing the outline of the entire program of the operation performed in the above-mentioned camera. When the execution of the above-mentioned storage program is started, first, in step (002), the state of the switch SW1 that is turned on by the first-step operation of the shutter button (not shown) is detected, and when the switch SW1 is off, step (002) In 005), all the flags inside the microcomputer PRS are set to "0", and all variables described later are set to "0".

The steps (002) and (005) are repeatedly executed until the switch SW1 is turned on. When the switch SW1 is turned on, the process proceeds to step (003).

Step (003) is a step for executing a subroutine of AE control. In this AE control subroutine, a series of controls such as photometric calculation processing, exposure control, shutter charge after exposure, film winding and the like are performed. The above AE control subroutine is not directly related to the present invention, so a detailed description thereof will be omitted, but the outline thereof is as follows. That is, while the switch SW1 continues to be turned on, photometric and exposure control calculations are performed each time this AE subroutine is executed, and during this process, the shutter release button is operated in the second step and the switch SW2 is turned on. Release operation is performed by interrupt processing,
The aperture or shutter time is controlled according to the exposure amount obtained by the exposure control calculation, and the shutter charging and film feeding operations are performed after the exposure is completed.

If it is assumed that the shutter button is operated in the first stage, this flow is step (00
After shifting to 3) and performing photometry and arithmetic processing, the routine proceeds to step (004). When the AF operation mode is the so-called “servo mode” or “manual mode”, the interrupt operation described above is permitted in the state where the first photometry is completed after the power switch is turned on, and the AF operation mode is the “one-shot mode”. When, the focus is detected in the state where the focus is detected. That is, in the "servo mode" or the "manual mode", the release operation can be performed at any time regardless of the focus adjustment operation, and in the "one-shot mode", the release is possible only after detecting the focus. In this example, the AF operation mode is selected by a mode switch (not shown). In the "one-shot mode", once the focus is achieved, the AF operation is not performed again until the switch SW1 is turned off, while in the "servo mode", the AF operation is always performed. I do.

At the step (004) of AF control, the focus state of the taking lens is detected and the "one-shot mode" is selected.
In the "servo mode", the photographing lens is driven so as to be in focus, and in the "manual mode", only the in-focus or out-of-focus display is performed. Further, in the manual mode, AF control by auxiliary light is not performed.

In the above flow, the AE control in step (003) and the AF control in step (004) are alternately repeated as long as the switch SW1 is on.

At step (004), the AF control subroutine shown in FIG. 3 is called and executed. First, in step (102), it is determined whether or not it is the first AF control immediately after the shooting of one frame by the shutter release operation is completed (that is, the film feeding is performed by the AE control subroutine performed immediately before). Judge whether or not),
It is determined whether or not the currently designated mode is the "continuous shooting mode". In this step (102), when it is determined that the current AF control is performed for the first time after the end of shooting, it is determined that the AF operation is in the continuous shooting mode, and the process shifts to step (103) to execute the microcomputer PRS. All flags in the above are set to "0", and all variables are set to "0". As a result, the AF control situation in the past is initialized during the "continuous shooting mode", and the next step (104)
The continuous shooting mode flag FAF is set to 1 to store that the AF control is in continuous shooting, and then the process proceeds to step (109). On the other hand, if the current AF control is not the first AF control immediately after the end of shooting, the process proceeds to step (105), and the set state of the flag PRMV is detected. This flag PRMV is a "lens drive flag".
If the lens has not been driven in the previous AF control, the flag PRMV remains set to "0", and therefore the process proceeds to step (109) and the AF control is performed again. On the contrary, when the lens is driven last time, it is determined in the next step (106) whether or not the lens is stopped.

In this step (106), as described in FIG. 1, the microcomputer PRS causes the lens LN to pass through the lens communication buffer circuit LCM and the lens internal control circuit LPRS.
It communicates with S and checks the state of the lens by receiving the monitor signal. If the lens is not stopped at this time, the AF control subroutine is ended because the lens is being driven. That is, when the lens is in the driving state, the determination as to whether or not the lens has stopped is continued in step (106). This is because in the present example, the AF control is not performed while the lens is being driven, and the AF control is performed only when the lens is stopped, and the next step ( 108) in PRM
V is set to "0" and the process proceeds to step (109).

At step (109), it is determined whether the mode is autofocus or not. If the mode is the manual mode ("manual" is selected by a mode switch (not shown)), the process proceeds to step (110). This step (11
In (0), the image signal input subroutine, the focus detection subroutine in step (111), and the display subroutine in step (112) are executed, and then the AF control subroutine is terminated in step (113) (each subroutine). See below).

On the other hand, if the manual mode is not set in step (109), it is determined in the next step (114) whether the "one-shot mode" is set. If it is the one-shot mode, the process proceeds to step (115), and the state of the focus flag JF is checked to determine whether or not it is the "previous focus state". If the AF control mode is the "one-shot mode" and "focusing last time", the AF control is ended in step (116). That is, in the one-shot mode, once the focus is achieved, the switch SW1 is turned off and the switch SW1 is turned on again.
No new AF control is executed until 1 is turned on. If the previous focus is not obtained at the step (115), the process proceeds to the next step (117). The reason why the above-described processing is performed in the "one-shot mode" is that the AF mode is determined to be the one-shot or the servo described in the step of the AF control.

Next, the routine proceeds to step (117) to execute the "image signal input" subroutine. The outline of the control performed in this "image signal input" subroutine is as follows. That is, first, the microcomputer PRS is the CSD
R is set to "H" and the SO signal is transmitted to the drive circuit SDR. The SO signal at this time is an accumulation start command, and in response to this command, the drive circuit SDR transmits the CLR signal to the line sensor device SNS to clear the image accumulation signal of the CCD line sensor, and then causes the accumulation operation for the image to be performed. On the other hand, the CDD line sensor CC of the line sensor device SNS
The image light fluxes that have entered through D ₁ and CCD ₂ through the taking lens respectively enter the sensors CCD ₁ and CCD ₂
The upper image position is given depending on the focus condition. That is, if the subject is in focus, the same image pattern should be projected at the same position on each of the sensors CCD ₁ and CCD ₂ , and if the detected focus state is the front focus state or the rear focus state. For example, the image patterns on CCD ₁ and CCD ₂ are projected at positions symmetrically displaced in accordance with the focus shift direction and the shift amount. Therefore, by detecting the position shift amount and direction between the image patterns on the CCD ₁ and CCD ₂ , the focus shift direction and the shift amount can be detected.

As described above, the image signal accumulated in the CCD ₁ and CCD ₂ for a predetermined time showing the image pattern projected at the position corresponding to the focus state is output from the drive circuit SDR as the α signal SH and the clock φ1, φ2 allows sensor device SNS
Supplied from The accumulation time of the image pattern is S
Output SAG of storage control sensor AGC (not shown) in NS
It may be determined based on C, which will be described later.

As described above, when the signal SH and the clocks φ1 and φ2 are supplied to the sensor device SNS, the image signal accumulated in each pixel of each sensor CCD ₁ and CCD ₂ is output from the output end of the sensor device SNS. A signal as a content is sequentially output as an OS in time series, and is sequentially input to an analog input terminal of the microcomputer PRS as a signal AOS amplified by an amplifier circuit in the drive circuit SDR. Then, the microcomputer PRS synchronizes with the clock CK output by itself and synchronizes with the above signal A.
The OS is A / D converted by the internal A / D conversion function, and digital image signals after that are sequentially stored in a predetermined address of the RAM.

With the above operation, the sensors CCD ₁ and CCD ₂
The image signal for each sensor corresponding to the above image pattern is stored as a digital value in the RAM in the microcomputer PRS.

Details of this "image signal input" subroutine are shown in FIG. That is, first step (20
When the image signal input subroutine is called in step (117) of the AF control routine in step 1), the flag AUXMOD (a flag set to 1 in the previous auxiliary light mode) is set in step (202). A
It is determined by the F control whether or not it is the "assist light mode". If it is not the auxiliary light mode, the process proceeds to step (207), and if it is the auxiliary light mode, the process proceeds to step (203).

In step (203), the flag AUXN is set.
It is determined by OEM (a flag that is set to 1 when light is emitted a predetermined number of times) whether or not the predetermined number of times of emission of the auxiliary light has been completed. If the predetermined number of times of light projection is completed, the process proceeds to step (207), but if not completed, the process proceeds to step (204).

In step (204), the flag AUXJ is set.
It is determined by F (a flag that is set to "1" at the time of focusing in the auxiliary light mode) whether or not the previous focusing was performed under the projection of the auxiliary light.
If in focus, step (207) without projecting auxiliary light
Move to, and if not in focus, next step (205)
Go to the auxiliary light on.

This flag AUXJF will be described in detail in the subroutine "focus detection". This flag AUXJF is a judgment as to whether or not focusing is performed under the projection of the auxiliary light. When AUXJF is "1", Even if the flag AUXOD is in the fill light mode and the number of times the flag AUXNOEM is projected is less than or equal to a predetermined number, the fill light is not projected.

In step (205), the auxiliary light is turned on. That is, by setting the output signal of the output terminal SAL of the microcomputer PRS to "1", the auxiliary light source AL
A current is passed through the transistor ATR to the ED and the optical system A
Light emission of auxiliary light is started from the LNS.

In step (206), the auxiliary light projection counter AUXCNT is incremented by one.

In step (207), the focus detection line sensor SNS device is caused to start accumulating an optical image. The microcomputer PRS sends a "storage start command" to the sensor drive circuit SDR.
Is sent, the sensor drive circuit SDR causes the sensor SN
The clear signal CLR to the photoelectric conversion element portion of S is set to "0" to start the charge accumulation.

In step (208), the accumulation time counter INTCNT is initially set to an initial value 0.

In step (209), the 1 ms timer counter of the accumulation time timer is reset. The counter uses the timer function of the microcomputer PRS.

In step (210), it is determined whether or not the sensor has completed the accumulation. That is, it is checked whether or not the signal INTEND from the sensor drive circuit SDR is "1". The sensor drive circuit SDR sets the signal INTEND to "0" at the same time when the line sensor SNS starts to store the sensor A for storage control in the sensor SNS.
By monitoring the output SAGC signal from the GC,
When the signal reaches a predetermined level, the sensor drive circuit SDR
The signal INTEND from the sensor drive circuit SDR is set to "1", and at the same time, the charge transfer signal SH from the sensor drive circuit SDR is set to "H" for a predetermined time so that the charge of the photoelectric conversion element section in the sensor SNS is transferred to the CCD section. To control.

If the accumulation is completed, the process proceeds to step (215). If the accumulation is not completed yet, the process proceeds to step (211).

In step (211), it is determined whether or not the 1 ms timer has reached the accumulation time of 1 ms. 1 ms
If it has not reached, the process returns to step (210) again, and if it has already reached 1 ms, the accumulation time counter INTCNT is incremented by 1 at the next step (212).

In step (213), the counter INT
The content of CNT is compared with a predetermined number MAXINT. MAX
INT is the longest accumulation time expressed in 1ms units,
If INTCNT is less than MAXINT, the process returns to step (209) to wait for the end of accumulation again. If INTCNT coincides with the longest accumulation time MAXINT, the accumulation is forcibly ended in the next step (214).

This end of storage is completed by sending a "storage end command" from the microcomputer PRS to the sensor drive circuit SDR. The sensor drive circuit SDR is a microcomputer P
When the "storage end command" is sent from RS, the charge transfer signal SH is set to "H" for a predetermined time to transfer the charges of the photoelectric conversion unit to the CCD unit.

In step (215), the flag AUXMO is set.
In D, it is determined whether or not the auxiliary light mode is set. The flag AUXM
If the OD is the auxiliary light mode, the emission of the auxiliary light is turned off in step (216). That is, the output terminal S of the microcomputer PRS
The AL signal is set to "0" to deactivate the light source ALED. Regarding step (216), step (215) may be omitted because there is no adverse effect even if this step is executed when the mode is not the fill light mode.

In step (217), the accumulation time counter INTCNT is compared with a predetermined constant AUXINT. This AUXINT is a low-brightness accumulation time expressed in accordance with the accumulation time, and INTCNT ≧ AUXINT
If it is, the low brightness flag LLFL is determined in step (219).
G is set to “1” and INTCNT is AUXIN
When it is less than T, the low luminance flag LLFLG is set to "0" in step (218).

In step (220), the A / D conversion of the signal AOS obtained by amplifying the image signal OS of the line sensor SNS in the drive circuit SDR and the R / D conversion in the microcomputer PRS are performed.
The operation of storing in the predetermined address of AM is performed.

In step (221), the contrast information of the image signal stored in the RAM within the predetermined address in step (220) is extracted and stored in a counter (not shown) as contrast information CON.

In step (222), the content of the contrast information CON is compared with a predetermined number LCNUM, and the CO
If N is less than LCNUM, it is determined that the current image signal is in the low contrast state where focus detection is impossible, and the process proceeds to step (223), and the flag LCON indicating the low contrast state is set to "1".

On the other hand, if not, step (224).
Then, the content of the contrast information CON is compared with four times the predetermined number LCNUM. This is to determine whether or not the contrast of the image signal this time is relatively low. In this embodiment, four times the low contrast judgment threshold value is used as a comparison value, but the value is not particularly limited to this. The CON is the predetermined number LC
If it is less than four times NUM, the process proceeds to step (225), and the flag NLCON indicating the contrast state that recommends the use of the auxiliary light source is set to "1", and step (227).
Move to. This flag is also set when the CON is less than the LCNUM.

In step (226), it is assumed that the contrast of the image signal of this time is an amount sufficient for accurate focus detection, the flags LCON and NLCON are set to "0", respectively, and step (227) is executed. Then, the procedure returns to the main routine, that is, step (118).

The operation after the execution of the image signal input subroutine has been completed will be described by returning to the main routine. In step (118), it is determined by the flag FAF whether or not the continuous photographing mode is set.

When the flag FAF is 1, that is, in the continuous photographing mode, the process proceeds to the next step (123), and when it is 0, the process proceeds to step (119). The step (119) and subsequent steps show the flow of auxiliary light. When the flag FAF is 1, the process proceeds to step (123),
No processing related to auxiliary light is performed for continuous shooting.

In step (119), the flag AUXMO is set.
At D, it is determined whether or not the auxiliary light mode is set.

If it is the auxiliary light mode in the previous AF control, the flag AUXMOD is 1, and the routine proceeds to step (130). When the flag AUXMOD is 0, step (1
The process proceeds to step 20), and it is determined whether the brightness is low or not by the flag LLFLG. This flag LLFLG is set in steps (218) and (219) in the above-mentioned subroutine "image signal input".
The flag is set in step 1. When the value is 1, that is, the brightness is low, the process proceeds to step (121).

In step (121), it is determined whether or not the focus can be detected by the flag LCON set in step (223) in the above-mentioned "image signal input" subroutine. If the flag LCON is 1, the state is low brightness and low contrast this time, so the process proceeds to step (130) to perform the focus detection operation using the auxiliary light source. If the flag LCON is 0, the auxiliary light source is currently selected in step (122). It is similarly determined whether or not the use of the contrast condition is better by the flag NLCON set in step (225) in the "image signal input" subroutine. If NLCON is 1, this time, the brightness is low and the contrast is relatively low. Therefore, the process proceeds to step (130) to perform the focus detection operation using the auxiliary light source. If 0, step (12
Go to 3).

Steps (130) to (137) are the auxiliary light control flow in the state where the brightness is low and the contrast is low.

First, in step (130), it is determined whether or not the auxiliary light unit is attached to the camera. Here, the determination is made by detecting the state of a switch (not shown) that is turned on when the auxiliary light unit is attached to the camera.

When the auxiliary light unit is not attached to the camera, the process proceeds to step (123) and the AUX
Set "O" to MOD to cancel the auxiliary light mode.

When the auxiliary light unit is mounted, in step (131), AUXMOD is set to "1" to set the auxiliary light mode. In this example, the step (131) is executed even when the flag AUXMOD is in the auxiliary light mode in the step (119), but this is done for the sake of the progress of the program sequence of this example.

In step (132), it is determined whether the auxiliary light projection counter AUXCNT is set to "0". This counter AUXCNT is a variable counted in step (206) in the above-mentioned "image signal input" subroutine, and counts one when the image signal is accumulated in the photoelectric conversion element under the projection of the auxiliary light. Up. The counter AUXCNT is a switch SW1.
When is off, it is cleared in step (005) of the flowchart shown in FIG. 2, and in AF control in the continuous shooting mode, it is cleared in step (103) shown in the flowchart of FIG. Therefore, the counter AUXCNT = "0" means that the auxiliary light mode is the first time in the sequence progress, and when the first auxiliary light mode is the current time, the step (13
Proceed to 3) to end the AF control by the image signal input this time. That is, the image signal input in the step (117) is discarded, and the next AF control is performed under the auxiliary light projection.
A new image signal is input. Therefore, in this case, it can be considered that the image signal input in step (117) was used only for determining the brightness.

On the other hand, if the counter AUXCNT is not "0", the auxiliary light mode has already been entered.
In step 34), the counter AUXCNT is set to a predetermined constant AUXNU.
Compare M. This AUXNUM is the light emission limit number of auxiliary light, and this counter AUXCNT is the light emission limit number A.
When it is less than UXNUM, the process proceeds to step (124) and the "focus detection" subroutine is executed. When the counter AUXCNT is AUXCNT ≧ AUXNUM (however, in the actual sequence, AUXCNT> A
No UXNUM occurs) Go to step (135).

At step (135), the flag AUXN is set.
The OEM determines whether or not the number of times the auxiliary light has been projected is completed. When it is determined by the flag AUXNOEM that the number of times of projecting the auxiliary light has already been projected by the predetermined constant AUXNUM, the number of times of projecting the auxiliary light is completed, and no more of the auxiliary light is projected. That is, even when the flag AUXMOD is 1 and the auxiliary light mode is set in the above-mentioned "image signal input" subroutine, when the flag AUXNOEM is 1 and the auxiliary light projection number is completed, the auxiliary light is no longer stored when the image signal is accumulated. Do not project light.

In step (135), the flag AUXN is set.
If the number of times of light emission is completed by OEM, the process proceeds to step (136).

At step (136), the flag LLFLG is set.
Determines whether the brightness is low. This flag LLFLG is a low brightness flag set by the "image signal input" subroutine as described above, and the fact that the number of times of light projection is completed by the flag AUXNOEM and the brightness is judged by the flag LLFLG means " In the auxiliary light mode, the brightness is determined in a state in which the auxiliary light has already been projected a predetermined number of times and is no longer projected. Here, when the flag LLFLG is not low in brightness, the process proceeds to step (123), the flag AUXMOD is set to "0" to cancel the auxiliary light mode, and the process proceeds to the next step (124). Here the flag LLF
Determining that the LG does not have low brightness means that the auxiliary light mode is canceled and the normal mode is resumed when the brightness increases in the above-described state, and the AF control in the normal mode is restarted.

When the process shifts to the step (123), when the flag FAF is 1 in the step (118) and it is determined that the continuous photographing is being performed (no auxiliary light is projected), and the flag AUXMOD is set in the step (119). This is when it is determined in the next step (120) that the flag LLFLG is "0" and the brightness is not low when the mode is not the fill light mode. That is AF
If the control is not in the auxiliary light mode and the subject brightness is not low, the process proceeds to step (123).

On the other hand, when it is determined that the flag LLFLG is 1 and the brightness is low, the process directly shifts to the "focus detection" of the next step (124).

In the above flow, step (135)
Thus, only when the flag AUXNOEM indicates that the number of times of projecting the auxiliary light is completed, the flag LLFLG determines whether or not the brightness is low.
The reason for this is that when the brightness of the object is detected under the projection of the auxiliary light, the object brightness is increased due to the influence of the auxiliary light, and the true object brightness is not judged. Therefore, the flag AUXNOEM is set to "0". This is because it is not determined whether or not the brightness is low when the number of times the auxiliary light is projected is less than or equal to a predetermined value.

Next, in step (124), the "focus detection" subroutine is called and executed. The contents of this "focus detection" subroutine will be described with reference to FIG.

First, at step (303), it is determined by a flag AUXJF whether or not focusing has been performed under the projection of the auxiliary light in the past. When the subject is in focus under the projection of the auxiliary light, the process proceeds to step (304).

At step (304), the flag LLFL is set.
The G state is detected, and it is determined whether the luminance in the state where the auxiliary light is not projected is low. Here the flag LLFL
When G is 1 and the brightness is low, the next step (30
The AF control is ended in 5). On the other hand, the flag LLFLG
When the brightness is 0 and the brightness is not low, the brightness of the subject has increased, and therefore the process proceeds to step (306), and the auxiliary light mode flag AUXMOD and the projection count completion flag AUXNOE.
M, clear the focus flag AUXJF to "0", and
The auxiliary light counter flag AUXCNT is cleared to "0".

In step (307), the defocus amount DEF of the photographing lens is calculated based on the image signal from the "image signal input" subroutine described above.

Here, on the basis of the digital values corresponding to the image patterns on the sensors CCD ₁ and CCD ₂ obtained in the above-mentioned "image signal input" subroutine, the shift amount up to focusing and the shift direction are defocused amounts. Calculate as DEF.
The specific method of calculating the defocus amount is not directly related to the purpose of the present application, and therefore its detailed description is omitted, but as described above, the degree of coincidence between the image patterns on the sensors CCD ₁ and CCD ₂ is in focus. Since it is determined by the state, the deviation amount and the deviation direction from the focused state, that is, the defocus amount DEF is obtained by comparing the digital value of each sensor corresponding to the pattern and obtaining the degree of coincidence of the data. It is a thing.

Next, in step (308), the calculation is performed based on the calculation result of the defocus amount DEF described above.
The "reliability determination" subroutine, which is one of the features of this example, is executed.

The contents of this "reliability determination" subroutine will be described with reference to FIG. First, in step (501),
Reliability flag FNC (1,2,3) F indicating the degree of coincidence of images
Clear each LG to "0" and then proceed to step (50
In 2), the image reliability is calculated. Note that here, the reliability flag FNC There are three FLGs of 1, 2, and 3 because in this example, the reliability is determined by using three threshold values as the determination criteria and performing four-step ranking. is there. That is, the reliability flag FNC (1) FLG
Is a flag that is given as a result when the judgment is made with the largest threshold value, and that it is “0” means that
It shows that the detected image has the highest reliability. Further, the reliability flag FNC (2) FLG is a flag given as a result of judgment with an intermediate threshold value, and the reliability flag FNC (3) FLG is given as a result of judgment with the smallest threshold value. It is a flag that is set. The image reliability calculation method performed in the above step (502) in this example is performed as follows. First, by utilizing the fact that the relative position displacement amount of two images and the defocus amount are proportional to each other, the image information is converted into the defocus amount DEF and the photographing lens drive amount up to the in-focus point is calculated as described above. is there.

On the other hand, the degree of coincidence of the image patterns (that is, the degree of similarity between the pair of images formed by the focus detection optical system and the degree of coincidence) is determined as follows. That is, first, the following formula (1) is calculated from the intensity data of each of the first image (image A in FIG. 5) and the second image (image B in FIG. 5) sampled by each CCD.

[0092]

[Equation 1]

As shown in FIG. 5, the value of U _{o in} this equation (1) is smaller as the coincidence between the two images is better (the shaded area in FIG. 5 is the U. value), and is the minimum when focusing. Becomes

Here, the calculation in which the sampling data of the image A is shifted (shifted) by p pixels relative to the image B with respect to the image B is represented by U in the above equation (1). Conducted in the same manner as obtaining a value to it a U _p value of the following equation (2).

[0095]

[Equation 2]

At the shift amount at which the images best match, the above U
_{Since p} should be the minimum, such a shift amount p
The relative position displacement amount of the two images can be obtained by locating (see FIG. 6).

However, the change of the defocus amount corresponding to the shift of one pixel is about 1 mm, and it cannot be used for focus detection as it is. Therefore, the displacement amount (h) of the image finer than the minimum size of one pixel is calculated by interpolation from the vicinity of the minimum U _p value in FIG.

Then, based on the image displacement amount (h) obtained by this calculation, the reliability data FNCLVL of the detected image is obtained.
Is calculated by the following equation (3), for example.

FNCLVL = h × 32 / MAXDER (3) In this way, the reliability data FNCLVL of the detected image is obtained. Next, the process proceeds to step (503), and the defocus amount obtained by the above-described processing. A constant value LRG that determines in advance whether DEF is a large defocus amount Compared with PRDCT, DEF ≧ LRG In the case of PRDCT, the process proceeds to step (504) to set the LRGPRDF flag to "1", while DEF <LRG In the case of PRDCT, that is, as it is (that is, the LRGPRDF flag is "0"), the process proceeds to step (505).

Then, in step (505), the condition setting for determining the reliability of the detected image based on the reliability data of the detected image, which is a characteristic important point of the present example, is determined according to the above determination result. Perform processing (weighting).
Depending on the condition setting performed here, the flag LRGPRD
When F is "0" (that is, when the detected defocus amount is small), the reliability threshold value FNC The weighting factor A of NOR is set to "1", and conversely when the flag LRGPRDF is "1" (that is, when the detected defocus amount is large), the reliability threshold value (FNC The NOR weighting coefficient A is saved as 2/3.

Next, in step (506), reliability comparison 1 is performed. That is, the reliability data FNCLVL of the detected image obtained in the step (502) is compared with the reliability threshold value A × FNC (1) NOR multiplied by the above coefficient to obtain F
When NCLVL ≦ A × FNC (1) NOR, the reliability flag FNC (1) FLG is set to “0” because the reliability of the pair of detected images is sufficiently high (the best rank) and the reliability is determined to be high. Proceed to step (512). On the contrary, FNC
When LVL> A × FNC (1) NOR, the reliability cannot be said to be sufficiently high, so the reliability flag FNC (1) FLG is set to “1” in step (507), and step (508) Proceed to.

In steps (508) to (511), reliability comparisons 2 and 3 are performed in the same manner as the reliability comparison 1 described above, the reliability data FNCLVL of the detected image is ranked, and the reliability flag FNC is determined. (2) FLG and reliability flag FNC (3) FLG is set to "1" (that is, the corresponding reliability is not satisfied) or "0" (that is, the corresponding reliability is satisfied).

As the coefficient A used for the calculation in the above reliability judgment, "2/3" is used in this example when the detected defocus amount is large as described above, and the detected defocus amount is If it is small, "1" is used. However, it goes without saying that the value of the coefficient "2/3" is not necessarily limited to this value depending on the characteristics of the focus detection optical system used and the like.

After the above reliability determination processing, the routine proceeds to step (512) to perform contrast determination of the detected image, and when it is determined that the contrast of the subject is sufficient, FIG. When it is determined that the contrast of the subject is insufficient, the reliability flags F described above in step (513) are returned to step (309).
After setting NC (1,2,3) FLG to all "1" (that is, the corresponding reliability is not satisfied) and setting low contrast L L FLG to "1", the step of FIG. Return to (309).

With the above, the processing of the "reliability determination" subroutine of the detected image is completed.

Next, returning to the "focus detection" subroutine of FIG. 4B, further explanation will be given.

When the process returns from the "reliability determination" subroutine, first, the defocus amount DEF described above in step (309) and JF which is a predetermined fixed focusing allowance value.
Compare FLD.

Here, when DEF> JFFLD,
Proceeding to step (310), the focus flag JF is set to "0", the focus detection subroutine is terminated, and the routine proceeds to step (124.1) in FIG.

On the other hand, if it is determined that JFFLD> DEF, then the focus flag JF in step (312) is set to "1", and the flow advances to step (313).

In step (313), the flag AUXMO is set.
At D, it is determined whether or not the mode is the fill light mode.

If the flag AUXMOD is the auxiliary light mode, the flag AUXJF is set to "1" by the focusing under the auxiliary light projection in step (314), and "focus detection" is performed.
End the subroutine and go to the next step (124.1)
Shift to, judge the flag LCON, LCON is "1"
If it is, it is determined that focus detection cannot be performed, and the process proceeds to step (124.2) to execute the subroutine "NG display". Since this sub-routine has a function of displaying that focus detection is impossible on a display member, detailed description thereof will be omitted. After executing "NG display", the step (124.
The process moves to 3) and the "AF control" subroutine is returned.

If LCON is "0" in step (124.1), it is determined that focus detection is possible, and the flow shifts to the subroutine "display" in step (125) in FIG.
The “display” of step (125) is a function of merely displaying the in-focus state of the AF control described above by the display member, and therefore the description thereof will be omitted.

At step (126), it is determined by the flag JF whether or not focus is achieved.

The flag JF is a flag stored in steps (310) and (312) in the above-mentioned "focus detection" subroutine, and when this flag JF determines that it is in focus (that is, "1"). That is, when the focus is achieved, the AF control is ended in step (129), and when it is determined that the focus is not achieved (that is, "0"), the step (127) is performed.
Then, the process proceeds to the lens drive subroutine.

In the lens driving subroutine, the microcomputer PR
No signal CLCM is set to "1" at S and buffer circuit LCM
Is specified. In addition, the lens drive amount is set to the SO signal and the circuit L
Control circuit LP transmitted to CM and SO signal as DCL signal
Tell RS. The circuit LPRS sets the signal LMF or LMR to H according to the lens driving amount, and the motor LMTR
Is rotated in a predetermined direction to move the taking lens LNS in the optical axis direction. The moving amount of this lens is motorized by the encoder ENC, outputs a signal SENC corresponding to the moving amount, and is compared with the signal SENC and the signal indicating the defocus or the like transmitted to the circuit LPRS. The signals LMF and LMR are set to L, the rotation of the motor LMTR is stopped, and the lens driving is ended.

With the above operation, the lens moves by the amount corresponding to the defocus amount, the lens driving subroutine is ended, the lens driving flag PRMV is set to "1" in step (128), and the lens driving flag PRMV is set in step (129). The AF control ends.

Next, the lens driving subroutine step (401) of FIG. 4D will be described in detail.

In step (402), the "coefficient S of the defocus amount to the extension amount" is input from the taking lens. The defocus amount is the above-mentioned DEF, and the extension amount represents the amount of movement of the lens for focus adjustment in the optical axis direction.

The coefficient S is S = 1 when the photographic lens is a single-lens extension lens, and when the photographic lens is a zoom lens, the coefficient S varies depending on the zooming position.

At step (403), the "feed amount PTH equivalent to one encoder tooth" is input. When the lens for focus adjustment moves, a pulse is generated from the encoder along with this movement, and the feed amount PTH corresponding to one tooth of the encoder represents the amount of movement of the lens per tooth.

In step (404), the moving amount of the focus adjustment lens is adjusted by the previously detected defocus amount DEF, the above-described coefficient S of the defocus amount to the moving amount, and the moving amount PTH corresponding to one encoder tooth. An integer value FP represented by the number of teeth of is calculated by the following formula.

FP = DEF × S / PTH In step (405), the integer value FP obtained in step (404) is sent to the photographing lens as a lens driving amount to instruct lens driving, and in step (406), The lens drive ends.

The operation of the present example configured as described above is summarized as follows.

In step (124), the distance is measured by AF, the defocus amount DEF is calculated, and it is determined in step (503) whether or not the result is large defocus. In this example, the reliability threshold values 1, 2 and 3 are weighted based on the determined result, and the value and the reliability of the detected image FNCLVL
And the reliability flag FNC (1.2,3) FLG
To decide.

Whether the defocused value is used or not is determined depending on the state of the reliability flag determined above.

Therefore, in this example, in the large defocus state as described above, the threshold value used as the criterion for reliability judgment is smaller than the threshold value used as the criterion in the small defocus state. (This is achieved by multiplying by a small coefficient in this example), so it is derived from the above-mentioned tendency (that is, the uncertainty of the detected image is large) that appears in the large defocus state compared to the small defocus state. It provides a camera that can solve the inconvenient problems that occur, can always realize stable and accurate reliability determination that is not affected by the amount of defocus, and that performs automatic focus detection operation and automatic focus adjustment operation without waste. it can.

Defocus was used for the determination of weighting the reliability threshold value.
For example, the magnitude of the image output can be determined and replaced with the determined value.

[0128]

According to the present invention, when the defocus of a pair of detection images detected by the focus detection optical system is determined to be large, the reliability of the detection image is set to be stricter than that in the small defocus state. As a result, it is possible to detect a highly accurate focus state even in the large defocus state, and it is also possible to eliminate the problem that the photographing lens is driven based on uncertain data that is originally unreliable.

[Brief description of drawings]

FIG. 1 is a circuit diagram for explaining an outline of a control system of a camera to which the present invention is applied.

FIG. 2 is a diagram briefly explaining an outline of operations performed by the control circuit of FIG.

FIG. 3 is a flowchart illustrating a main routine of an AF control operation performed in the above embodiment.

FIG. 4A is a flow chart for explaining the operation of an “image signal input” subroutine performed in the above embodiment.

FIG. 5B is a flow chart for explaining the operation of the “focus detection” subroutine performed in the above embodiment.

FIG. 6 (c) is a flowchart explaining the operation of a “reliability determination” subroutine performed in the above embodiment.

FIG. 7D is a flow chart for explaining the operation of the “lens drive” subroutine performed in the above embodiment.

FIG. 8 is a diagram for explaining the matching of image patterns of a pair of images formed on a pair of line sensors.

FIG. 9 is a diagram for explaining the matching of image patterns of a pair of images formed on a pair of line sensors.

[Explanation of symbols]

PRS ... Microcomputer SDR ... SNS drive circuit

─────────────────────────────────────────────────── ───

[Procedure amendment]

[Submission date] July 23, 1993

[Procedure Amendment 1]

[Document name to be amended] Statement

[Name of item to be corrected] Brief explanation of the drawing

[Correction method] Change

[Correction content]

[Brief description of drawings]

FIG. 1 is a circuit diagram for explaining an outline of a control system of a camera to which the present invention is applied.

FIG. 2 is a diagram briefly explaining an outline of operations performed by the control circuit of FIG.

FIG. 3 is a flowchart illustrating a main routine of an AF control operation performed in the above embodiment.

FIG. 4 is a flow chart for explaining the operation of an “image signal input” subroutine performed in the above embodiment.

FIG. 5 is a “focus detection” performed in the above embodiment.
The flowchart figure explaining operation | movement of a subroutine.

FIG. 6 is a flow chart for explaining the operation of a “reliability judgment” subroutine performed in the above embodiment.

FIG. 7 is a flowchart illustrating the operation of a “lens drive” subroutine performed in the above embodiment.

FIG. 8 is a diagram for explaining the matching of image patterns of a pair of images formed on a pair of line sensors.

FIG. 9 is a diagram for explaining the matching of image patterns of a pair of images formed on a pair of line sensors.

[Explanation of symbols] PRS ... Microcomputer SDR ... SNS drive circuit

[Procedure Amendment 2]

[Document name to be corrected] Drawing

[Name of item to be corrected] Fig. 4

[Correction method] Change

[Correction content]

[Figure 4]

[Procedure 3]

[Document name to be corrected] Drawing

[Name of item to be corrected] Figure 5

[Correction method] Change

[Correction content]

[Figure 5]

[Procedure amendment 4]

[Document name to be corrected] Drawing

[Name of item to be corrected] Figure 6

[Correction method] Change

[Correction content]

[Figure 6]

[Procedure Amendment 5]

[Document name to be corrected] Drawing

[Name of item to be corrected] Figure 7

[Correction method] Change

[Correction content]

[Figure 7]

[Procedure Amendment 6]

[Document name to be corrected] Drawing

[Correction target item name] Figure 8

[Correction method] Change

[Correction content]

[Figure 8]

[Procedure Amendment 7]

[Document name to be corrected] Drawing

[Correction target item name] Figure 9

[Correction method] Change

[Correction content]

[Figure 9]

Claims (2)

[Claims] 1. A first image and a second focus detecting optical system for forming an image in which a relative positional relationship between the images changes depending on a focus state of an optical system for forming an image of a subject, and these images. Image signal output means for outputting respective image signals photoelectrically converted corresponding to the first image and the second image based on these image signals.
Of the image difference between the image signals of the first and second images and a calculation means for calculating the defocus amount of the focus detection optical system and a predetermined threshold value. In the focus detection device, the image reliability determination means determines the reliability of the image formed by the focus detection optical system in comparison with A focus detection device comprising means for varying a threshold value. 2. The means for varying the threshold value according to claim 1, wherein the threshold value is set to a small value when the defocus amount is larger than a predetermined fixed value, and the defocus amount is less than the fixed value. In some cases, the focus detection device has a large threshold value.