JPH03164727A - Focus detecting photometric device - Google Patents

Abstract

PURPOSE:To obtain the photometric value optimum for the region of the optimum group used for the computation of focus detection by rearranging the results of the focus detection of plural focus detecting photometric regions to the group considered to be the same subject and emphatically executing photometry of the selected group. CONSTITUTION:This device has a focus detecting means 2 which detects the defocusing quantity of the present image plane to the respective prescribed planes in the plural focus detecting regions, a photometric means 3, a grouping means 4 which rearranges the plural focus detecting regions to plural groups according to the defocusing quantity of the regions, a selecting means 5 which selects the one optimum group from the plural groups, an optimum defocusing quantity computing means 6 which calculates the defocusing quantity in accordance with the defocusing quantity of the focus detecting region belonging to this optimum group, and a photometric computing means 7 which obtains the optimum arithmetic value of photometry by increasing the weighting in the photometric output of the region belonging to the optimum group. The computation of the optimum photometric value is possible in this way even in the case in which the main subject is across the plural discrete photometric regions.

Description

DETAILED DESCRIPTION OF THE INVENTION A. Field of Industrial Application The present invention relates to a focus detection photometry device that calculates an optimal photometry value by performing photometry in a plurality of areas in conjunction with focus detection.

B. Prior art As a conventional device of this type, Japanese Patent Application Laid-Open No. 63-58324,
The one disclosed in Japanese Patent Application Laid-open No. 1-105221 is known. These devices have means for detecting focus on multiple areas or points in a subject, and photometry means for performing spot photometry on multiple areas that are the same as the focus detection areas or points, or areas that include them. When the focus detection area in which the photographic lens should be focused is selected based on the result, the photometric value of the photometric area corresponding to the selected area is selected, and exposure display or control is performed based on the selected photometric value. It is something to do.

Problems to be Solved by the Invention In such conventional devices, the focus value to be selected is selected from one of the focus detection outputs obtained by each single focus detection area. Therefore, the area adopted as the photometric value in conjunction with the focus detection result also corresponded to the single focus detection area selected above. Therefore, the middle part is like the subject, etc. ! If there are multiple subjects scattered around, both of which are at approximately the same distance and also the main subject, it would be a mistake to use the photometric output of the photometric area corresponding to the focus detection area facing one of the subjects as the photometric value. It is.

SUMMARY OF THE INVENTION An object of the present invention is to make it possible to calculate an optimal photometric value even when a main subject straddles a plurality of discrete photometric areas.

D, ill! means to solve the problem This will be explained with reference to FIG. 1, which is a complaint correspondence diagram.

The focus detection photometry device according to the present invention includes a photographing optical system 1 for forming a subject image on a predetermined plane, and a plurality of focus detection areas AREA(j) (where j=
1 to f), the current defocus amount DEF(j) of the image plane with respect to the above-mentioned predetermined plane (however, j=1 to
a focus detection means 2 for detecting f), a photometry means 3 for obtaining a photometry output Bv(j) for each of the plurality of focus detection areas AREA (j) and the area between the two areas, and a plurality of focus detection areas A
a grouping means 4 for grouping REA (j) into a plurality of groups GRP(i) (where i=l to e) according to the defocus amount of the area;
a selection means 5 for selecting one optimal group GRP (w) in which an image of a subject to be focused is expected to be formed from among i); Optimal defocus amount calculation means 6 calculates the defocus amount DEFX based on the focus amount, and increases the weighting of the photometry output of the area belonging to the optimal group GRP (w) to obtain the optimal photometry calculation value B vans
The above-mentioned technical problem is solved by providing the photometric calculation means 7 for obtaining the above-mentioned.

A part of a subject image formed by a photographing optical system 1 such as an E0 effect camera is guided to a focus detection means 2 comprising a well-known focus detection optical system, an image sensor, a microcomputer (CPU), and the like.

The focus detection means 2 detects a plurality of focus detection areas (AREA(j), j=l) set on the photographing screen as shown in FIG.
~f (f=8 in FIG. 5)) is subjected to well-known focus detection calculations to obtain defocus amounts (oEF(j)) of a plurality of focus detection areas.

The photometer 3 measures the light in the eleven focus detection areas AREA(j) and the area between them, and obtains a photometry output By(j). The grouping means 4 groups the plurality of focus detection areas AREA (j) into several groups for each focus detection area that seems to capture the same subject based on the defocus amount (DEF (j)). (GRP (i), i=l~e)
summarized in.

The optimal group selection means 5 selects a plurality of groups (GRP(i
)), the optimal group (GRP(w)
).

The optimum defocus amount calculating means 6 is an optimum group (GRP).
Based on the defocus amount of the focus detection area belonging to (w)), the optimum defocus amount DEFX for the entire system is calculated.

The photometric calculation means 7 increases the weighting of the photometric output of the area belonging to the optimal group (GRP (W)) to obtain the optimal photometric calculation value B vans.

As a result, a stable defocus amount with little variation can be obtained, and an optimal photometric value can be obtained for the region of the optimal group used in the focus detection calculation.

F. Example of Part 1 of Embodiment The configuration of a first example in which the focus detection device according to the present invention is applied to an interchangeable lens type eye reflex camera system will be described with reference to FIG.

An exchangeable lens 10 can be detachably mounted on a camera body 20. When the lens 10 is attached, a part of the photographic light beam coming from the subject passes through the photographic lens 11 and is reflected by the main mirror 21 provided on the camera body 20 and guided to the finder, and the other part is reflected by the main mirror 21 provided on the camera body 20. The light passes through the main mirror 21, is reflected by the sub-mirror 22, and is guided to the AF module 23 as a light beam for focus detection.

As shown in FIG. 3, the AF module has a field of view mass 70,
Field lens 27. It is composed of a focus detection optical system 24 consisting of a half mirror 120 and a pair of re-imaging lenses 28A and 28B, and a photoelectric conversion device 25 such as a CCD consisting of a pair of light receiving sections 29A and 29B.

In the above configuration, the exit 1l of the photographic lens 11
A pair of regions 1 symmetrical with respect to the optical axis 17 included in f16
The luminous flux passing through 8A and 18B respectively is.

A primary image is formed near a field mass 70 having an aperture shape corresponding to the entire focus detection photometry area shown in FIG. 5A. A portion of the primary image formed at the opening of the field mass 70 is further formed by a field lens 27 and a pair of reimaging lenses 28A,
28B, a pair of light receiving parts 29A of the photoelectric conversion device 25,
29B as a pair of secondary images.

As is well known, the amount of defocus of the photographic lens 11 can be detected by detecting the relative positional relationship of the paired secondary images in the direction in which the light receiving sections are lined up on the photoelectric conversion device 25. Furthermore, by performing this positional relationship for each of a plurality of focus detection areas set on the photographing screen as shown in FIG. 5A, it is possible to detect the amount of defocus for each focus detection area.

FIG. 4 shows the arrangement of light receiving sections on the photoelectric conversion device 25.

The light receiving sections 29A and 29B each have n light receiving elements AP+B.
p (p=1 to n), and when the -order image coincides with the film surface, the outputs of each corresponding pair of light receiving elements (Al and B1, A2 and 82,...) are equal. It is arranged like this.

The light-receiving elements forming the light-receiving sections 29A and 29B are composed of charge storage type elements such as photodiodes,
By accumulating charges for a charge accumulation time corresponding to the illuminance on the photoelectric conversion device 25, it is possible to control the light receiving element output signal to an output level suitable for focus detection calculations to be described later.

By configuring the focus detection optical system as described above, a plurality of focus detection areas as shown in FIG. 5 are set on the photographing screen.

Returning to FIG. 2 again, the explanation will be continued.

On the other hand, the photometric light beam reflected by the half mirror 120 in the AF module is imaged onto the photometric photoelectric conversion device 134 by the re-imaging lens of the photometric optical system 132 in the photometric module 130. Photoelectric conversion device 13 for photometry
4, the light-receiving surface of the photodiode is divided as shown in FIG. 5B, and a current-voltage conversion circuit is disposed in each divided area, and each photometering area is located between the focus detection divided area and the path. It is set so that Each output of the above current-voltage conversion circuit is the AECPUloo boat P7 to P
Sent to 14th.

The sensor control unit 26 is an AFCPU (CPU for AF)
It receives and receives charge accumulation start and end commands from the boat P4 of 30, and controls the charge accumulation time of the photoelectric conversion device 25 by giving control signals corresponding to the commands to the photoelectric conversion device 25. It also provides a transfer lock signal and the like to the photoelectric conversion device 25 to transfer the light-receiving element output signal to the AFCPU 30 in time series, and sends a synchronization signal synchronized with the start of transfer of the light-receiving element output signal to the boat P4 of the AFCPU 30. A
In synchronization with this signal, the FCPU 30 starts A/D conversion of the light receiving element output signal input to the boat P3 using the built-in A/D converter, and obtains A/D conversion data corresponding to the number of light receiving elements. When the A/D conversion is completed, the obtained data is subjected to data processing, which will be described later, to determine the optimum defocus amount.

That is, the focus detection means 2 in FIG. Grouping means 3. Optimal group selection means 4. The operation of the optimum defocus amount calculating means 5 is realized by the program of A F Cr' U30.

The AF CPU 30 controls the display form of the display units 41, 42, 43° 44 of the AF display device 40 using the port P5 based on the optimum defocus amount.

Furthermore, the AFCPU 30 controls the driving direction and driving amount of the AF motor 50 as described below based on the optimum defocus amount to move the photographing lens 11 to the in-focus point.

First, the AFCPtJ30 generates a drive signal from the port P2 to rotate the AF motor 50 in a direction in which the photographing lens 11 approaches the in-focus point according to the sign of the defocus amount (front focus, rear focus). The rotational movement of the AF motor 50 is caused by the rotation of the AF motor 50.
0, a body-side coupling 53 provided on the mount portion of the lens 10, and a lens-side coupling 14 fitted thereto, and a lens transmission system 12 comprising gears etc. built into the lens 10. Finally, the photographing lens 11 is moved in the focusing direction.

The amount of drive of the AF motor 50 is determined by converting the amount of rotation of gears and the like constituting the body transmission system 51 into a pulse train signal by an encoder 52 composed of a photointerrupter or the like.

The pulse train signal is fed back to port P1,
The AFCPU 30 calculates A by counting the number of pulses.
The drive amount of the F motor 50 is detected and controlled.

The lens 10 has a built-in lens CPU 13, and is connected to the port P via a communication bus 64 formed from a lens side contact 15 and a body side contact 63 provided on the mount.
6 connects to the AFCPU 30 and controls the AF of the lens 10.
Related information is sent to AFCPU30.

The focus detection mode selection device 80 selects a focus detection mode (center weighted) automatically or manually selected.

Port P
7, and the AFCPU 30 switches focus detection processing based on this information.

AECPUloo receives the photoelectric output of each region from the photoelectric conversion device 134 for photometry from ports P7 to P14, sequentially performs AD conversion, multiplies them by a predetermined constant, logarithmically compresses them, and obtains the photometric value Bv(j) of each region. Calculate.

In addition, the defocus amount and group of each area calculated in the AFCPU 30 from the port P9 of the AFCPU 30,
Information on optimal group and optimal defocus amount is sent to port P.
Step 2 is to calculate the optimum photometric value using the photometric value and each piece of information. Namely.

The operations of the photometric means 3 and the photometric calculation means 7 in FIG.
This is realized by the ECPUloo program.

After calculating the optimal photometric value, AECPUI 00 is set to port P3.
, P4, and P5 to control the shutter control device 140, aperture control device 150, and strobe control device 160 individually or in parallel to provide proper exposure to the film.

The photometry mode converter 180 sends information on the automatically or manually selected photometry mode to port P6, and converts it to AECP.
Uloo switches the optimum aperture value calculation process based on this information.

The above is an overview of the configuration and operation of an embodiment in which the focus detection photometry device according to the present invention is applied to a single-lens reflex camera.

Next, details of focus detection, grouping, optimal group selection, calculation of optimal defocus amount for a plurality of areas performed within the AFCPU 30, and photometry calculation performed within the AECPU 100 will be described.

Focus Detection Process> First, the focus detection process will be explained using FIGS. 6 and 7.

The light receiving element AP obtained through A/D conversion by the AFCPU30,
The light receiving element output data corresponding to Bp (p=1 to n) is assumed to be aP*bp (p=1 to n).

First, the correlation amount C(j
, L) is obtained.

C(j, L)=ΣIa(r+[,) −b(r)l
- (1) However, in equation 1), L is an integer and is the relative shift amount (shift amount) of the pair of light receiving element output data in units of the pitch of the light receiving elements, and J represents the focus detection area. . Further, in the integration calculation of equation (1), the range taken by the parameter r is determined as appropriate depending on the shift amount and the focus detection area j.

When the light receiving element output data aPs bp corresponds to a matrix, for example, as shown in FIG. 6, (1)
The combination of light-receiving element output data in the equation, that is, the range of the parameter r can be determined. In FIG. 6, the shift scale is moved in the range of -2 to +2, and the area surrounded by thick lines represents the position on the matrix of the combination of light receiving element output data at which the correlation calculation is performed.

For example, if the shift amount is 0, the parameter r in equation (1)
The range taken for each focus detection area j is as shown in the following equation.

When j=1 r=1~t When j=2 r=t +1~X When j=6 r=u +1~V j;4 When r=v +1~W When j=5 r= w +1 to X When j=6 r=x+1 to y When j=7 r=y+1 to 2 When j=8 r=z+1 to n... (2) The light receiving element output data in each focus detection area The shift amount xm (j) with the highest correlation is determined by applying the three-point interpolation method disclosed in Japanese Patent Application Laid-Open No. 60-37513 by the applicant to the result of equation (1) (3) It can be obtained as shown in the formula.

However, in equation 3), D and 5LOP are such that the minimum value of the discretely determined correlation amount C(j, L) is the shift amount L=
Assuming that it can be obtained at Can be done.

DEF(j)=KX(j)XPY(j)Xxm(j)+
CZ(j) (6) In equation (6), PY (j) is the pitch in the direction in which the light receiving elements are lined up for each focus detection area, and a value corresponding to the area is stored in the AFCPU 30.

KX (j) is a coefficient determined for each focus detection area by the configuration of the focus detection optical system shown in FIG. 3, and a value corresponding to the area is stored in the AFCPU 30.

cz (j) is an offset value for each area, and is a correction value determined by the amount of aberration of the photographing optical system (read from the lens CPU 13) and the position adjustment state of the AF module 23 with respect to the body 20 (stored in the EEPROM for each body). (remembered).

Furthermore, the parameter 5LOP (j) obtained by equation (5) is an amount roughly proportional to the contrast of the subject image, and the larger the value, the deeper the depression near the minimum value of the correlation amount C (j, L). This shows that the correlation is large, and therefore the reliability of the obtained defocus amount DEF(j) is high.

In addition, if the minimum value xm (j) is not found and the defocus amount DEF (j) cannot be determined, or if the defocus amount DEF (j) is found but 5LOP (j) is small and the reliability is low, It is determined that the focus cannot be detected and the defocus amount DEF (j)"ω is set.

FIG. 7 is a flowchart showing the operation of the focus detection process described above.

First, in step #95, the focus detection area is initialized by setting j=1. Focus detection tilt ft1AREA in step #lOO
A correlation calculation is performed in (j) to obtain the correlation amount C(j, L). xm (j) by three-point interpolation in step #1o5
Find +5LOP (j).

In step #110, it is determined whether xm(j) has been found, and if it has not been found, the process proceeds to step #125, and if it has been found, 5LOP(j) is set to the predetermined value TS.
If it is determined that it exceeds TS, that is, if it is determined that there is no reliability, the process proceeds to step #125, and if it is determined in step #115 that it is greater than TS.

The process proceeds to step #120, where the defocus amount DEF(j) is determined, and the process proceeds to step #130. Meanwhile, step #12
If you proceed to step 5, it is determined that focus cannot be detected and DEF (j
)=ω, and proceed to step #130. Step #13
0, the focus detection area is updated to the next area with j=j+1, and in step #135, j=9 (the focus detection area is finished).
Determine if j is not 9, and if j is not 9, proceed to step #
Return to 1oO and perform correlation calculation in the next focus detection area. If j = 9 in step #135, focus detection has been completed in all areas, so exit the focus detection process and perform grouping in step #140. Proceed to processing.

As described above, the amount of defocus can be determined in all focus detection areas. In the above description, the focus detection area shown in FIG. 5 is divided into eight parts, and the defocus amount of each area is detected, but the shape of the detection area and the number of divisions are not limited to this.

Furthermore, the detection areas may overlap each other, or the area boundaries may be changed depending on the intensity distribution of the subject image.

Grouping process> The grouping process groups multiple focus detection areas according to the multiple defocus amounts obtained by the focus detection process.
Group images that are likely to capture the same subject.

FIG. 8 is a flowchart showing one embodiment of the grouping process.

In step #200, first, as shown in FIG. 9, the focus detection areas are rearranged in ascending order (in order of closest distance) according to the amount of defocus of the area, and are numbered from 1 to k in order. However, focus cannot be detected (defocus amount DEF (j
)=O)) The area is not numbered, and if it is undetectable in all areas, the number k is set to O. Also, the final number is KL. Therefore, when KL=O, it means that focus cannot be detected in all areas.

In step #205, it is determined whether KL=O, that is, the focus cannot be detected in all regions. If it is not possible, the process proceeds to step #210, and the subsequent processing is canceled as it is determined that the focus cannot be detected in all the regions. to start the next focus detection operation. If it is not possible, in step #215, the size of the zone on the defocus axis used in the grouping process, that is, the depth of the image plane, is determined as follows.

Example 1: The depth of field changes according to the F value of the photographic optical system at the time of shooting, and when the F value is small, the depth is deep and the size of the zone ZONE that can be considered as the same subject increases, so the shooting F
Determine the size of the zone according to the value.

Example 2: Fix the size of the zone to a predetermined value so that the image plane that is considered to be the same subject always falls within the predetermined depth.

Example 3: The defocus amount obtained by the focus detection process described above has a certain range of uncertainty.

The width changes depending on the reliability 5LOP value of the defocus amount and the contrast of the subject image in that area. For example, the width of uncertainty is 5LOP confidence level.
The size of the zone is determined in proportion to the reciprocal of the 5LOP value of the area. By sizing the zones in this way, areas with a highly reliable amount of defocus are more likely to form a group independent of other areas, making it more likely that the area will The probability of focusing is increased. For example, if the predetermined zone width is ZN
Let the size of each zone be ZONE (j)=Z
Let it be N/5LOP (j). In this case, the upper and lower limits of the size of the zone may be set, and the contrast CON (j) may be calculated by calculating the sum of the absolute values of the differences between the output data of adjacent light-receiving elements within that area. This value may be used in place of the 5LOP value.

In step #22o, as initialization of the grouping process,
Group number i is set to 1, area defocus order number k is set to 1, and in step #225 group GRP (i
) to register the area to the number.

In step #23o, it is determined whether the number k has reached the final number KL, and if it has reached KL, the grouping process is ended and the process proceeds to the optimal group selection process in step #235. If it has not reached KL, the grouping process is continued. Continue and proceed to step #24o.

In step #240, if the size of the zone is determined as in Example 1 or Example 2, is the defocus amount of the area numbered +1 included in the zone on the infinite side from the defocus amount of the area numbered 1? judge.

For example, in the case shown in FIG. 10(a), the two areas are determined to be outside the zone and are placed in different groups, and in the case shown in FIG. 10(b), the two areas are determined to be inside the zone and placed in the same group. If the size of the zone is determined as in Example 3, the difference between the amount of defocus in the area with number 1 and the amount of defocus in the area with +1 in number is the size of the zone determined by the 5LOP value of the area with +1 in number 1. For example, in the case shown in Figure 11(a), the two areas are determined to be outside the zone and are placed in a separate group, and as shown in Figure 11(b).
In such a case, the two areas are determined to be within the zone and are in the same group.

If it is determined that the group is a different group in step #240, the group number is updated in step #245, and the group number is updated in step #245.
Proceed to step 0, and if it is determined that they are the same group, proceed to step #
245 is skipped and the process immediately proceeds to step #250. In step #250, the number k is updated, and then in step #22
Return to step 5.

Repeat the above process and when you get to step #235,
For example, areas are grouped as shown in FIG. That is, in the case of one embodiment of the grouping process, the spread of one group on the defocus axis is flexible, and if there are multiple areas within a group, one area belonging to that group is 10(b) or 11 for at least one other area where
It suffices if the relationship is as shown in Figure (b). This type of grouping allows the group to be spread out widely, so even if the focus position changes continuously over a wide area, such as a poster hung on a wall viewed from an angle, the entire subject can be captured in one image. Can be captured as a group. Of course, you could set a certain upper limit on the size of the group.

The group may be updated when the sum of the 5 LOP values of the areas added to the group becomes a constant value.

The grouping process described above is not limited to the above, but may be a grouping process as shown in Japanese Patent Application No. 63-331748 filed by the present applicant or other processes.

<Optimal Group Selection Process> In the optimal group selection process, one group that is likely to capture the subject intended by the photographer is selected from among the multiple groups created by the grouping process.

FIG. 12 is a flowchart of the optimal group selection process.

In step #60o, the area point E(j) for each focus detection area is determined as shown in Table 1 based on the information obtained from the focus detection mode selection device [80].

In other words, in the case of center-weighted mode, unless focus detection is impossible, the focus detection results are changed by giving a region point of 1 only to the region near the center of the entire focus detection region, and setting region points of other regions to O. I try not to let it affect me. In that case, the priority order of the areas is (area 4.
5) → (area 3.6) → (area 2, 7) → (area 1, 8
) is set.

In Table 1, if the central area cannot be detected, the area points are set so that the detection area moves to the edge areas sequentially, but it is also possible to detect only the central area or move the detection area. The area points may be set so as to terminate the process in the middle.

Other detection modes (proximity priority, reliability priority)

Focus priority), the area at the end of the column (area 1, 2, 7°8)
Give area points 0.5 to the central area (areas 3, 4,
5.6) is given 1 area point. In the edge area, there is a high possibility that the subject will move in and out due to camera shake, etc., which may cause the focus detection result to fluctuate, so the area point of the edge area is set lower than the area point of the solid area.

In step #605, group number i is initialized to 1. In step #610, a group defocus amount is calculated for each group from the defocus amount of the area belonging to that group as follows.

For example, areas 2 and 3.4 belong to group 1, and their group defocus amounts are each DEF (2).

When DEF (3) and DEF (4) are used, the group defocus amount GDEF (1) becomes as shown in equation (7).

GDE F(1)=(DE F(2)+DE F(3)
+DE F(4))/3 or GDE F(1) = (DE F(2) * E(2)+
DE F (3) * E (3) + DEF (4) Umbrella E (4)
)/(E(2)+E(3)+E(4))...(7) In other words, the group defocus amount is a simple average of the defocus amounts of areas belonging to the group or a weighted average weighted by area points. It is required as. However, if the sum of area points is O when calculated as a weighted average, the group defocus jlDEF (i) is made undetectable.

In step #615, the closest priority point P for each group
1 is determined based on the group defocus amount of that group. 6 For example, if the focus detection mode is the closest priority mode, the closest priority point is determined for the group with the maximum group defocus amount, that is, the closest one, as shown in Table 2-1. 1 is given, and the other groups are given points α (0<α×1) or O. The value of α is set smaller as the proximity priority is higher. In addition, in Table 2-1, other groups are given points of a predetermined value α, but the degree of closeness is calculated for each group according to the order of group defocus amount and the deviation from the nearest group. The closest priority point P1 may be changed and given.

In addition, in other detection modes (center-weighted, reliability priority 9 focusing priority), 1 as shown in Table 2-2, the group is given a closest priority point of 0 or 1 depending on whether the group defocus amount can be detected or not. 2 Therefore, the group If the amount of defocus can be detected, points will be given to the law, so each group will be treated fairly regardless of the value of the amount of defocus. It turns out.

By setting the closest priority point P1 as described above, the closest group is prioritized in the closest priority mode,
In other modes, the nearest group no longer takes priority.

In step #620, one focus priority point P2 is determined for each group based on the group defocus amount of that group. For example, when the focus detection mode is focus priority mode, as shown in Table 3-1, focus priority point 1 is given to the group with the smallest absolute value of the group defocus amount, that is, the group closest to focus, and the other groups are is point β
(0 × β × 1) or 0 is given. The value of β is set smaller as the focus priority is higher. In addition, in Table 3-1, points of a predetermined value β are given to other groups, but the degree of closeness to in-focus is calculated according to the magnitude of the absolute value of the group defocus amount and the order, and the degree of closeness to focus is calculated for each group. The focus priority points may be changed and given.

Table 3-1 Focusing priority points and other detection modes (center weighted, reliability priority, close proximity priority) can be determined in the same way as Table 3-1 above.Table 3-2
A focusing priority point of 0, γ, or 1 is given to a group according to the absolute value of the group defocus amount, as shown in FIG. However, γ is set to β<γ<1. In this way, even in modes other than the focus priority mode, by giving some priority to groups that are close to focus, stability near focus can be increased.

By setting the focus priority point P2 as described above, priority is given to the group closest to focus in the focus priority mode, and stability is improved in other modes as well.

In step #625, for each group, reliability points P
3 is determined based on the reliability 5 LOP (j) of the defocus amount of the area j belonging to the group. For example, if the focus detection mode is reliability priority mode, Table 4-1
The group with the highest sum of reliability is given reliability priority point 1, and the other groups are given reliability priority point 4-1. Points (0 < points minus 1) are given according to the amount.

There are also other detection modes (center weighted, closest priority).

In focus priority), reliability points O or 1 are given to a group only depending on whether the group defocus amount can be detected as shown in Table 4-2.

Therefore, if the group defocus amount can be detected, points are given uniformly, so each group is treated fairly regardless of the size of the sum of the reliability of the regions belonging to the group.

Furthermore, in other detection modes, reliability points may be given as shown in Table 4-3. In this case, if the sum of reliability in the group is greater than or equal to the predetermined value δ, 1 reliability point is given uniformly, and if it is less than δ, the sum of reliability is
Reliability points are given according to the amount divided by. In this way, even in modes other than reliability priority mode, groups with a certain degree of reliability or higher are treated fairly, and groups with low reliability are treated fairly.
By treating groups with weights according to their reliability, it is possible to reduce the influence of fluctuations in reliability points due to uncertainty in the amount of defocus.

By setting the reliability point P3 as described above, priority is given to the group with the highest reliability in the reliability priority mode, and stability is improved in other modes as well.

In step #630, the total point PG for each group is calculated as the product or sum of the closest priority point P1, the focusing priority point P2, and the reliability point P3. Group number i is updated in step #635, and step #640
Then, it is determined whether the group number i has reached KL+1, that is, whether the group number has ended. If the group number has not ended, the process returns to step #610 and continues to calculate the total points of the next group. When the above process is repeated and the group is completed, the process proceeds to step #645, and the group with the maximum total point PG (i) is selected as the optimum group, and the process proceeds to step #650, which calculates the optimum defocus amount.

In the optimal group selection process described above, multiple selection rules (
We set points according to the center weight, close focus 9 focus priority, reliability priority, etc.), determine the points for multiple selection rules in each group, and finally select the group with the maximum total points. , has the following advantages.

a) By appropriately changing the point value within the same rule, the effectiveness (degree of priority) of the selected rule can be easily controlled without modification of the program or other changes.

b) The strength (degree of priority) between one selected rule can be controlled by appropriately changing the point value ratio between different rules. Therefore, there is no need to provide a separate program for selecting each focus detection mode.

The focus detection mode can be changed simply by changing the point value.

C) Since points are set for each selection rule and the optimal group is selected by the product or sum of these points, different rules can be easily mixed. Also, new rules can be easily added. For example, when the size of an area is added as a selection rule, it is sufficient to set 1 area point setting and give a point value according to the number of areas belonging to a group. In addition, in order to prevent a specific pattern, for example, a so-called hollow pattern, the hollow is a point P.
5 is set, and pattern judgment is performed for hollowing out (determine whether end areas 1.2 and 7.8 of the horizontal row belong to the nearest group)
If the cutout is a pattern, the cutout in the nearest group will have only 1 point, and the cutout in other groups will have 0 points, and if the cutout is not a pattern, the cutout in all groups will be done. You can set the point to 1.

By introducing the point-based optimal group selection method as described above, the system can flexibly respond to mode changes, additions, and optimization of selection rules.

In the above explanation, one optimal group is ultimately selected, but the position of the area belonging to that group may be displayed in the viewfinder field of view, etc., so that the photographer can confirm the area in the selected group. In this case, if the subject intended by the photographer is not in the displayed area, the selected group is canceled and the group with the next highest total points is selected in order. You can.

Also, in the above explanation, various points distribution methods are explained.

The focus detection mode was changed according to the selected focus detection mode, but the ROM can be changed to the body by allowing the photographer to set the point directly or by setting the point using ROM data in which predetermined points are set in advance. The points may be changed by configuring and replacing the ROM. In addition, without fixing the point value, the characteristics of the finally selected group (area position, group defocus amount, contrast, etc.) are fed back,
A learning function may be provided to dynamically change point values to suit the photographer's intentions.

The above is an example of the optimal group selection process, and of course the optimal group may be selected using other methods6.For example, the absolute distance of each group can be calculated using the group defocus amount and the absolute position of the photographic lens. It is also possible to determine the optimal distance and group based on the difference between the closest or farthest distance and the closest and farthest distance.6 Also, the focus detection mode can be changed according to the distance distribution pattern of the area or group. In the optimal defocus amount calculation process, the defocus amount of the focus detection area belonging to the optimal group selected by the optimal group selection process is calculated. Based on this, the overall optimal defocus amount is calculated.

FIG. 13 is a flowchart showing the optimum defocus amount calculation process.

In step #900, the weight W (j) for the focus detection area AREA (j) belonging to the optimal group is determined. For example, the weight W(j) can be determined according to the focus detection mode as shown in Table 5. In the reliability priority mode, the focus detection area AREA (j
) is the weight W (j) of the area point E (j) and the reliability of the amount of defocus for that area 5LOP (j), and in other modes, E (j) umbrella 5LO
If P (j) is larger than the predetermined value V, the weight W (j)
is set to 1, and when it is smaller than a predetermined value V, the weight W(j) is set to E (j) *5LOP (j).By setting in this way, in the reliability priority mode, the weight W(j) is set to In other modes, if there is a certain level of reliability, the same weighting will be applied.

In step #905, the optimum defocus amount DEFX is calculated as a weighted average obtained by weighting the defocus amounts DEF (j) of the focus detection areas belonging to the optimum group with the weight W (j) as shown in equation (8).

0 (8) However, in equation 8), j is the number of the focus detection area belonging to the optimal group. By taking the weighted average in this way, variations in the optimum defocus amount can be reduced and stability can be improved.

In step #910, A
It controls the display form of the F display device 40, controls the amount of drive of the Af motor 50, and drives the photographing lens 11 to the in-focus position. As a result, a focused subject image is formed in the focus detection area belonging to the optimal group on the photographic screen. When the process of step #910 is completed, the next focus detection cycle is started again.

<Photometry calculation process> In the photometry calculation process, the photometry value of each area obtained by the photoelectric conversion device 134 for photometry is based.

The photometric values of each region are weighted based on the information obtained in the optimal group selection process, and the optimal photometric value is determined from the weighted average value. FIG. 14 is a flowchart showing photometric calculation processing.

First, in step #1000, set area number j to 1.

The photometric integrated values Q1 to Q4 (hereinafter abbreviated as Q1 to Q4) are initialized to 0, and the photometric index values R1 and R2 (hereinafter referred to as R1 and R2) are initialized to predetermined values. Next, in step 31005, it is determined whether the focus detection area AREA (j) (hereinafter referred to as AREA (j)) belongs to the optimal group. If it does not belong to the optimal group, proceed to step #1400 and calculate the photometric value Bv(j) of the photometric area corresponding to AREA(j).
)(below.

Bv(j)) is added to Q4 as a background photometric value. If AREA (j) belongs to the optimal group, the process advances to step #1010.

In step #1010, average mode photometry calculation is performed.

In other words, the weighting is given by the coefficient F(j) (hereinafter F(
j) = 1, and multiply By (j) by Ql
Add to.

Next, in step #1015, photometric calculation in center-weighted mode is performed. This gives importance to the photometric value in the center area of the screen in the optimal group, and, similar to the idea of grouping in focus detection, it has the effect of preventing the influence of unnecessary objects entering and exiting at both ends. Specifically, F shown in Table 6
The value of (j) is multiplied by By (j) according to the condition and added to Q2.

Next, in step $1020, a photometric calculation is performed in the wide-area subject weight mode. This is because when the optimum group is separated and the subject is a hollow part, the photometric value of the one that is reflected on the screen is given more importance. For example, when photographing a person in the shade of a tree, it is possible to prevent the effects of twigs from appearing in the photograph. Furthermore, similar to the center-weighted mode described above, the effect of preventing effects at both ends can also be obtained. Specifically, F (
By (j) is multiplied by the value of j) according to the condition and added to Q3.

Next, in step $1025, photometric calculation is performed using the highest brightness reference mode, which means that the highest value in BY (j) is set to R1.
It is something to remember. Similarly, in step #1030, Bv (j) is determined by photometric calculation in the lowest brightness reference mode.
The lowest value among them is stored in R2.

Table 6 Center-weighted mode weighting is a coefficient Table 7 Wide-area subject weighting mode weighting is a coefficient The area number is updated in step $1035, and the area number is updated in step $1.
At 040, it is determined whether the area number has reached 9, that is, whether the area number has ended, and if it has not ended, step #
Return to step #1005 and continue processing the next area. Repeat 0 or more processes to process the next area. If the area is finished, proceed to step #10.
Proceed to normalization of photometric values in step 45.

Here, the integrated photometric value is normalized as the photometric value of each mode. First, let's take Ql as an example.

When Bv (2), By (3) and By (5) are integrated, the photometric value in average mode is: By (Ql) = (Bv (2) +Bv (3) +B
The 0 denominator for obtaining v (5) )/3=Q1/3 is the number of integrated data, and the same applies to By(Q4).

In the case of Bv (Q2) and By (Q3), each data is multiplied by the coefficient F (j), so Bv (Q2) = Q2/ (F (2) + F (3) + F (5
)) and normalize.

For Bv (R1) and By (R2), By (R1) = R1, taking into account the latitude of each film.
−1,5Xb By (R2)=R2+1.5Xc. Here, c is a correction coefficient depending on the type of film, and is determined by, for example, a signal from a DX code discrimination circuit (not shown).

Next, the optimal photometric value B vans(
Hereinafter, proceed to the calculation of B vans). Here, B vans is calculated based on the photometric values of each mode calculated as described above. That is, By (Ql) ~Bv (Q4)
The values of By (R1) and By (R2) are set based on a preset optimal weighting rate, or by changing the weighting rate according to the setting of the photometry mode selection device 180.
Determine ns. Bvans is as follows.

Bvans = (QXBv (Ql) + mXBv (
Q2)+nXBv (Q3)+oXBv (Q4)+p
XBv (R1)+qXBv (R2))/(Q+
m+n+o+p+q) For example, if it is set as 1 and m=q=O, average mode photometry will be used. Further, if Q=1, O=0.25 and other values are set to =O, the average light metering mode takes into account the out-of-focus area.

In addition, it is also possible to consider the background photometry value by adding the lowest brightness reference mode to the wide-area subject weighted mode, or by adding the highest brightness reference mode to the center weighted mode. In this case, if the auxiliary mode is set to about 0.7 to 0.35, and the background photometric value is set to about 0.25 to 0, the characteristics of each photometric mode can be utilized.

In addition, each weighting rate may be changed depending on the value of R1, which is the highest brightness data, or the weighting rate may be further changed depending on the value of the maximum and minimum brightness difference between R1 and R2, or information on focus detection may be added. A learning function may be provided to dynamically change the weighting rate to match the photographer's intention by feeding back the characteristics of the photographic subject (characteristics of the group, changes over time of each part, etc.).

The photometric calculation process described above may determine the optimum photometric value in accordance with a method other than this. For example, it is also possible to add a close-range priority photometry mode in which the weighting rate is gradually decreased in order from the closest areas based on the defocus value of each area within the optimum group. Further, it is also possible to set a predetermined limit value for the maximum photometric value of each area, and substitute the limit value for those exceeding that value to prevent the influence of direct reflection of sunlight.

Second Embodiment Next, a second embodiment in which the focus detection photometer according to the present invention is applied to an interchangeable lens type eye reflex system will be described with reference to FIG. 15. Components similar to those in the first embodiment shown in FIG. 2 are designated by the same reference numerals, and description thereof will be omitted.

In this embodiment, the photometric value of each area is directly obtained from the focus detection light receiving element output data. In other words, the optical conversion element for focus detection and the photoelectric conversion element for photometry are shared.

First, the light receiving element Ap obtained through AD conversion by the AFCPU 30
, Bp (p=l to n), and the charge accumulation time are obtained from the ports P1 and P2. The value divided by is multiplied by a predetermined coefficient d to compensate for the loss of the optical system, and the result is logarithmically compressed.
Obtain photometric values for each element.

Bv (ap)=log, ((ap/1)Xd)Bv
(bp)=log, ((bp/1)Xd) Note that the value of d may be a function of P or a variable based on other factors due to the characteristics of the optical system or the convenience of photometry.

It can also be done as follows.

By(ap)=log2((ap/1)Xd(ap))
By(b p)=1 o gz ((b p/1)Xd
(b p)) Next, the average value for each area is calculated according to the area division. For example, when the shift scale is O as in the first embodiment, the photometric value BV (1) in the area j = 1 is B v (1) = (B y (a 1) + · - + B
v(at)+B v(b l) +-+ B v (
b t, ) / (2X t). Bv
The same applies to (2) and thereafter.

Here, By (ap) and By (bp) are photometric values of secondary images created by light fluxes that have passed through different exit pupil regions of the photographic lens 11, and are the same because they are photometric values of images of the same subject. Therefore, the photometric value By (1).

B y(1)"(B y(a l)+...+ B v
(a t)/t or.

B v(1)"(B y(b 1)+-・+B v(b
t)/t. The same applies to By (2) and thereafter.

Based on Bv(j) thus obtained, the calculations described in the first embodiment may be performed thereafter.

Note that the second embodiment is not limited to the above-mentioned processing, and other methods may be used. For example, when calculating Bv(j), By(ap) or B
Alternatively, y(bp) may be averaged as it is as an antilog, and then logarithmically compressed.

Further, R1. described in the first embodiment. By(j
) instead of By (ap) or By (hp) data may be used in this case.

This method has the advantage of being able to obtain more detailed data than data for each region, and more accurately obtaining luminance information of the subject.

G1 As described in detail of the invention, the focus detection device of the present invention collects the focus detection results of a plurality of focus detection photometry areas into groups that are considered to be the same subject, and selects the one that most matches the photographer's intention from among the groups. Select a group that is thought to be present, perform statistical averaging among the selected groups, and obtain the final focus detection result. At the same time, focus on the selected group and perform calculations to obtain the final photometric value. This is what you get.

By grouping in this way, even if multiple subjects exist in the focus detection area on the screen, they can be accurately identified, and by accurately identifying the subjects, optimal group selection processing can be performed. The selection result will be more suitable for the photographer's intention. Furthermore, since the optimum defocus amount is determined based on the defocus amount determined within a group that is somewhat coherent, variations are reduced and the results are not far different from the individual results within the group.

Furthermore, it is possible to calculate the photometric value of the main subject that matches the photographer's intention by focusing on the photometric value of the optimal group corresponding to the main subject and using a processing method that most closely matches the situation of the main subject. It is. Therefore, both the focus detection result and the S light result match the photographer's intention, and the photographing result is also good.

In addition, in one or more embodiments, AFCPU3o and AEcPtJl
Although oo is assumed to be separate, it is of course possible to replace each process with a single CPU that can process each process in chronological order or in parallel.

[Brief explanation of the drawing]

Fig. 1 is a diagram corresponding to claims, Fig. 2 is a block diagram of a focus detection photometer according to the present invention, Fig. 3 is an explanatory diagram of a focus detection optical system used in the present invention, and Fig. 4 is a block diagram of a photoelectric conversion device. ,
Fig. 5A is an explanatory diagram of the focus detection area, Fig. 5B is an explanatory diagram of the photometry area, Figs. 6 and 7 are an explanatory diagram and flowchart of the focus detection process, and Figs. 8 to 11 are an explanatory diagram of the grouping process. Figure 8 is a flowchart and Figure 9 is a flowchart.
10 and 11 are explanatory diagrams, FIG. 12 is a flowchart of optimal group selection processing, FIG. 13 is a flowchart of optimal defocus amount calculation processing, FIG. 14 is a flowchart of photometry calculation processing, and FIG. 15 is an overall configuration diagram of the second embodiment. 10: Photographing lens 13: Lens CPU 20: Camera body 24: Focus detection optical system 25: Photoelectric conversion device 30: AFCPU 40: AF display unit 50
:AF motor 8o: Focus detection mode selection device

Claims (1)

[Claims] 1) A photographing optical system for forming a subject image on a predetermined plane, and a plurality of focus detection areas set within the photographing screen,
focus detection means for detecting the amount of defocus of the current image plane with respect to each of the predetermined planes; photometry means for obtaining a photometry output for each area substantially the same as the plurality of areas; a grouping means for grouping into a plurality of groups according to the amount of focus; a selection means for selecting one optimal group in which an image of a subject to be focused is expected to be formed from among the plurality of groups; an optimal defocus amount calculating means for calculating a defocus amount based on a defocus amount of a focus detection area belonging to an optimal group; What is claimed is: 1. A focus detection photometry device comprising: a photometry calculation means for obtaining a photometry. 2) A focus detection photometry device according to claim 1, wherein the photometry means and the focus detection means have a common photoelectric conversion element.