JP2671206C - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION (Field of application of the invention)   The present invention relates to a focus detection device that detects a focus state from a relative positional relationship between two images of an object.
The present invention relates to an improvement of an optical device including: (Background of the Invention)   Conventionally, as one type of camera focus detection device, the exit pupil of the taking lens is divided.
Observe the relative position displacement of the pair of images formed by the light flux passing through each pupil area
A so-called “image shift type” for determining a focused state is known.   A signal processing method for detecting an image shift amount from an image signal is disclosed in Japanese Patent Application Laid-Open No. 58-1423.
No. 06, U.S. Pat. No. 4,333,007 and the like are disclosed.
Focus detection processing is performed using all or a fixed range of signals output from the sensor.
It was a thing. For this reason, the focus detection range provided in the viewfinder
The distance measurement frame is smaller than the whole area of the sensor, and it is misaligned in the sensor row direction.
The focus cannot be detected even if the subject is within the focusing frame,
There is an inconvenience of detecting the focus of an object outside the frame. In other words, the eyes
If the field of view you are looking at and the field of focus actually detected are different,
Accurate focus detection could not be performed.   Conventionally, in order to deal with the above-mentioned points, the sensor and the distance measurement frame are
The alignment was done one by one. For this reason, it takes a lot of time and work efficiency.
It was bad. Also, for example, you want to change to a different distance measurement frame for some reason
However, in such a case, the above was remarkable. (Object of the invention)   An object of the present invention is to solve the above-mentioned problems and to perform mechanical adjustment of a photoelectric conversion unit.
The focus detection area of the observation means and the focus detection area on the pixels of the photoelectric conversion means
An object of the present invention is to provide an optical device including a focus detection device that can be easily matched.
You. (Features of the invention)   In order to achieve the above object, the present invention relates to a photoelectric conversion means comprising a plurality of pixels,
An optical system for forming an image of the light passing through the image optical system on the photoelectric conversion unit;
Performing a predetermined focus detection process on the photoelectric conversion signal from the means,
A focus detection device including a calculation unit for detecting a focus state of a system, and the imaging optical system
For observing an object through a light source and including an observation means having a focus detection area
The optical device generates an effective photoelectric conversion signal corresponding to the focus detection area.
Storage means for storing an effective pixel range in advance in the adjustment step, and
Match the focus detection area with the focus detection area on the pixel by the effective pixel range
As a result, the photoelectric conversion signal of the effective pixel range stored in the storage means
It is characterized in that focus detection processing is performed. (Example of the invention)   First, the principle of focus detection in this type of apparatus will be described with reference to FIG.
. Field lens FLD with the same optical axis as the imaging lens FLNS to be focused
Is arranged. Two secondary imaging lenses at the rear and symmetrical positions with respect to the optical axis
FCLA and FCLB are arranged. Further, the sensor arrays SAA and SAB are arranged behind it. secondary
Stops DIA and DIB are provided near the imaging lenses FCLA and FCLB. Field lens
FLD has the exit pupil of the photographing lens FLNS almost in the pupil plane of the two secondary imaging lenses FCLA and FCLB.
Form an image. As a result, the light beams incident on the secondary imaging lenses FCLA and FCLB, respectively,
Each secondary image on the exit pupil plane of the shooting lens FLNS Shoot from the non-overlapping areas of equal area corresponding to the lenses FCLA and FCLB.
It will be issued. The aerial image formed near the field lens FLD is secondary
When the image is re-imaged on the sensor rows SAA and SAB by the image lenses FCLA and FCLB,
The two images on the sensor arrays SAA and SAB change their positions based on the displacement of the aerial image position of
And Therefore, if displacement (deviation) of the relative positions of the two images is detected, the photographing lens
You can know the focus state of FLNS.   Signal processing for detecting an image shift amount from image signals output from the sensor arrays SAA and SAB
As the method, JP-A-58-142306 and JP-A-59-107313 are used.
And Japanese Patent Application Laid-Open No. 60-101513 are disclosed by the present applicant.
. Specifically, the number of pixels constituting the sensor array SAA or SAB is N, and the i-th (i = 0)
,..., N-1) when the image signals from the sensor arrays SAA and SAB are A (i) and B (i)Or Is given by k₁≤k≤k_TwoIs calculated. M is represented by (M = N− | k | −1)
Is the number of operation pixels to be calculated, and k is called a relative displacement amount.₁, K_TwoIs usually -N / 2, N / 2. Here, the operator max {a, b} is a, b
Represents the extraction of the largest of the two, and the operator min {a, b}
This means extracting small things. Therefore, the term X in the above equations (1) and (2)₁(k),
X_Two(k), Y₁(k), Y_Two(k) can be considered as a correlation amount in a broad sense. Further, the above (1),
Looking at equation (2) in detail, X₁(k), Y₁(k) is actually the above in (k-1) displacement
The correlation amount according to each definition is expressed as X_Two(k), Y_Two(k) is the correlation amount at the displacement of (k + 1)
, Respectively. Therefore, X₁(k), X_TwoThe evaluation amount X (k) which is the difference between (k) is
It means the amount of change in the correlation amount between the image signals A (i) and B (i) at the relative displacement amount k.   X₁(k), X_TwoThe correlation amount (k) has the highest correlation between the two images as is clear from the above definition.
When the minimum. Therefore, the change X (k) is “0” when the correlation is the highest.
And the slope should be negative. However, since X (k) is discrete data,
actually     X (kp) ≧ 0, X (kp + 1) <0 (3) In addition, the peak of the correlation amount is found in the relative displacement section [kp, kp + 1] where X (kp) -X (kp + 1) is maximum.
Think it exists, By performing the interpolation calculation of, the image shift amount PR of a pixel unit or less can be detected.
Wear.   On the other hand, Y₁(k), Y_TwoWhen the correlation between the two images is the highest from the above definition,
X₁(k), X_TwoContrary to (k), it becomes the maximum. Therefore, the amount of change, Y (k), has the highest correlation.
In this case, it should be “0” and the slope should be positive. Y (k) is the same as X (k)     Y (kp) ≦ 0, Y (kp + 1)> 0 (6) And Y (kp) -Y (kp + 1) is maximum By performing the interpolation calculation of, the image shift amount PR of a pixel unit or less can be detected. Wear.   Further, the image shift amount can be detected using any of the focus evaluation amounts X (k) and Y (k).
However, as can be seen from JP-A-60-101513, | X (kp) -X (kp + 1) |
> | Y (kp + 1) -Y (kp) |, the focus evaluation amount X (k) is calculated as |
1) In the case of −Y (kp) |, it is better to obtain the image shift amount PR using the focus evaluation amount Y (k).
The accuracy is good.   FIG. 1 is a block diagram showing an example of a camera focusing device suitable for carrying out the present invention.
FIG. PRS is a camera control circuit, for example, a CPU (central processing unit) inside
, RAM, ROM, EEPROM (electrically erasable programmable ROM), input / output port and A
1-chip microcomputer with analog input port with / D conversion function
The camera sequence, AF (auto focus), and AE (auto exposure) are stored in the ROM.
Source: Control software in the EEPROM, necessary parameters for AF and AE control
Is stored. SHT keeps the data while the control signal CSHT is input from the control circuit PRS.
Data input via the data bus DBUS is received and not shown based on the data.
APR is a control signal for controlling the travel of the front curtain and rear curtain of the shutter.
While R is being input, it accepts data input via the data bus DBUS,
The aperture control circuit controls the aperture mechanism (not shown) based on the data.
While inputting, it accepts data input via the data bus DBUS, and
A display circuit that displays various shooting information based on the information, SWS is a release switch (not shown),
Switches for setting various information such as shutter and aperture, located outside and inside the camera
Switch group.   SPC is a photometric circuit, and an analog photometric signal SSPC, which is an output of the circuit, is provided by the control circuit P.
RS is sent to the analog input port with A / D conversion function, A / D converted and
Used as photometric data to control shutter control circuit SHT and aperture control circuit APR
Can be LCOM is input via the data bus DBUS while the control signal CLCOM is input.
Data and receive serial communication with a lens unit described later based on the data.
A lens communication circuit that communicates with the optical axis of the taking lens FLNS in synchronization with the clock signal LCK.
The lens driving data DCL indicating the amount of movement in the direction is sent to the lens control circuit described later.
At this time, at the same time, the control circuit in the lens Lens information DLC such as a coefficient of the amount of defocus versus the amount of lens movement is serially input. BSY shooting
This signal informs the camera whether the shadow lens FLNS is moving or not.
When the signal is "H" (high level), the serial communication becomes impossible.   LNSU is a lens unit, LPRS is a motor M based on serially input data DCL
ENC is a control circuit in the lens that drives TR and moves the taking lens FLNS in the optical axis direction.
For example, a pulse generated by the movement of the lens barrel holding the taking lens FLNS
Signal is detected and the encoder pulse signal is
Encoder circuit that outputs the signal EPL to the in-lens control circuit LPRS.   SDR has two sensor arrays SAA according to the signals STR and CK input from the control circuit PRS.
Sensor drive circuit that controls line sensor SNS such as CCD with SAB and SAB
is there.   Next, the operation will be described with reference to FIGS. Note that the shutter control
The operations of the circuit SHT, the aperture control circuit APR, the display circuit DSP, and the photometric circuit SPC are not directly related to the present invention.
Since there is no connection, detailed description is omitted here. Also, in this embodiment, the camera
The “AF” flow is called as a subroutine from the sequence flow.
ing.   When the AF operation is started, first, two flags, a flag RCFLG and a flag NRSDFLG,
Is set to "N" (meaning NO) (FIG. 2, step 10). The flag RCFLG
The function of NRSDFLG will be described later. Next, the image signal reading subroutine "IMA
GE "(step 11). Here, the sensor drive circuit SDR
The in-sensor SNS is driven, and image signals A (i) and B (i) of two images are obtained. At this time
Brief description of the operation of the sensor drive circuit SDR and the line sensor SNS with reference to FIGS.
I do. When an "H" accumulation start signal STR is output from the control circuit PRS (see FIG. 3).
50), a clear signal CL is output from the sensor drive circuit SDR to the line sensor SNS,
The charges in the photoelectric conversion units of the sensor arrays SAA and SAB are cleared. Then line sensor SN
A secondary imaging lens or the like arranged at the previous stage at S (not shown in FIG. 1,
15 arranged on the sensor arrays SAA and SAB.
The photoelectric conversion of the light image and the charge storage operation are started. After the above operation started
When a predetermined time has elapsed, a transfer signal is sent from the sensor drive circuit SDR. SH is output to the line sensor SNS, and the charge stored in the photoelectric conversion unit is transferred to the CCD unit.
It is. At the same time, an "H" accumulation end signal IEND is generated in the sensor drive circuit SDR,
The signal is input to the control circuit PRS (step 51). After that, the control circuit PRS switches to CC
When the D drive clock CK is output, the CCD drive signal φ is output from the sensor drive circuit SDR.₁, Φ_Two
Is output. As a result, the analog image is obtained from the line sensor SNS according to this signal.
The signal SSNS is output to the control circuit PRS, and in response, the control circuit PRS receives the signal SSNS.
A / D conversion of the analog image signal SSNS in synchronization with the clock CK, and image signals A (i) and B (i) of two images
Is stored at a predetermined address in the RAM (steps 52 to 55). Here Sen
It is assumed that the number of pixels in the sub-arrays SAA and SAB is 40.   Returning to FIG. 2 again, in step 12, all pixels are processed according to the out-of-focus state.
Perform focus detection calculation as a frame or focus detection calculation with some pixels as the processing target range
, That is, the flag VSN for setting the processing target pixel range for performing the focus detection calculation,
When "WD" (meaning a large amount of defocus), focus detection subroutine "WPRED"
When "NR" (meaning near focus), focus detection subroutine "NPRED"
Call (step 12). For example, as shown in FIG.
If it is out of focus, use the subroutine "WPRED" and focus near the focus as shown in Fig. 5.
Then, the subroutine "NPRED" is adopted. The power-on flag VSN is set to “WD
(Because it is the first time and the out-of-focus state is not known), the AF
Is started, “WD” or “NR” depends on the focus state at that time.
Is reset to This resetting will be described later.   Focus detection processing when subroutine "WPRED" is called is shown in the flow of Fig. 6.
Therefore, it will be described. In steps 100 to 105, the relative displacement amount k is set to “−20 to 20”.
The focus evaluation amount X (k) when changed within the range is obtained. Here, the relative displacement k
The reason why the number of pixels of the sensor rows SAA and SAB is "20 to 20" is as described above.
40 ", but this processing target pixel range is
Variable according to the focal length of the lens FLNS.   First, in step 101, the number M of operation pixels is calculated by the equation M = 39− | k |.
The number of calculation pixels M is variable according to the relative displacement amount k, and becomes smaller as the absolute value of k becomes larger.
Become. This is because as the relative displacement k increases, the output of the corresponding sensor Because they are missing. In step 102, the sign of the relative displacement k (positive or negative)
, And then, according to the sign, start the pixel position P at which the calculation of the A and B images is started.
A and PB are calculated in step 103 or 104. In Step 105 focus evaluation
The calculation of the value X (k) is performed.   Here, the processing steps in steps 100 to 105 will be described with reference to FIG.
FIG. 7A shows two image signals A (i) and B (i). FIG. 7 (b) shows that k = −20.
Represents the correspondence between the sensor rows in the correlation operation in this case, and at this time, M = 39− | 20 | =
At 19, PA is "0" and PB is "20". That is, the relative displacement k is a negative value.
Therefore, the B image is relatively displaced to the left by k pixels (−20 pixels). This response
From the relationship, X (-20) is obtained by calculating the corresponding relationship in which the B image is further displaced to the left by one pixel.
Conversely, the first term of the equation was calculated based on the correspondence that the image A was displaced to the left by one pixel.
Correspond to the second term. In the first and second terms, the A image and the B image are each one pixel.
When calculating the number of calculated pixels M, M = 40− |
Instead of k |, M = 39− | k |. FIG. 7 (c) shows the correspondence when k = 0.
Represents. FIG. 7 (d) shows the correspondence of the correlation operation at k = 20.
Contrary to FIG. 7 (b), the image A is displaced to the left by 20 pixels.   An example of a plot of the focus evaluation amount X (k) calculated as described above is shown in FIG.
Shown in the figure.   Returning to the flow of FIG. 6, in step 110, the two images A,
The peak value kp of the image shift amount in the pixel unit of B is detected. Hereinafter, steps 120-1
In step 46, an image shift amount PR of a pixel unit or less is obtained.
In steps 120 to 135, two focus evaluations are performed based on the peak value kp obtained in
Recalculate the quantities X (k) and Y (k). There are two reasons for this. One is where to find kp
In the step, the number of calculation pixels M is variable according to the relative displacement amount k.
Interpolation of the focus evaluation amount X (k) calculated in the calculation range of
When the PR is obtained, there is a possibility that an error due to the mismatch of the number of calculation pixels M may be included.
There is. Another method is to calculate X (k) rather than calculating the image shift amount PR using only the focus evaluation amount X (k).
, Y (k) in the earlier application (Japanese Patent Application Laid-Open No. 60-101513)
As described above, depending on the signal pattern of the subject, is it superior in S / N? It is. From the above, in steps 120 to 135, the calculation pixel is calculated based on kp.
The number M is kept constant (step 120), and the focus evaluation amounts X (k) and Y (k) are simultaneously obtained.
You.   First, in step 120, the calculation of M = 38− | kp |
Decide. Next, in steps 130 to 135, k =
The focus evaluation amounts X (k) and Y (k) are calculated at the three points kp-1, kp and kp + 1 in the same manner as above.
You. Note that when calculating the number of calculation pixels M, M = 38− | kp |
Number of calculated pixels in the relative displacement amount having the largest absolute value among the three points of 1, kp, kp + 1
This is for fixing to M. Next, the focus evaluation amounts X (k), Y (
From k), the image shift amounts kpx and kpy in pixel units due to the respective focus evaluation amounts are detected again.
(Steps 140 and 141). At this time, the contrast of each focus evaluation amount X (k), Y (k)
XD (XD = X (kpx) -X (kpx + 1)) and YD (YD = Y (kpx + 1) -Y (kpx
)) Is also required. This is in accordance with the calculation method disclosed in the prior application in this embodiment.
Therefore, instead of comparing the magnitude of the image signal at the end for each relative displacement (computation processing becomes complicated)
When the amount of out-of-focus is large, the contra
By looking at the strike evaluation amounts XD and YD, it is as if looking at the image signal at the end for each relative displacement.
This is because the information is used while paying attention to the fact that information as if
The larger the strike evaluation amounts XD and YD, the better the S / N ratio. Therefore
In step 142, the two contrast evaluation amounts XD and YD are compared, and XD ≧ YD is satisfied.
At the time, the focus evaluation amount X (k) is adopted (step 143). When XD <YD, the focus evaluation amount X (k) is used.
The quantity Y (k) is adopted (step 144). In steps 145 and 146,
ZD (Z₁-Z_Two), Using kz     PR = kz + | Z₁/ ZD | To calculate the image shift amount PR of a pixel unit or less. This process is shown in FIG.
Is shown. In the example as shown in this figure, since the relationship of XD <YD is satisfied, the focus evaluation amount Y (k
) Is adopted, and kz = kpy, Z at the time of calculating the image shift amount PR in pixel units or less.₁= Y (kp)
, Z_Two= Y (kp + 1) is used.   Steps 150 to 158 are processing for determining the focus state.
When 0, the four flags LCFLG, SDFLG, NJFFLG, JFFLG are set to "N". To qualitatively describe the meaning of each flag, “LCFLG” indicates that the subject image signal is low.
This is a flag indicating that it is the last, and the shift amount of the two images “SDFLG” is relatively small
That is, a flag indicating that the out-of-focus amount of the photographing lens FLNS is relatively small,
"NJFFLG" has a very small displacement between the two images, that is, the photographing lens FLNS is almost in focus.
"JFFLG" indicates that there is almost no shift between the two images.
That is, this is a flag indicating that focus is achieved.   Next, the contrast evaluation amount ZD of the adopted focus evaluation amount is compared with a predetermined value LCTH.
(Step 151). As a result, when ZD <LCTH, the contrast is considered to be low.
Then, set “LCFLG” to “Y” (meaning YES) and end the subroutine “WPRED”
I do. If ZD ≧ LCTH, contrast is considered to be sufficient to perform focus detection
No, proceed to the next step 153. In step 153, the image shift amount PR is
The value PR is compared with the predetermined value “3”, and if PR> 3, that is, the image shift amount is 3 pixels or more.
If so, the subroutine "WPRED" ends. Relatively disqualified if PR ≤ 3
Since the amount of focus is small, “SDFLG” is set to “Y” (step 154), and then PR and
The fixed value "1" is compared (step 155). As a result, if PR> 1, ie, no image
When the shift amount is one pixel or more, the subroutine "WPRED" ends here. PR
If> 1, it is determined that the object is substantially focused and "NJFFLG" is set to "Y" (step 156).
Finally, the PR is compared with a predetermined value JFTH (step 157). The predetermined value JFTH is not imaged
The shift amount PR is a value that can be regarded as being in focus. As a result, if PR> JFTH,
Exit the routine "WPRED". If PR ≤ JFTH
“JFFLG” is set to “Y” (step 158), and the subroutine “WPRED” ends.
. The functions of the four flags LCFLG, SDFLG, NJFFLG, and JFFLG will be described later.
I do.   Next, the subroutine "NPRED" will be described. "NPRED" is near focus
Equivalent to the distance measurement frame in the viewfinder among all sensors
The focus detection process is performed only on the data within the range. This is what the viewfinder and
This will be described with reference to FIG. 10 showing the positional relationship between the sub-rows SA. In the figure, FFRM is
This is the distance measurement frame on the viewfinder. In the subroutine "NPRED", the range of the distance measurement frame FFRM
And R₁, R_Two, R_ThreeIs divided into three areas as shown by Area R in the range frame FFRM₁, R_Two, R_ThreeIs located in each pixel of the sensor row SA.
Pixel position NR corresponding to the beginning of the part₁, NR_Two, NR_ThreeBeforehand in the adjustment process.
Stored in the EEPROM in the control circuit PRS. At this time, the NNPX included in each part is the same
(R₁= R_Two= R_Three), And this value NNPX is stored in the EEPROM as well. this
In the embodiment, NNPX = 12 and NR₁, NR_Two, NR_ThreeAre "8", "14", "2"
0 ”, and (NNPX + k) is the processing target range in this case (the hatched portion in FIG. 12).
Minutes). These values depend on the installation of the focus detection optical system and the line sensor SNS.
There may be some pixel shift depending on the situation, but since it is stored in the EEPROM as described above,
The range of the calculation for performing the focus detection processing can be matched with the range frame FFRM. The fifth
R shown for the image signal in the figure₁, R_Two, R_ThreeIs the above subroutine "NPRED" processing
Are the divided areas.   The operation when the subroutine "NPRED" is called according to the flow of Fig. 11.
State. In step 200, the number of calculation pixels M is set to NNPX. The subroutine "
In “WPRED”, the number M of calculated pixels was variable according to the relative displacement amount k.
In outgoing processing, it is fixed to NNPX. This is because "WPRED" uses the output of all pixels for calculation
On the other hand, NPRED uses some outputs of the sensor array SA,
This is because data at the end of the corresponding sensor output is not lost. Therefore
In “NPRED”, the number of calculation pixels M performed in steps 120 to 141 of “WPRED” is one.
No recalculation is performed. Also, since recalculation is not performed,
From the beginning, the evaluation amounts X (k) and Y (k) are simultaneously obtained.   Steps 210 to 221 correspond to region R₁This is a focus detection process for the image data.
In the figure, the same processing parts as “WPRED” are given the same step numbers. Subroutine
In “NPRED”, the range of the relative displacement amount k is (−4 to 4). This is "NPRED
Is applied near the in-focus state, and the image shift amount is originally small.
You. In this case, as shown in steps 212 and 213, the region R₁Start position NR₁But
Is considered. This process is shown in FIG. 13 (a). The solid line in the figure indicates the focus evaluation amount X (
k) and the dashed line represents Y (k). Processing similar to that described in “WPRED” (step
140 to 146), the region R₁PR at image₁Ask for. Step 221
And image shift PR₁And contrast evaluation amount ZD₁In the control circuit PRS Store it in RAM. Similarly, in steps 230 to 241, the region R_TwoIn
Image shift PR_Two, Contrast evaluation amount ZD_Two(See Fig. 13 (b))
In steps 250 to 261, the region R_ThreePR at image_Three, Contrast evaluation amount
ZD_Three(See FIG. 13 (c)), and these are also stored in the RAM in the control circuit PRS.
.   In steps 270 to 278, the three regions R determined as described above are used.₁, R_Two, R
_ThreeImage shift PR₁, PR_Two, PR_ThreeOf which image shift amount is the final image shift amount
Judgment as to whether to make PR is made. That is, in the present embodiment, of the three regions
, The contrast is sufficiently high, and the maximum image shift amount is adopted.
. The reason for adopting the maximum image shift amount here is that when the image shift amount is positive, the back focus is set.
, The most focused object is the closest object, and the
It is to match. The final image shift amount PR and the contrast evaluation amount ZD
After the determination, the focus state is determined in step 280. The process here is "WPRED"
Since the processing is the same as that described above, the description thereof is omitted.   As described above, if the out-of-focus amount is large, the subroutine "WPRED"
In the near-focus state, the subroutine "NPRED"
The amount of shift, that is, the amount of out-of-focus of the photographing lens FLNS can be detected.   Returning to FIG. 2, the description from step 15 will be continued. Step 15 is the flag LC
The control during the focus detection processing of “WPRED” or “NPRED” performed earlier after looking at the FLG
Check the last. When LCFLG is "Y", the contrast is low and the next time
In the focus detection processing, the flag VSN is set to “WD” so as to use “WPRED” (step 35).
), Execute the display subroutine "DISP" and the lens control subroutine "LENS" (
Steps 19 and 20). The display subroutine "DISP" is not directly related to the present invention.
The lens control subroutine "LENS" will be described later.
You.   In step 16, the flag VSN is checked again. When VSN is “WD”,
The processing after step 17 is performed, and when "NR", the processing after step 28 is performed.   First, processing when VSN = “WD” will be described. VSN = "WD" means the focus detection processing
Represents what was done with "WPRED". In step 17, check flag RCFLG I do. RCFLG is “Y” when performing focus detection processing twice in the “AF” routine.
ing. In this embodiment, when executing the “AF” routine, the first “AF”
Execute either “WPRED” or “NPRED” with the VSN set in (in the release
Initially, the VSN was "WD", but the resulting image
If the amount of deviation is not appropriate for each condition, reset VSN and set flag RCFLG.
As "Y", a different focus detection process "NPRED" or "WPRED" will be executed.
Swell. RCFLG is the flag for that. The resetting of the VSN will be described later.   When the flag RCFLG is “N”, that is, at the time of the first focus detection processing, the flag SDFLG is checked.
(Step 18). SDFLG is "WPRED" or "NPRED" as described above
(The SDFLG in this step is set in "WPRED"
), When this flag is “Y” (within 3 pixels), the current focus state is relatively close to in-focus.
Means that Therefore, VSN is “WD” and SDFLG is “N” (3 pixels or more).
If the "WPRED" is appropriate, display the subroutine "DISP" and control the lens.
Move to execution of subroutine "LENS". On the other hand, if SDFLG is “Y”, “NPR
It is more appropriate to perform recalculation with "ED", that is, it is possible to obtain an accurate image shift amount PR.
In consideration of this, VSN is set to "NR" (step 21), and the image shift amount PR is stored in the buffer BPR.
Is temporarily stored (step 22), RCFLG is set to “Y” (step 23), and
Routine from step 12 (actually step 12 → step 14 → step 1
5 → Step 16 → Step 28 → Step 33 →). Said
The reason for temporarily storing the image shift amount PR in the buffer BPR in step 22 is as follows.
If the result calculated in subroutine "NPRED" is not appropriate, that is, "NPRED"
When SDFLG at step 33 is “N” (3 pixels or more)
This is because the image shift amount PR obtained in the first subroutine “WPRED” is used.
Step 34 → Step 27 →).   When RCFLG is "Y" in step 17, that is, at the time of the second focus detection process,
Go to Step 24 and check SDFLG here. As a result, if SDFLG is "N",
Since the focus detection process at this time is “WPRED”, the focus detection process here is appropriate.
Execute the display subroutine "DISP" and the lens control subroutine "LENS" Move to. When SDFLG is “Y”, RCFLG is “Y” and VSN is “WD”.
Therefore, the first focus detection processing is performed with “NPRED” and the result is
It would have been painless. As described later, the first time of "NPRED", the flag NJFFLG
Is "Y", that is, it is judged that it is not appropriate unless it is approximately in-focus.
If "Y", the flag NRSDFLG is set to "Y" (not in focus but within 1 pixel)
ing. In other words, if NRSDFLG is “Y”, the first focus of “NPRED” is almost in focus
It does not mean that it was relatively close to focusing. At this time, the statue at "NPRED"
The shift amount PR is temporarily stored in the buffer BPR. So in step 24 SDFLG
If “Y”, it means that the focus is relatively close in the second “WPRED”, so “N”
PRED "Checks the flag NRSDFLG in step 25 to refer to the first result
. If NRSDFLG is "N" (one or more pixels), the first time is not appropriate.
And the display subroutine “DISP” using the second image shift amount PR of “WPRED”
Shifts to the execution of the end control subroutine "LENS". If NRSDFLG is “Y”, then
In “AF”, VSN is set to “NR” in order to set the first focus detection process to “NPRED”.
Step 26) The image shift amount P in the first time of “NPRED” already stored in the buffer BPR
Take out R (step 27), display subroutine "DISP", lens control sub
The process temporarily shifts to the execution of the routine “LENS”.   If the VSN is determined to be “NR” in step 16 by going up the flow, step
At 28, the flag RCFLG is checked. Here, when RCFLG is “N”, “NPRED” once
Eyes, "Y" means "NPRED" for the second time. In the case of “N”,
Check SDFLG at step 29. When SDFLG is “N”, “NPRED” is not appropriate
VSN is set to “WD” so that the second focus detection process is performed by “WPRED”.
Step 30), temporarily stores the image shift amount PR in “WPRED” in the buffer BPR (Step 30).
Step 22), set RCFLG to “Y” (step 23), and return to step 12
To the execution of the routine (actually, step 12 → step 13 → step 15
The process shifts to execution of → step 17 → step 24 →). SDFLG at step 29
Is "Y", the flag NJFFLG is checked in step 31. This is the previous
As described in step 25, in the first “NPRED”, focus detection processing is performed by NJFFLG.
To determine if the process was appropriate. NJFFLG in step 31 If "N" (1 pixel or more and 3 pixels or less), it is not nearly in-focus but relatively close to in-focus
So, the second focus detection process "WPRED",
Set the flag NRSDFLG to "Y" for use in step 25.
Move to 30. "NPRED" temporarily stored in the buffer BRP in step 22
The first image shift amount PR may be changed as needed, that is, the “WPRED”
Used in   When RCFLG is "Y" in step 28, it is the second time of "NPRED".
As you can see, the first time means that the result was judged to be inappropriate in "WPRED"
I do. Also, as described above, the SDFLG is checked in step 33, and if "Y", "NPR
ED "is determined to be appropriate for the second time, and the display subroutine" DISP
", And shifts to execution of the lens control subroutine" LENS ". SDFLG is “N”
If the "NPRED" second time is inappropriate, the focus detection process in the next "AF" will be "W
Set VSN to “WD” to set to “PRED” (step 34), and temporarily set the first time to “WPRED”
The image shift amount PR stored in the storage buffer BPR is extracted (step 27).
Move to execution of display subroutine "DISP" and lens control subroutine "LENS"
I do.   To briefly summarize the flow of the above “AF” processing, when “AF” is called,
The image signal of the sensor array SA is read (“IMAGE”), and the focus detection processing “WPRED” or “NP
WPRED is used when the amount of defocus is large, while NPRED
Based on the assumption that it will be applied near the focus, the detected image shift amount is
If it is not appropriate, recalculate the same image signal with an appropriate processing method,
When the defocus amount is large, the image shift amount PR to be adopted is determined by “WPRED”, and the focus is determined to be close to the focus.
Sometimes it is calculated by "NPRED". In other words, when the out-of-focus amount is large,
If the focus detection processing is performed by “WPRED”, the out-of-focus state is better understood.
If it is determined that the result is near the focus, the focus detection processing is performed by “NPRED”.
This is because the focus detection accuracy can be improved.   Next, the subroutine "LENS" will be described using the flow of FIG. S
At step 60, the flag LCFLG is checked. This flag is used for focus detection as described above.
When the contrast evaluation amount of the image signal to be processed is high, “N” is set. When LCFLG is "N", the image signal had sufficient contrast for focus detection processing
The control of the photographing lens FLNS is performed based on the image shift amount PR as the processing result.
First, the signal BSY input to the lens communication circuit LCOM is checked (step 61).
While this signal BSY is "H", communication with the in-lens control circuit LPRS is not possible.
Wait until the signal BSY becomes "L" (low level). When the signal BSY becomes "L"
From the control circuit LPRS in the lens to the relationship between the defocus of the taking lens FLNS and the lens movement amount.
The number S is input from the signal DLC (step 62). Next, based on the image shift amount PR,
Then, the defocus amount DEF of the flash FLNS is calculated by the formula of DEF = K · PR (step 6).
3). K is a value set by the focus detection optical system, and is set in advance. Continued
FP = DEF from the defocus amount DEF and the coefficient S.
/ S (step 64). FP corresponds to the count number of the encoder circuit ENC.
Hit. Then, in step 65, the signal BSY is checked again, and the signal BSY is set to "L".
”, The lens movement amount FP (signal DCL) is communicated to the lens control circuit LPRS.
(Step 66). On the other hand, when LCFLG is “Y”, the contrast is not sufficient.
It is determined that the moving amount FP of the photographing lens FLNS for performing the search operation is
It is calculated by CDEF / S (step 67). CDEF can be moved by one search operation.
The lens movement amount is converted into a defocus amount, which is set in advance.   When the lens movement amount FP is input, the in-lens control circuit LPRS, based on the information,
To control the driving of the photographing lens FLNS. When the shooting lens FLNS stops, the image
A series of "AF" processing of focus adjustment that has been started from the input of the signals A (i) and B (i) ends.   According to this embodiment, the range (effective range) of the pixel corresponding to the finder ranging frame is adjusted.
Because it is checked in advance in the adjustment process and stored in the EEPROM,
No adjustment is required, and the work efficiency is very good. In addition, the same configuration
Cameras that differ only in the viewfinder ranging frame (matte surface, split image
Even if it is desired to change to (e), the change becomes easy. In addition, after assembly
If the sensor is found to have sensitivity unevenness, or if there is aberration in the AF optical system
It is easy to avoid using that part when you know You. (Correspondence between invention and embodiment)   In this embodiment, the photographing lens FLNS is provided with a field lens in the imaging optical system of the present invention.
FLD and secondary imaging lenses FCLA and FCLB in the optical system, and sensor rows SAA and SAB
The CPU corresponds to the arithmetic means, and the EEPROM corresponds to the storage means. (Modification)   In the present embodiment, the “image shift type” apparatus has been described, but the present invention is not limited to this.
Rather than detecting the change in contrast at the focal plane,
It can be applied to a point detection device. (The invention's effect)   As described above, according to the present invention, corresponding to the focus detection area of the observation unit,
Storage means for storing in advance an effective pixel range for generating an effective photoelectric conversion signal in an adjustment step
Having a focus detection area and pixels on the effective pixel range in the observation means.
And the focus detection areas of the effective images stored in the storage means are adjusted.
Since the focus detection processing is performed using the photoelectric conversion signal in the elementary range, the photoelectric conversion means
Without mechanical adjustment of the focus detection area of the observation unit and the pixels of the photoelectric conversion unit
Can be easily matched with each other.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a block diagram showing an example of a focus adjusting device for a camera suitable for carrying out the present invention, FIG. 2 is a partial flowchart thereof, and FIG. FIGS. 4 and 5 are flowcharts showing examples of image signals output from the sensor array.
FIG. 7 is a partial flowchart, FIG. 7 is a diagram showing the correspondence between two images at the time of focus detection calculation, FIG. 8 is a diagram showing a change in the evaluation amount, and FIG. 10, FIG. 10 is a partial flowchart, FIG. 11 is a diagram showing a positional relationship between a finder ranging frame and a sensor array, FIG. 12 is a diagram showing a correspondence relationship between two images at the time of focus detection calculation, and FIG. FIG. 14 is a diagram showing a change in a part of the evaluation amount, FIG.
FIG. 5 is an arrangement diagram showing an optical system of a general secondary imaging type focus detection device. PRS: Control circuit, LCOM: Lens communication circuit, LNSU: Lens unit, SDR
… Sensor drive circuit, SAA, SAB… Sensor array, FLNS… Shooting lens, k… Relative displacement.

Claims (1)

Claims: (1) A photoelectric conversion means comprising a plurality of pixels, an optical system for forming an image of light having passed through an imaging optical system on said photoelectric conversion means, and a photoelectric conversion signal from said photoelectric conversion means being predetermined. A focus detection device having a calculation means for detecting a focus state of the imaging optical system, and a focus detection area for observing a target through the imaging optical system. An optical device including an observation unit provided with a storage unit in which an effective pixel range for generating an effective photoelectric conversion signal corresponding to the focus detection area is stored in advance in an adjustment step, and the focus detection in the observation unit is performed. An optical device, wherein an area and a focus detection area on a pixel according to the effective pixel range are matched.