JP2671489B2 - Focus detection device - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a focus detection device for an optical device such as a camera that detects a focus adjustment state of an imaging optical system that forms an optical image of a subject.

[Conventional technology]

Conventionally, two light images of almost the same part of the subject are
Leading on a pair of photoelectric conversion element arrays consisting of an array of a plurality of photoelectric conversion elements, calculating the output of this photoelectric conversion element array to detect the relative displacement of the optical image on the photoelectric conversion element array, the detection result 2. Description of the Related Art There is an automatic focus detection device that detects the focus from the camera and adjusts the focus of a photographing lens.

Further, as this type of automatic focus detection device, a width of the focusing range is provided with hysteresis to prevent slight vibration of the taking lens and to perform stable photographing (JP-A-59-340).
However, if the contrast of the subject is low, the focus detection of the subject becomes unstable and the operation may become unstable.

In consideration of such a case, there is a method in which a stable focus detection is performed by changing the determination level of the detection depending on whether or not the focus detection is possible in the previous detection result (JP-A-60-247210). Gazette).

[Problems to be solved by the invention]

However, in the conventional automatic focus detection device, the determination level is set only in one focus detection block,
If the previous focus detection cannot be performed by the focus detection block, the focus detection determination of the subject may become unstable.

Therefore, there is a problem that stable focus detection cannot be performed.

An object of the present invention is to provide a focus detection device that can perform stable focus detection.

[Means for solving the problem]

In order to achieve the above-mentioned object, the present invention is a focus detection device that detects a focus state based on an optical image of a subject incident on a focus detection region, in a plurality of regions obtained by dividing the focus detection region, for each region. A determination level setting means for setting a determination level, and a determination for each area to determine whether focus detection is possible by comparing the set determination level with a calculation result based on the data in each area. Means, storage means for storing the determination result, and changing means for changing the determination level according to the stored previous determination result.

Further, in claim 2, the focus detection device includes a light receiving element array including a plurality of photoelectric conversion elements, and the determination level is changed according to a contrast value obtained by summing up differences of data of adjacent photoelectric conversion elements. I did it.

Further, in claim 3, the focus detection device includes a light receiving element array including a plurality of photoelectric conversion elements that receive a pair of optical images obtained from an optical image of an incident subject, and the determination levels are adjacent to each other. The contrast value, which is the sum of the differences of the data of the conversion elements, and the best correlation value of the correlation values obtained by the mutual comparison of the pair of optical images are changed.

In the fourth aspect, the determination level at the first focus detection operation is set to a lower level than when the focus detection operation is determined to be impossible last time after the second focus detection operation.

(Operation) According to the focus detection device having the above configuration, the focus detection area is divided into a plurality of areas, and it is premised that the calculation for focus detection and the like can be performed in each area. The determination level of is changed based on the previous determination result, and the determination level for each area is compared with the calculation result based on the data of the area to determine whether the focus can be detected.

Further, the determination level is changed according to the contrast value obtained by summing the differences between the data of the adjacent photoelectric conversion elements.

Further, the determination level is also changed according to the ratio between the contrast value obtained by summing the differences between the data of the adjacent photoelectric conversion elements and the best correlation value of the correlation values obtained by the mutual comparison of the pair of optical images.

Further, in the first focus detection operation, the determination level is relaxed after the second and subsequent times when it is determined that focus detection cannot be performed the previous time.

〔Example〕

FIG. 1 shows an embodiment of a schematic configuration of a camera provided with an automatic focus detection device according to the present invention. The automatic focus detecting device 1 includes a photographing lens 2, focus detecting means 3, defocus amount determining means 4, lens position detecting means 5, lens driving means 6, and control means 7. The photographing lens 2 guides a light beam 9 emitted from the subject 8 to the focus detecting means 3. The focus detection means 3 has a photoelectric conversion element array (not shown), and processes the output of this photoelectric conversion element array to detect the relative displacement of the optical image by the subject 8 on the photoelectric conversion element array. is there.

Further, the defocus amount determining means 4 determines whether the photographing lens 2 is based on the displacement detection signal output from the focus detecting means 3.
Is determined. The lens position detecting means 5 detects the current position of the taking lens 2. The lens driving unit 6 drives the photographing lens 2 based on the defocus amount signal of the defocus amount determining unit 4. The control means 7 controls the lens driving means 6 so that the photographing lens 2 is driven to a position corresponding to the defocus amount, and when the focus detection means 3 outputs a focusing signal, the photographing lens 2 Is stopped.

FIG. 2 shows an embodiment of the mechanism configuration of the focus detecting means 3. The focus detecting means 3 includes a main mirror 10, a sub mirror 11, and a focus detecting optical system 12. This primary mirror 10
Is for branching the light beam 9 that has passed through the taking lens 2 into a finder optical system (not shown) and a sub-mirror 11. The sub-mirror 11 further splits the light beam 9 split by the main mirror 10 into a focus detection optical system 12 and a film surface 13.

The focus detection optical system 12 is a photoelectric conversion element array 141, 142,
143, separator lens 15, aperture mask 16, module mirror 17, condenser lenses 181, 182, 183 and field stop 19
It is composed of Photoelectric conversion element arrays 141, 142, 14
Reference numeral 3 denotes an arrangement of a plurality of CCD image pickup devices and the like, which are formed on one chip on the focal plane 14 of the separator lens 15. The separator lens 15 is a separator lens pair 151, 152,
The photoelectric conversion element array 14 that has the branched light flux 9
1, 142, 143. The aperture mask 16 has circular or oblong openings 161, 162, and 163, and limits the light flux 9 input to the separator lens 15. The module mirror 17 guides the light beam 9 passing through the condenser lenses 181, 182, 183 to the separator lens 15.
The field stop 19 has rectangular openings 191, 192, and 193 and is disposed near the focal plane, and limits the field of view of the light beam 9 input to the focus detection optical system 12.

FIG. 3 shows a light receiving section of the photoelectric conversion element arrays 141, 142, 143 (hereinafter, referred to as a CCD including a light receiving section, a storage section, and a transfer section of the photoelectric conversion element). The photoelectric conversion element array 141 includes a reference part 141a, a reference part 141b, and a photoelectric sensor 141c. Similarly, the photoelectric conversion element array 142 includes a reference part 142a.
The photoelectric conversion element array 143 includes a reference portion 143a, a reference portion 143b, and a photoelectric sensor 143.
c. The photoelectric sensors 141c, 142c, and 143c are disposed on one side of the reference portions 141a, 142a, and 143a, respectively, in the longitudinal direction, and output signals according to the amount of received light. The accumulation time of the CCD in the accumulation unit is controlled by a signal corresponding to the amount of received light. As shown in FIG. 4, the photoelectric conversion element arrays 141, 142, and 143 are focus detection areas 21, 22, and 23 (hereinafter, referred to as first islands 21 and 22, respectively) located at the center of the photographing screen 20 in the viewfinder.
The displacement detection signal is output based on the light image of the subject 8 projected on the second island 22 and the third island 23).

A region 24 indicated by a dotted line in the center of the photographing screen 20 displays a region where focus detection is performed for the photographer. The display unit 25 is turned on when focusing is performed.

The number of pixels X of the reference portions 141a, 142a, 143a and the number of pixels Y of the reference portions 141b, 142b, 143b are, for example, as shown in FIG. In this example, the number of pixels X in the reference portion is 34, and the number of pixels Y in the reference portion is 44.
In 142, the number of pixels X in the reference portion is 44 and the number of pixels Y in the reference portion is Y
Consists of 52 pieces.

In the automatic focus detection device 1, the reference portions 141a, 142a, 143a
Is divided into a plurality of blocks, and each of the divided blocks is compared with all the data of the reference units 141b, 142b, and 143b to perform focus detection. Then, among the focus detection results in each block, the data of the rearmost pin is used as the focus detection data of each of the islands 21, 22, and 23, and the focus detection data of each of the islands 21, 22, and The photographing lens 2 is driven based on the data of the photographing magnification to bring the lens into focus.

Next, a range (a defocus range) for detecting the amount of defocus (that is, the amount of defocus in the optical axis direction between the planned focal plane of the imaging optical system and the object image plane) from the photoelectric conversion element arrays 141, 142, and 143. Will be described with reference to FIGS. 5 to 7.

FIG. 5 is an enlarged view of the difference data area of the reference portions 141a, 142a, 143a corresponding to the islands 21, 22, 23 shown in FIG. In the figure, each of the reference portions 141a, 142a,
The numerical value shown in 143a is the reference part 141a, 1 shown in FIG.
Data for each pixel 42a and 143a is detected every third pixel, and the number of differences (that is, spatial frequency components) obtained by taking the difference between them is shown. Further, this difference data indicates a specific spatial frequency component extracted (film ring) from the pixel data (first spatial frequency component extraction means). The difference data is two or one.
It may be every other. However, the spatial frequency components are different from every third spatial frequency component, and the numerical values are also different.

The numerical values in the reference portions 141a, 142a, 143a of the islands 21, 22, 23 are 30 in the first island 21, 40 in the second island 22, and 30 in the third island 23. In addition, the reference parts 141a, 142a,
It is obtained in the same manner as the numerical value of the difference data of 143a. That is,
The numerical values of the reference portions 141b, 142b, 143b are
Forty, the second island 22 has 48, and the third island 23 has forty.

Next, block division in each of the islands 21, 22, and 23 will be described. That is, the first island 21 is divided into two blocks, K1 (1 to 20) and K1 (11 to 30), from the difference data at the upper end, and sets them as a first block BL1 and a second block BL2, respectively. In the second island 22, three of K2 (1 to 20), K2 (11 to 30), and K2 (21 to 40) are obtained from the left end difference data.
Divided into three blocks, the third block BL3 and the fourth block, respectively.
Block BL4 and fifth block BL5. Third island
In 23, K3 (1-20), K3 (11-
30) divided into two blocks, and the ninth block BL
9, and the tenth block BL10.

Further, the second island 22 detects difference data whose extraction frequency has been changed in order to focus on a subject composed of low-frequency components (second spatial frequency component extraction means), and converts the difference data into sixth block BL6, 7th block BL7 and 8th block
It is divided into blocks BL8 and used for focus detection data. That is, the data of each pixel of the reference unit 142a and the reference unit 142b is, for example, a difference is obtained every seven, and the sum of the difference data is obtained between adjacent data. The sum data is used as the sixth block BL6. That is, the number of difference data every seventh is 36 in the reference unit 142a and 44 in the reference unit 142b.
The number of data in the sixth block BL6 is 35 in the reference unit 142a,
In the reference part 142b, the number is 43. And this reference part 142a
The 25 blocks on the left side are the seventh block BL7, and the 25 blocks on the right side are the eighth block BL8.

In the above description, the intervals of the difference are set at seven intervals, but the larger the interval, the more low frequency components can be extracted.
That is, the intervals of the differences may be intervals other than every seven intervals.

In this focus detecting means, the rear focus is set when the image interval when the optical images of the reference portion and the reference portion match is larger than a predetermined interval, the front focus is set when the image interval is smaller than the predetermined interval, and the focus is set at the predetermined interval. Therefore, the defocus range in the divided blocks is such that the blocks farther from the optical center in each island cover the rear focus side.

FIG. 6 shows the reference portion 142a and the reference portion of the second island 22.
The actually used area of 142b is enlarged and shown, and the defocus range of the block BL4 divided into blocks will be described with reference to FIG. That is, in order to focus on the image of the fourth block BL4 of the reference unit 142a, the data of the block BL41 located at the center of the reference unit 142b, that is, the data of the image located at the 15th to 34th positions from the left end of the reference unit 142b and This is when the data K2 (11 to 30) of the four blocks BL4 match. And
From this, when the image coincidence is on the left side of the reference portion 142b, it becomes the front pin, and the maximum displacement data number of the front pin (hereinafter referred to as displacement pitch) becomes 14. On the other hand, if the image coincides with the right side of the reference portion 142b, the rear focus is set, and the maximum shift pitch of the rear focus is 14 pieces.

The same applies to the first island 21 and the third island 23. That is, as shown in FIG. 7, in the third block BL3, the front pin side shift pitch is 4, and the rear pin side shift pitch is 24. In the fifth block BL5,
The front pin side shift pitch is 24, and the rear pin side shift pitch is 4. Further, the first block BL1 and the ninth block BL9
In the example, the front pin side shift pitch is 5 and the rear pin side shift pitch is 15, and the second block BL2 and the tenth block BL1
In the case of 0, the front pin side shift pitch is 15, and the rear pin side shift pitch is 5. Further, in the sixth block BL6, there are four shift pitches on both the front pin side and the rear pin side.

In the seventh block BL7, the front-pin-side shift pitch is four,
The rear pin side shift pitch is 14, and in the eighth block BL8, the front pin side shift pitch is 14, and the rear pin side shift pitch is 4, but since the shift pitch overlaps with the sixth block BL6, In the seventh block BL7, 4 to 14 shift pitches on the rear pin side are used.
The defocus range is set to the shift pitches of 14 to 14 pieces.

FIG. 8 shows an embodiment of a camera circuit block in which the automatic focus detection device according to the present invention is configured using a microcomputer.

A central control circuit (hereinafter, referred to as a microcomputer) 26 controls the entire camera, calculates exposure and the like, extracts a first spatial frequency component (primary filter means), and focuses on the basis of the extracted spatial frequency component. Detection (first focus detection means), and further, a second spatial frequency component is extracted based on the extracted spatial frequency component (secondary filter means), and the spatial frequency extracted by the secondary filter means is extracted. Focus detection is performed based on the component (second focus detection means), and when the first focus detection means determines that the focus cannot be detected, the secondary filter means and the second focus detection means are operated. (Determination means). Lens circuit 27
Denotes a photographing lens 2 mounted on a camera body (not shown)
(Interchangeable lens) This is for transmitting data unique to the microcomputer 26. The focus detection circuit 28 corresponds to the focus detection means 3 in FIG. 1 and converts the analog output of the photoelectric conversion element array into a digital output and transmits it to the microcomputer 26. The brightness detection circuit 29 detects the brightness of the subject 8 by measuring the light flux 9 that has passed through the taking lens 2, and transmits the apex value BV0 corresponding to this brightness to the microcomputer 26. The film sensitivity reading circuit 30 reads the film sensitivity from a film (not shown) and transmits the apex value SV corresponding to the film sensitivity to the microcomputer 26.

The display circuit 31 displays the exposure information and the focus state of the taking lens 2. The encoder 32 detects the rotation amount of the lens drive motor 33 and outputs a pulse number corresponding to a predetermined rotation amount of the lens drive motor 33 to a lens drive control circuit 34 described later. The lens drive control circuit 34 is a microcomputer
The motor drive direction signal and the motor stop signal are input from 26, and the drive of the lens drive motor 33 is controlled based on these signals.

Further, a counter is provided inside the microcomputer 26. This counter detects the extension position of the photographing lens 2 from the infinity position, and counts up or down with respect to the pulse number. In addition, a power switch (main switch) described later
SW ₁ is turned on, further, when being driven taking lens 2 is the taking lens 2 has been retracted to the infinity position, the counter is adapted to be reset.

The power supply battery 35 directly supplies power to the microcomputer 26 and switches described later, and the other circuits include a power supply circuit described later.
The power is supplied via 36. The power supply circuit 36 supplies power to circuits such as the lens circuit 27 and the focus detection data output circuit 28 based on a control signal from the microcomputer 26.

Power switch SW ₁ is intended to start or stop the operation of the camera by being opened and closed. One-shot circuit 37, in conjunction with the opening and closing of the power switch SW ₁ generates a predetermined pulse, the pulse microcomputer 26
Is to be entered. Shooting preparation switch SW ₂ is intended to be opened and closed by operation of the release switch (not shown), when the shooting preparation switch SW ₂ is turned on, so that the focus detection operation is started. Limit switch
SW ₃ is turned on when the taking lens 2 is kneaded to the infinity position or when it is extended to the leading edge. The selection switch SW ₄ selects an automatic focus adjustment mode (hereinafter, referred to as AF mode) or a focus detection display mode (hereinafter, referred to as FA mode). When the selection switch SW ₄ is turned on, the FA mode is set. When turned off, the camera enters AF mode.

 Next, the operation of the camera having the above configuration will be described.

First, the power switch SW ₁ is once turned on, the one-shot circuit 37 outputs pulses to the interrupt input terminal INTO of the microcomputer 26, the microcomputer 26 executes the flowchart of the interrupt operation shown in FIG. 9.

That is, the microcomputer 26 prohibits the focus detection operation by the on the shooting preparation switch SW ₂ (step # 1), or not this interruption operation due to the on power switch SW _1, or by off interrupt input of the microcomputer 26 The determination is made based on the voltage input to the terminal IP1 (step # 2). And
If this input voltage is a high voltage, steps # 3 to # 11
Stopping the power to the lens circuit 27 etc. without performing the procedure of
The microcomputer 26 enters an operation standby state (steps # 12 and # 1).
3). If this input voltage is a low voltage, the power switch S
W ₁ determines the ON state, and prohibits the operation of the counter in the microcomputer 26 (Step # 3), the flag to be used for execution of the flowchart, and initially sets the output terminal such microcomputer 26 (Step # 4).

Next, the output terminal OP1 of the microcomputer 26 is set to a high voltage to supply power to circuits such as the lens circuit 27 (step #
5). Next, the retraction signal of the photographing lens 2 is input to the lens drive control circuit 34, and the photographing lens 2 is retracted in the direction of the infinite position (step # 6). Then, the photographing lens 2 is retracted to the infinity position, and the limit switch SW ₃
Is turned on (step # 7). When the limit switch SW ₃ is turned on at step # 7, it stops the driving of the photographic lens 2 (step # 8). Along with this, the counter in the microcomputer 26 is reset and the number of pulses corresponding to the rotation amount of the lens drive motor 33 is counted (steps # 9, # 10). And shooting preparation switch
The interruption of the focus detection operation by turning on SW ₂ is enabled (step # 11).

Then, the lens circuit 27
It stops power to such, to wait the operation to the photographing preparation switch SW ₂ and the like is depressed (step # 12, # 13).

The photographing preparation switch SW ₂ when the operation waiting state of the step # 13 when it is turned on, the microcomputer 26 executes the flowchart of the interrupt operation shown in FIG. 10.

That is, the microcomputer 26 initially sets flags used for execution of the flowchart and output terminals of the microcomputer 26 (step # 21). Further, the timer provided in the microcomputer 26 is reset, and then this timer is started (step # 22). Then, a flag AFSF indicating that this is the first focus adjustment operation is set (step # 23), and the output terminal OP1 is set to a high voltage to set the lens circuit 27.
And the like (Step # 24). Next, data including a coefficient for converting the focal length data of the photographing lens 2, the open aperture value, the defocus amount, and the like into a pulse number for driving the lens is input from the lens circuit 27 (step # 25). Then, the sensor of the focus detection circuit 28 is driven to integrate the subject light, and this pixel data is captured and stored (steps # 26, # 27). Next, the defocus amount of each of the islands 21, 22, 23 described above is calculated, the exposure calculation is further performed, and the calculation results are displayed on the display circuit 31 (steps # 28 to # 30).

Next, it is determined whether the mode is the FA mode or the AF mode (step # 31). If the mode is the AF mode, the driving amount of the photographing lens 2 is calculated from the defocus amount, and the photographing lens 2 is driven based on the calculated amount. (Step # 32). On the other hand, if the mode is the FA mode in step # 31, the process proceeds to step # 34 without performing the process of step # 32. At step # 34, it determines the opening and closing of the shooting preparation switch SW _2. Then, if photographing preparation switch SW ₂ is turned on, and resets the flag AFSF (step # 35), the process returns to step # 25, step #
Repeat the process from 25. On the other hand, if the shooting preparation switch SW ₂ is turned off at step # 34, turns off the power supply circuit 36 stops power to the lens circuit 27 and the like (step # 36),
Shooting preparation switch SW ₂ and the like to wait operation until pressed (step # 37). When the shooting preparation switch SW ₂ and the like is depressed, the process proceeds to the flow chart of Figure 9.

Next, each island 2 shown in step # 28 of FIG.
A defocus amount calculation subroutine of 1, 22, 23 will be described with reference to FIGS. 11 to 14.

FIG. 11 shows the defocus amount of each of the islands 21, 22, and 23 as the first island 21 (step # 41) and the second island 22.
(Step # 41), Third Island 23 (Step # 41)
12 to 14 show a specific flowchart of the defocus amount calculation of each of the islands 21, 22, and 23.

FIG. 12 shows a subroutine for calculating the defocus amount of the first island 21 (step # 41). As described above, the first island 21 includes the first block BL1 and the second block BL1.
A predetermined value "-K" is stored in variables DF1 and DF2 which are divided into blocks BL2 and store the respective defocus amounts of these blocks BL1 and BL2 (steps # 51 and # 52). This predetermined value “−K” is a value of a front focus state that cannot be obtained in each of the blocks BL1 and BL2, and is output as a defocus amount when focus detection cannot be performed on the first island 21 (hereinafter referred to as low contrast). You.

Next, the flag LCF1 indicating the state of the low contrast on the first island 21 is set (step # 53). Then, the focus state (front focus state, rear focus state, focus state) of the first block BL1 is detected and the defocus amount DF is calculated (step # 54). If focus detection is possible from this calculation result,
The flag LCF1 is reset, the obtained defocus amount DF is stored in the variable DF1 that stores the defocus amount of the first block BL1 (steps # 55 to # 57), and the process proceeds to step # 58.

On the other hand, if it is determined in step # 55 that focus detection cannot be performed, step # 58 is skipped without performing steps # 56 and # 57.
Move to

Next, the focus state of the second block BL2 is detected and the defocus amount DF is calculated (step # 58). If focus detection is impossible from the calculation result, it is determined whether the flag LCF1 is set. When the flag LCF1 is reset, that is, when the focus detection is enabled in the determination in step # 55, the process proceeds to step # 62.
Move to When the flag LCF1 is set in step # 60, the defocus amount calculation subroutine for the second island 22 in step # 42 shown in FIG.
(Figure).

On the other hand, if it is determined in step # 59 that focus detection is possible, the process in step # 60 is not performed, the flag LCF1 is reset in step # 69, and the calculated defocus amount is stored in DF2 in step # 70. Then, the process proceeds to step # 62.

In step # 62, an inter-block defocus amount (average processing width) ΔDF used in an averaging processing routine described later is determined. Next, the defocus amount DF1 and the defocus amount DF2
The defocus amount of the subject closer to the photographing lens 2 (camera) is determined as the defocus amount DFIS1 of the first island 21. That is, the defocus amount DF1 is equal to the defocus amount DF2.
If the defocus amount DF1 is larger than the defocus amount DF2, the defocus amount DF2 is set to the defocus amount DFIS1 of the first island 21. Defocus amount DFI
S1 is set (steps # 63 to # 67).

When the difference between the defocus amount DF1 and the defocus amount DF2 is smaller than the average processing width ΔDF in steps # 64 and # 65, the defocus amount DF1 and the defocus amount DF2 are averaged, and the average value is calculated. And the defocus amount DFIS1 of the first island 21 (step # 68). And step # 6
After completing the processing of 6, # 67 and # 68, step # 42 in FIG.
Move to

FIG. 13 shows a subroutine for calculating the defocus amount of the second island 22.

First, a variable DF for storing the defocus amount of each of the third block BL3, the fourth block BL4, and the fifth block BL5.
A predetermined value "-K" is set to 3, DF4 and DF5 (steps # 71 to # 73), and a flag LCF2 indicating the state of the low contrast on the second island 22 is set (step # 74). Then, the focus state detection and the defocus amount DF of each block BL3, BL4, BL5 are calculated (steps # 75, # 79, # 8).
3). That is, if the focus can be detected from the calculation result of the defocus amount DF of the third block BL3 in step # 75, the flag LCF2 is reset and the obtained defocus amount DF
Is stored in the variable DF3 (Steps # 76 to # 78), and the routine goes to Step # 79. Conversely, if it is determined that focus detection cannot be performed, the processing in steps # 77 and # 78 is not performed and step # 79 is performed.
Move to

Similarly, for the defocus amount DF of the fourth block BL4, if the focus can be detected from the calculation result of step # 79, the flag LCF2 is reset, and the defocus amount DF is stored in the variable DF4 (steps # 80 to ##). 82), proceed to step # 83. Conversely, if it is determined that focus detection is impossible, step #
The process proceeds to step # 83 without performing the processes of 81 and # 82.

Similarly, if the defocus amount DF of the fifth block BL5 can be detected from the calculation result of step # 83, the flag LCF2 is reset, and the defocus amount DF is stored in the variable DF5 (steps # 84 to # 84). 86), proceed to step # 87. Conversely, if it is determined that focus detection is impossible, step #
The process proceeds to step # 87 without performing the processes of 85 and # 86.

In step # 87, it is determined whether or not the low contrast flag LCF2 is set. When the flag LCF2 is not set, that is, in steps # 76, # 80, #
When it is determined that the focus can be detected in each of the determinations in step 84, the average processing width Δ used in the averaging processing routine described below is used.
Determine the DF (step # 88). Next, each block BL
The magnitude of each of the defocus amounts DF3, DF4, DF5 of 3, BL4, BL5 is determined, and the largest defocus amount MAXDF is extracted (step # 89). If there is no other block whose difference from the defocus amount MAXDF is smaller than the average processing width ΔDF, the defocus amount MAXDF is set to the second island 22.
(Steps # 90 and # 91), and if there is at least one other block in which the difference is smaller than the average processing width ΔDF, the defocus amount MAXDF and the difference are the average processing width. Averaging is performed only with the defocus amounts of other blocks smaller than ΔDF (average processing), and the average value is used as the defocus amount DFIS2 of the second island 22 (step # 92). When the processes of steps # 91 and # 92 are completed, the process proceeds to the defocus amount calculation subroutine (FIG. 14) for the third island 23 in step # 43 shown in FIG.

When it is determined in step # 87 that the flag LCF2 is set, every third difference data is re-organized into every seventh difference data in order to focus on a subject having low frequency components (step # 87). # 93). That is, for example, if the pixel data is l ₁ , l ₂ , l _n , ..., The difference data for _every three pixels is dD _n = l ₁ -l ₅ , ..., l ₅ -l ₉ , ..., l _n −
l _{n + 4} , ... Also, every 7th difference data is dD _m =
l ₁ −l ₉ , ..., L _m −l _{m + 8} ,. This every 7th difference data is obtained by taking the sum of every 3rd difference data dD _n for every 3rd dD _m . In other words, the difference data of 7 every _{is, dD m = dD 1 + dD} 5, ..., dD m + dD m + 4, ... = l 1 -l 5 + l 5 -l 9, ..., l n-4 -l n + l _n- 1
_{n + 4, ... = l 1} -l 9, ..., l n-4 -l n + 4, ... = l 1 -l 9, ..., l m -l m + 8, ... become. Here, n = m + 4.

Further, the difference between the adjacent difference data dD _m is calculated, and a new data sequence dDW (m) = dD _m + dD _{m + 1} is calculated. Using this, focus detection is performed in the sixth block BL6, and the focus is calculated. The state is detected and the defocus amount DF is calculated (step # 9
4) If focus detection is possible, the flag LCF2 is reset and the defocus amount DF6 of the sixth block BL6 is set as the defocus amount DFIS2 of the second island 22 in step # 43.
(Steps # 95 to # 97). Meanwhile, step #
If the focus cannot be detected at 95, the focus is detected at the seventh block BL7 to detect the focus state and the defocus amount DF.
Is calculated (step # 98), and if focus detection is possible (step # 99), the flow returns to step # 96 to set the flag LCF2
Is reset, and the defocus amount of the seventh block BL7 is set as the defocus amount DFIS2 of the second island 22, and the process proceeds to step # 43. On the other hand, if focus detection is not possible in step # 99, focus detection is performed in the eighth block BL8,
The focus state detection and the defocus amount DF are calculated (step # 100), and if focus detection is possible (step # 10)
1) Returning to step # 96, the flag LCF2 is reset, and the defocus amount of the eighth block BL8 is set as the defocus amount DFIS2 of the second island 22, and the process proceeds to step # 43. On the other hand, if the focus cannot be detected in step # 101, the process proceeds to step # 43 without performing the processes of steps # 96 and # 97.

FIG. 14 shows a subroutine for calculating the defocus amount of the third island 23 (step # 43). First, a predetermined value “−K” is stored in variables DF9 and DF10 for storing the respective defocus amounts of the ninth block BL9 and the tenth block BL10 (steps # 111 and # 112). Next, a flag LCF3 indicating the state of the low contrast in the third island 23 is set (step # 113). Then, the focus state detection of the ninth block BL9 and the defocus amount DF are calculated (step # 11).
4) If the focus can be detected, the flag LCF3 is reset and the obtained defocus amount DF is stored in the defocus amount DF9 of the ninth block BL9 (steps # 115 to # 11).
7). On the other hand, if it is determined in step # 115 that the focus detection cannot be performed, the process proceeds to step # 118 without performing the processes of steps # 116 and # 117.

Next, the detection of the focus state of the tenth block BL10 and the defocus amount DF are calculated (step # 118). If the focus detection is not possible from this calculation result, it is determined whether or not the flag LCF3 is set. Then, when the flag LCF3 is reset, that is, when the focus detection is enabled in the determination of step # 115, the process proceeds to step # 122. When the flag LCF3 is set, the process proceeds to the exposure calculation subroutine of step # 29 shown in FIG. 10 (steps # 119, # 120). on the other hand,
If focus detection is determined to be possible in step # 119, the flag LCF3 is reset without performing the processing in step # 120, and D
The defocus amount obtained in step # 118 is stored in F10 (# 129, # 130), and the process proceeds to step # 122.

In step # 122, an average processing width ΔDF of an averaging processing routine described later is determined. Next, the magnitude of the defocus amount DF9 and the defocus amount FG10 is determined, and the one with the larger defocus amount is defocus amount DFIS3 of the third island 23.
(Steps # 123 to # 127).

If the difference between the defocus amounts DF9 and DF10 is smaller than the average processing width ΔDF in steps # 124 and # 125, the defocus amounts DF9 and DF10 are averaged. Is the defocus amount DFIS3 of the third island 23 (step # 128). Then, when the processes of steps # 126, # 127 and # 128 are completed, the process proceeds to step # 29 in FIG.

Next, the exposure calculation subroutine shown in step # 29 of FIG. 10 will be described with reference to FIG.

First, an open luminance value (apex value) BV0 corresponding to the brightness of the subject 8 is input from the luminance detection circuit 29 to the microcomputer 26 (step # 131). Subsequently, a film sensitivity value (apex value) SV corresponding to the film sensitivity is input from the film sensitivity reading circuit 30 to the microcomputer 26 (step # 132). Next, these open brightness value BV0, film sensitivity value SV, and microcomputer 26 are determined in advance in step # 27 of FIG.
The exposure value E is calculated from the sum of the open aperture value AV0 input to
V (EV = BV0 + SV + AV0) is calculated (step # 13)
3).

Next, the control aperture value AV and the shutter speed TV are determined based on the exposure value EV (step # 134), and thereafter, the process proceeds to step # 30 shown in FIG.

Next, the process of calculating the drive amount of the photographing lens 2 shown in step # 32 of FIG. 10 will be described. In the process of step # 32, how the object is distributed is divided into patterns based on the defocus amounts of the islands 21, 22, and 23, and an optimal defocus amount calculation procedure is selected for each pattern. Thus, an optimal driving amount of the photographing lens 2 is obtained. Here, the means for selecting the calculation procedure will be briefly described.

First, it is determined whether the mode is the FA mode or the AF mode. In the case of the FA mode, the distance measurement of the second island 22 is prioritized. If the distance measurement of the second island 22 is possible, the driving amount of the photographing lens 2 is determined based on the defocus amount DFIS2 of the second island 22. If it is determined that the distance cannot be measured for the second island 22, the driving amount of the photographing lens 2 is determined based on the defocus amount of the nearest island. In other words, in the FA mode, shooting is often performed with a still subject in the center, and if the defocus amount is obtained based on ranging over a wide range, it is not clear which island is selected and displayed. To measure the distance in the center of the shooting screen,
The defocus amount DFIS2 of the island 22 is given priority.

On the other hand, in the case of the AF mode, when any one of the subjects 8 among the subjects on each of the islands 21, 22, and 23 can detect the focus, the defocus of the closest island, that is, the island having the largest defocus amount is performed. The photographing magnification is calculated based on the amount, the focal length data of the photographing lens 2 and the distance to the subject, and the calculation procedure of the defocus amount is changed according to the calculation result. In other words, basically, if the shooting magnification is large, the main subject always exists at the center of the shooting screen, and the defocus amount DFIS2 of the second island 22 is prioritized. Also, if the shooting magnification is small, the shooting will include the background, and the distribution of the distance to the subject will be large, and the main subject is often located close to the camera. I do. FIG. 16 shows an example of a value used as a guide for determining the photographing magnification and a procedure for selecting a calculation means based on the value.

That is, for example, in the case of the AF mode, the calculation means is selected when the focal length f of the photographing lens 2 is 35 mm. If the focal length f is equal to or greater than 35 mm and the imaging magnification βdf of the nearest island is smaller than a predetermined magnification βH (for example, 1/25), the photographic lens 2 is adjusted based on the defocus amount of the nearest island. Determine the drive amount. On the other hand, when the photographing magnification βdf exceeds the predetermined magnification βH, the second island 2
The second defocus amount DFIS2 is prioritized, and if the distance measurement of the second island 22 is impossible, the driving amount of the taking lens 2 is determined based on the defocus amount of the closest island.

Further, if the focal length f is less than 35 mm, the driving amount of the photographing lens 2 is determined based on the defocus amount of the nearest island. This is because, as the focal length f becomes shorter, the depth of field becomes deeper. Therefore, even if the object on the closest island is focused on in the distance distribution, the object detected on the other island is within a considerable range within the depth of field. Because it can be included in

In the case of the FA mode, regardless of the focal length f,
The defocus amount DFIS2 of the second island 22 is prioritized. If the second island 22 cannot be measured, the driving amount of the photographing lens 2 is determined based on the defocus amount of the closest island.

Here, the calculation means of the above-described photographing magnification βdf will be described.

The photographing magnification βdf is given by the following equation based on the focal length f and the subject distance x from the camera.

βdf = f / x Since the data of the focal length f is input from the photographing lens 2, when the subject distance x is obtained, the photographing magnification βdf is calculated. Further, the subject distance x is obtained by using the defocus amount DF _x from the infinity position of the taking lens 2 to the subject position as in the following formula.

x = f ² / DF _x However, since the taking lens 2 is not a single thin ideal lens, the principal point is at the front and back, and the principal point differs depending on the change in the focal length, so the subject distance x is approximated from the above formula. Calculated as a value.

On the other hand, the defocus amount DF _o from infinity position of the taking lens 2 to the current position is determined from the rotation amount (number) number of pulses corresponding to the N of the lens driving motor 33 stored in the counter of the microcomputer 26 , The relationship is as shown below.

N = k · DF _o DF _o = N / K However, the value of the coefficient k is determined based on the number of pulses and the like.

Then, the defocus amount DF _x is DF _x = DF _o + DF from the defocus amount DF _o and the defocus amount DF from the current position of the taking lens 2 to the in-focus position of the subject. From these expressions, the subject distance x is x = f ² / DF _x = f ² / (N / k + DF) Therefore, the imaging magnification βdf is βdf = f / x = (N / k + DF) / f.

Further, separately from the above equation, the photographing magnification βdf is the drive amount ΔN from the current position of the photographing lens 2 to the subject position (ΔN = DF ·
Using k), βdf = (N + ΔN) / f · k.

Next, the subroutine for performing the averaging process will be described with reference to the flowchart of the second island 22 (steps # 88 and # 9 in FIG. 13).
This will be described using 2) as an example.

The average processing width ΔDF of step # 88 in FIG. 13 is obtained according to a subroutine shown in FIG. First, it is determined whether or not the previous averaging process has been performed by using the flag WZF2 (step # 141). That is, when the previous averaging process is performed, the flag WZF2 is set. If the previous averaging process has been performed, the coefficient k ₁ is set to "1.5", and if the previous averaging process has not been performed, the coefficient k ₁ is set to "1" (steps # 142, # 143). . The reason for setting the coefficient k ₁ is
In order to prevent the averaging process from being performed or not being performed due to the variation in the distance measurement values, the averaging process may be expanded once and the averaging process may be performed again after the averaging process once ( This is to increase the probability).

Next, the shooting magnification βdf detected at the previous distance measurement is determined, and when the shooting magnification βdf is “1/20” or more, the coefficient k ₂
Is set to “1”, and the shooting magnification βdf is from “1/20” to “1/50”, the coefficient k ₂ is set to “0.5”, and the shooting magnification βdf is set to “1”.
/ 50 "coefficient k ₂ in the following cases" 0 "(step # 145 to # 149), the reason for setting to. The coefficient k ₂ proceeds to step # 150, the high imaging magnification βdf This is because there is a high possibility that the same subject will be occupied by a plurality of blocks in the same island.

Then, in step # 150, the reference average processing width ΔDF1 is set based on the aperture value FNo. That is, the depth of focus δ obtained by the product of the aperture value FNo. At the time of shooting and the permissible circle of confusion diameter ε is set as the reference average processing width ΔDF1 for storing the resolution of the shot image.

Then, the average processing width ΔDF1 is determined by multiplying the reference average processing width ΔDF1 by the coefficients k ₁ and k ₂ , and the flag WZF2 is reset (step #
151). Thereafter, the flow shifts to step # 89 shown in FIG.

The flag for determining whether or not the previous averaging process has been performed is determined by the first island 21 and the third island 21.
23 has the same flag WZF2. When the previous averaging process is performed, the flag WZF2 is set. When the previous averaging process is not performed, the flag WZF2 is reset.

Here, regarding the calculation of the defocus amount of each block and the focus detection impossibility judgment according to the present invention, the third block BL
This will be described with reference to FIG. 18 (a), taking 3 as an example (steps # 75 to # 78 in FIG. 13).

First, in the defocus range from the position where the front pin side shift pitch is four (shift pitch “−4”) to the position where the rear pin side shift pitch is 24 (shift pitch “24”), the reference unit 142
While shifting (shifting) the data of b, the data of the third block BL3 at each shift position and the reference unit 142
Obtain the correlation function H3 (I) with the data of b (step # 16
1).

 This correlation function H3 (I) is expressed by the following equation.

Here, K2 indicates a difference data string of the reference unit 142a, and S2 indicates a difference data string of the reference unit 142b.

Then, the correlation function H3 (I) at each shift position described above
Among them, the shift position IM having the best correlation, that is, the value of the correlation function H3 (I) is small, is extracted (step # 16).
2). Next, the interpolating operation between the pitches is performed using the value of each point of the correlation function H3 (I) (step # 163).

The interpolation pitch XM obtained by this interpolation calculation is as shown in the following equation (see Japanese Patent Application Laid-Open No. 60-247211).

XM = IM + {H3 (IM-3) -H3 (IM + 1)} / [MIN {H3 (IM-
1), H3 (IM + 1)}-H3 (IM)] / 2 The reason for performing this interpolation calculation is that the displacement of one pitch is converted to the defocus amount, which is as large as about 1000 μm. This is for accurate distance measurement.

The interpolation pitch XM obtained in this way is multiplied by a coefficient α determined by the pitch length of the optical system and the photoelectric conversion element array, and converted into a defocus amount DF (DF = XM · α) (step # 164). Next, the contrast (brightness / darkness) value C (3) of the data of the third block BL3 is obtained as the sum of the differences between the adjacent data of the reference unit 142a (step # 165).

 This contrast value C (3) is expressed by the following equation.

Then, the minimum interpolation value YM of the correlation function H3 (I) is obtained, and the ratio value R (3) between the interpolation value YM and the contrast value C (3) is obtained (step # 167).

The interpolated value YM and the ratio value R (3) are given by the following equation.

YM = H3 (IM)-| H3 (IM-1) -H3 (IM + 1) | / 2 R (3) = C (3) / YM The contrast value C (3) thus obtained and the ratio value R Focus detection is possible only when both (3) satisfy the set value, and the contrast value C (3) or the ratio value R
If either one of (3) is not satisfied, focus detection is disabled (low contrast determination). That is, first, the third
A determination is made as to whether the block BL3 has been able to perform focus detection in the previous routine processing, using the detection flag NLB3 (step # 168). This is to prevent the instability of the determination as to whether or not the focus can be detected for the subject at the end of the focus detection determination level, by summing the determination levels when the focus detection was possible last time. That is, with respect to the subject at the edge of the focus detection determination level, predetermined values C1 and R1 are set to the determination levels Cth and Rth, respectively (step # 169).

Then, in step # 168, if the focus cannot be detected last time or the first routine processing, that is, if the detection flag NLB3 has been reset, the photographing magnification βLS when photographing at the current position of the photographing lens 2 is performed. An operation is performed (step # 170).

The photographing magnification βLS is obtained by calculating the above-described photographing magnification βdf, that is, by substituting “0” into the defocus amount DF of βdf = (N / k + DF) / f.

Then, a product of the photographing magnification βLS and the focal length f is obtained, and when the product is a predetermined value f · βth (for example, βth is 1/15 or more in the case of a 300 mm lens), the determination level Ct
Predetermined values C4 and R4 greater than the prescribed values C1 and R1 for h and Rth, respectively
Is set (steps # 171, # 172). That is, the determination levels Cth and Rth are more severe than in step # 169.

When it is less than the predetermined value f · βth in step # 171,
The maximum local contrast value CLth of the third block BL3 is calculated (step # 173). Then, if the local contrast value CLth is larger than a preset predetermined value, that is, if a flag CL described later is set, the determination level Cth is set to a predetermined value C2, and if it is smaller than the predetermined value,
That is, if the flag CL has been reset, the determination level Cth is set to a predetermined value C3 (steps # 174 to # 174).
However, the relationship between the predetermined values C2 and C3 is set in advance so that C1 <C2 <C3 <C4.

Next, it is determined whether or not the flag AFSF is set. If it is set, the determination level Rth is set to a predetermined value R2, and if it is reset, the determination level Rth is set.
Is set to a predetermined value R3 (steps # 177 to # 179). However, the relationship between the predetermined values R2 and R3 is set in advance so that R1 <R2 <R3 <R4.

Then, after a predetermined value is set for each of the determination levels Cth, Rth (steps # 169, # 172, # 178, # 17).
9) If the contrast value C (3) is greater than the determination level Cth and the ratio value R (3) is greater than the determination level Rth, the detection flag NLB3 is set to set the contrast value C (3) or the ratio When either one of the values R (3) of the above is smaller than the determination level Cth or the determination level Rth, the detection flag NLB3 is reset (step #
180 to # 182). After that, processing such as calculation of the defocus amount of the fourth block BL4 and determination of the inability to detect the focus, which are shown in step # 79 and thereafter in FIG. 13, are performed.

As the detection flag NLB3, the detection flag NLB may be set for each island, and the determination level may be relaxed based on the detection flag NLB. In other words, if the focus is detected once and a large defocus amount is detected,
By driving the lens, the image interval approaches the basic image interval.
Therefore, the image itself of the reference portion is also, for example, the fourth block,
The image in the fifth block may be shifted to the third block and the fourth block by expanding the image interval by driving the lens. In such a case, focus detection becomes impossible in the third block. Therefore, a flag indicating low contrast for each island
The LCF may be stored in another flag LCFN, and this flag LCFN may be determined at the time of distance measurement of each island instead of the detection flag NLB.

Here, the reason why the determination levels Cth and Rth are tightened when the value is equal to or larger than the predetermined value f · βth in step # 172 will be described.

That is, in the case of a long focal length lens having an extremely large lens driving range (lens extension amount of 40 mm or more), a high magnification is set so that the lens position satisfies the above condition. If you try to detect the focus for the subject,
The spatial frequency component of the subject has a considerable high frequency component,
The image interval in the reference portion 142a or the reference portion 142b is extremely small, and a completely different image is projected between the reference portion 142a and the reference portion 142b. Further, since the magnification is low, a completely different subject is projected.
In such a blurred state, the optical system passes only an extremely low frequency component, so that focus detection cannot be normally performed.
However, as shown in FIG. 18 (b), when the persons A and B are arranged side by side, the data (high-frequency component) for distinguishing the persons A and B does not pass through the optical system.
Each of the images of the persons A and B projected on the reference portion 142b and the reference section 142b has substantially the same waveform, and the image of the person A and the image of the person b may be erroneously determined to determine that the focus can be detected.

On the other hand, if the lens position is at infinity and the magnification is low, the focus is to be detected on a near-view, high-magnification subject.
Even in such a case, the optical system passes only very low frequency components, so that detailed data does not pass. However,
Since the magnification is high, the image interval in the reference portion 142a or the reference portion 142b is extremely large, and different portions of the same subject are projected on the reference portion 142a and the reference portion 142b. Therefore, since the standard portion 142a and the reference portion 142b do not project different subjects, the determination levels Cth and Rth are set to be strict.

Also, in step # 161 and step # 165,
The correlation function H3 (I) and the contrast value C (3) of the third block BL3 were calculated, but as shown in FIG. 19, the other blocks BL1, BL2, BL4, BL5, BL9, BL10 are similarly obtained. be able to.

Next, the local contrast value CLt in the step # 173
The operation of h will be described. A noise amount Cnoise generated by the photoelectric conversion element array or the like is superimposed on the contrast level determination level Cth in step # 173. The magnitude of the noise itself of the noise amount Cnoise does not change, but the waveform of the noise varies. Therefore, when the minimum true subject contrast value required for focus detection is CMIN, the determination level Cth is expressed by the following equation.

Cth = CMIN + MAX {Cnoise} However, CMIN << MAX {Cnoise} Therefore, the noise amount Cnoise is small, and the contrast value is the judgment level C although the focus detection is possible originally.
In some cases, it is determined that the focus cannot be detected because it does not exceed th. Therefore, when the maximum noise amount Cnoise occurs,
Since it is predicted that noise exists in each part of the third block BL3, the contrast limit is determined for each difference data of the third block BL3 according to the following equation.

However, the value P is a predetermined value set in advance, and is selected from values equal to or less than “20”.

In the determination of the contrast limit, when the local contrast value satisfies a predetermined value, the determination level of the contrast value can be determined in the same manner as when the noise is small. That is, as shown in FIGS. 20 to 22, there are a plurality of determination routines for determining the contrast limit. That is, in the determination routine of FIG. 20, for example, a data block subdivided for every seven difference data from the third block BL3 is extracted, and if even one of the data blocks is equal to or more than the predetermined value CLth1, In addition, the judgment level of the total contrast value is relaxed.

That is, this determination routine first resets the flag CL and also resets the variable q (step #
191). Next, the data of the third block BL3 is replaced with the data K2
Data block CLOC consisting of difference data (7) from (1) to K2 (8) Is extracted, and this data block CLOC is compared with a predetermined value CLth1 (steps # 192, # 193). If the data block CLOC is greater than or equal to the predetermined value CLth1, the flag CL is set (step # 194), and the determination level of the contrast value is relaxed.

On the other hand, each difference data of the data block CLOC is a predetermined value C
If it is less than Lth1, the variable q is incremented to extract the data block CLOC consisting of the difference data from the data K2 (2) to K2 (9) of the third block BL3, and this data block CLOC is compared with the predetermined value CLth1. Do (step #
195, # 193). Similarly, data K2 (13) to K2 (2
This is continued until the data block CLOC consisting of the difference data up to 0) is extracted (steps # 192 to # 196). If even one piece of data in the data block CLOC is greater than or equal to the predetermined value CLth1, the flag CL is set, and if all data blocks CLOC are less than the predetermined value CLth1, the flag CL remains reset and the 18th. The process moves to the flowchart of FIG.

In FIG. 21, the maximum value and the minimum value of the data of the block are extracted, and the presence or absence of local contrast is determined from the difference between the maximum value and the minimum value.

That is, this determination routine extracts the maximum value [MAX {K2 (j)}] and the minimum value [MIN {K2 (j)}] from the data K2 (1 to 20) of the third block BL3, and determines these maximum values. Difference between [MAX {K2 (j)}] and minimum value [MIN {K2 (j)}] CLOC (CLOC = MAX {K2 (j)}-MIN {K2
(J)｝) is obtained (steps # 201 to # 203). Next, the flag CL is reset (step # 204), and the difference CLOC
Is compared with a predetermined value CLth2 (step # 205). And
If the value is equal to or larger than the predetermined value CLth2, the flag CL is set (step # 206), and the judgment level of the contrast value is relaxed. On the other hand, if the difference CLOC is less than the predetermined value CLth2, the flag C
While L remains reset, the process moves to the flowchart of FIG. 18 (a).

Further, in FIG. 22, the presence or absence of local contrast is determined by determining whether or not there is a predetermined number or more of data greater than a predetermined level among the data of the block.

That is, in this determination routine, first, the variable j is set to "1", the variables CL1 and CL2 are reset, and the flag CL
Is reset (step # 211). Next, the data K2 (1) of the third block BL3 is compared with the predetermined value CLth3, and when the data K2 (1) is equal to or more than the predetermined value CLth3, "1" is added to the variable CL1. In the case of, "1" is set to the variable CL2
Is added (steps # 212 to # 215). And the variable
The product of CL1 and the variable CL2 is calculated, and the products CL1 and CL2 are set to the predetermined value CL.
If th4 or more, the flag CL is set (step # 21)
7) Relax the contrast level judgment level. on the other hand,
If the product CL1 · CL2 is less than the predetermined value CLth4, the variable j is incremented and all the data K2 (2
With respect to (0), the processes from steps # 212 to # 215 are performed (steps # 212 to # 219). As a result, if the products CL1 and CL2 are less than the predetermined value CLth4 even after all the data have been processed, the flow shifts to the flowchart of FIG. 18A with the flag CL kept reset.

Next, the averaging process routine in step # 92 of FIG. 13 will be described. There are several processing procedures in this averaging processing routine, and the respective processing means are shown in FIGS.
This will be described with reference to FIG.

In the averaging process routine of FIG. 23, the contrast value C (3) obtained in steps # 165 and # 167 of FIG.
And a weighting function W (I) based on the ratio value R (3), and a weighted average of the defocus amount of each block is calculated using the weighting function W (I). In addition,
The weight function W (I) is, for example, a contrast value C
(3) itself or the ratio value R (3) itself may be used.

That is, first, the variable I is set to "3" (step # 217). Then, the largest defocus amount M among the respective defocus amounts DF3, DF4, DF5 of the blocks BL3, BL4, BL5.
The difference between AXDF and the defocus amount DF3 is obtained, and the difference is compared with the average processing width ΔDF. When the difference exceeds the average processing width ΔDF, the weight function W (3) is set to “0”. And
The variable I is incremented, the difference between the defocus amount MAXDF, the defocus amount DF4 and the defocus amount DF5 is determined, and this difference is compared with the average processing width ΔDF. When the difference exceeds the average processing width ΔDF, , The weight function W (4) and the weight function W (5) are set to “0” (steps # 222 to # 22).
Five). That is, the averaging process is performed using the defocus amount within the average processing width ΔDF among the defocus amounts DF3, DF4, and DF5 of the blocks BL3, BL4, and BL5. Next, weighted averaging is performed using the weight function W (I) processed as described above, and this average value is set as the defocus amount DFIS2 of the second island 22 (step # 226).

 That is, the defocus amount DFIS2 is given by the following equation.

DFIS2 = {DF3 ・ W (3) ＋ DF4 ・ W (4) ＋ DF5 ・ W (5)} /
{W (3) + W (4) + W (5)} Then, the wide zone flag WZF2 is set (step # 227). After that, the flow shifts to the defocus amount calculation subroutine of step # 43 in FIG.

In the averaging processing routine shown in FIG. 24, among the correlation functions, a plurality of blocks are combined at the maximum correlation position IM and the correlation positions IM + 1 and IM-1 before and after the maximum correlation position IM, and the correlation function is recalculated, and the correlation function is calculated again. It is designed to interpolate.
That is, if the blocks to be averaged (hereinafter referred to as target blocks) obtained in step # 90 in FIG. 13 are all blocks BL3, BL4, BL5, the data K2 (1-40) of the second island 22. And the correlation functions H3,4,5 (IM), H3,4,5 (IM
-1), H3,4,5 (IM + 1) are recalculated, and re-interpolation calculation is performed to set this value as the defocus amount DFIS2 of the second island 22 (steps # 231, # 232). If the third block BL3 and the fourth block BL4 are target blocks,
Correlation functions H3,4 (IM), H3,4 (IM
-1), H3,4 (IM + 1) are recalculated, re-interpolation is performed, and this value is set as the defocus amount DFIS2 (steps # 233, # 2).
34). Furthermore, the fourth block BL4 and the fifth block BL5
If is the target block, the correlation functions H4,5 (IM), H4,5 (IM-1), H4,5 (IM + 1) are recalculated with the data K2 (11 to 40) and reinterpolation calculation is performed. The lever value is set to the defocus amount DFIS2 (steps # 235, # 236). Also, step # 235
Then, if the third block BL3 and the fifth block BL5 are target blocks, the process proceeds to step # 232, and the wide zone flag WZF2 is set (step # 237), and then step # 43 of FIG. Then, the process proceeds to the defocus amount calculation subroutine.

Here, the calculation formulas of the correlation functions H3,4,5 (IM), H3,4 (IM), H4,5 (IM) in the above steps # 232, # 234, # 236 will be shown.

In addition, in the averaging processing routine of FIG. 25, the correlation functions H (IM), H (IM-1), H of each block to be subjected to the averaging processing are
Each of (IM + 1) is added, and the processing after the interpolation calculation is performed using the addition result. That is,
If all blocks BL3, BL4, BL5 are target blocks,
Correlation functions H3,4,5 (IM), H3,4,5 (IM-1), H3,4,5 (IM
+1) is calculated by the following formula (steps # 241, # 24)
2).

H3,4,5 (IM-1) = H3 (IM-1) + H4 (IM-1) + H5
(IM-1) H3,4,5 (IM) = H3 (IM) + H4 (IM) + H5 (IM) H3,4,5 (IM + 1) = H3 (IM + 1) + H4 (IM + 1) + H5
(IM + 1) If the third block BL3 and the fourth block BL4 are target blocks, the correlation functions H3,4 (IM), H3,4 (IM-
1), H3,4 (IM + 1) are calculated by the following equation (steps # 243, # 244).

H3,4 (IM-1) = H3 (IM-1) + H4 (IM-1) H3,4 (IM) = H3 (IM) + H4 (IM) H3,4 (IM + 1) = H3 (IM + 1) ) + H4 (IM + 1) Further, if the fourth block BL4 and the fifth block BL5 are target blocks, the correlation functions H4,5 (IM), H4,5 (IM
-1), H4,5 (IM + 1) are calculated by the following equation (steps # 245, # 246).

H4,5 (IM-1) = H4 (IM-1) + H5 (IM-1) H4,5 (IM) = H4 (IM) + H5 (IM) H4,5 (IM + 1) = H4 (IM + 1) ) + H5 (IM + 1) In step # 245, the third block BL3 and the fifth block
If the block BL5 is the target block, the correlation function H3,5
(IM), H3,5 (IM-1), H3,5 (IM + 1) are calculated by the following equation (step # 247).

H3,5 (IM-1) = H3 (IM-1) + H5 (IM-1) H3,5 (IM) = H3 (IM) + H5 (IM) H3,5 (IM + 1) = H3 (IM + 1) ) + H5 (IM + 1) Then, when the calculation of the correlation function in steps # 242, # 244, # 246, # 247 is completed, re-interpolation calculation is performed and this value is set to the defocus amount DFIS2 (step # 248). ). That is, the defocus amount DFIS2 is given by the following equation.

DFIS2 = α · IM + α · 1/2 · {H (IM-1) −H (IM + 1)} / [MIN {H (IM-1), H (IM + 1)} − H (IM)] and , The wide zone flag WZF2 is set (step # 249), and then the defocus amount calculation subroutine of step # 43 in FIG. 11 is entered.

Next, the sixth block BL of steps # 93 to # 95 in FIG.
The defocus amount calculation subroutine 6 will be described with reference to FIG.

First, the difference data string K2 used in the blocks BL3, BL4, BL5
Differences are taken at every seventh from (j) and S2 (j), and the sum data strings KW (j) and SW (j) between adjacent differences are obtained (step # 251). The sum data strings KW (j) and SW (j) are expressed by the following equation using the difference data string input in step # 26 of FIG.

KW (m) = K2 (m) + K2 (m + 1) + K2 (m + 4) + K2
(M + 5) SW (l) = S2 (l) + S2 (l + 1) + S2 (l + 4) + S2
(L + 5) Next, the correlation function H6 (I) is obtained, and the shift amount IM (IM = MIN {H6 (I)}) that minimizes the value of the correlation function H6 (I) is extracted (steps # 252, #). 253). This correlation function H6 (I) is expressed by the following equation.

Here, the correlation function H6 (I) is obtained by removing the data at both ends from the data string of the reference unit 142a as in the above equation, in order to obtain the correlation function H6 (I
Instead of using M-1), H6 (IM), H6 (IM + 1), correlation functions H6 (IM-2), H6 (IM-1), H6 (IM + 1), H6 (IM
+2) to use the correlation functions H6 (IM-2), H6 (IM + 2)
This is to correct a large shift in the position corresponding to the data of the reference portion 142b in (1) (see Japanese Patent Laid-Open No. 60-247211).

In step # 254, these correlation functions H6 (IM-2), H6
(IM-1), H6 (IM + 1), H6 (IM + 2) are obtained. That is, these equations are: Becomes

These correlation functions H6 (IM-2), H6 (IM-1), H6 (IM +
1) and H6 (IM + 2) are used to perform the interpolation calculation as shown in the following equation (step # 255).

XM = IM + {H6 (IM-2)-(H6 (IM + 2)} / [MIN {H6 (IM-2), H6 (IM + 2)}-MAX {H6 (IM-1), H6 ( IM + 1)}] / 2 The defocus amount D is obtained by multiplying the interpolation pitch XM by a coefficient α.
Convert to F (DF = XM · α) (step # 256). Next, the contrast value C of the data of the sixth block BL6
(6) is calculated by the following equation (step # 257).

Then, the minimum interpolated value YM1 of the correlation function H6 (IM) is obtained, and the value R (6) of the ratio is obtained by the following equation using the interpolated value YM1 and the contrast value C (6) (step # twenty five
8).

YM1 = MAX {H6 (IM-1), H6 (IM + 1)} / {| H6 (IM-2) -H6 (IM + 2) |} / 2 R (6) = C (6) / {YM1 -OF} Next, predetermined values are set to the determination levels Cth6 and Rth6, respectively, and focus detection is enabled only when both the contrast value C (6) and the ratio value R (6) satisfy the predetermined values. If either 6) or the ratio value R (6) is not satisfied, focus detection is disabled (steps # 259, # 260).

After that, the flow shifts to the flowchart of FIG. 13 to perform the process of step # 96 or step # 98.

Next, regarding the means of the interpolation calculation in step # 256,
This will be described with reference to the example of FIG.

First, the data string of the reference unit 142a is set to “3, 2, 1, −1, −2, −3”.
And the data string of the reference unit 142b is “4, 3, 2, 1, −1, −2, −3,
-4 ", the correlation function is H (0) = 0, H (-1) = 6, H
(1) = 6. Here, noise is added to the data sequence of the reference unit 142b, and the data sequence of the reference unit 142b is “4, 3, 2, 0, −1,
-1, -3, -4 ", the correlation function becomes H (0) = 2, H
(-1) = 5,5, H (1) = 6, and the influence of noise on the focal point, that is, the correlation function H (IM) is the largest, and the correlation functions H (IM-1), H (IM + The effect on 1) is relatively small. Therefore, unless the correlation function H (IM) is a periodic function, the correlation functions H (IM-2) and H (IM-) before and after the correlation function H (IM) are used without using the correlation function H (IM).
1), H (IM + 1), H (IM + 2) are used to reduce the influence of noise.

Further, in step # 258, when the normal interpolation calculation as shown in FIG. 18 (a) is performed, an offset amount is generated, and therefore the interpolation value YM1 to MAX {H6 (IM-1), H6 (IM + 1)} Difference between the correlation function H6 (IM) and OF1 or MAX {H6 (IM-2), H
Using the difference OF2 between 6 (IM + 2)} and MAX (H6 (IM-1), H6 (IM + 1)}, or the value OF set based on the value obtained by subtracting the contrast value C (6) The ratio value R (6) is calculated.

(Effect of the Invention) According to the present invention, the determination level is set for each of a plurality of areas into which the focus detection area is divided, so that the determination level can be set differently for each area. In addition, since the determination level is changed according to the previous determination result, it is determined that the focus detection is possible even if the subject detection condition slightly changes due to lens drive or camera shake after the focus is detected in the previous determination. Therefore, stable focus detection can be performed.

That is, when a moving subject is photographed by a camera equipped with the present focus detection device, once the focus of the subject is detected, the focus detection of the subject can be continuously performed thereafter. Therefore, the subject can be continuously shot.

Further, since the determination level is changed according to the contrast value obtained by summing the differences between the data of the adjacent photoelectric conversion elements, it is possible to prevent the noise generated in the photoelectric conversion element array or the like from being erroneously determined as the subject, A reliable focus detection can be performed.

Further, the determination level is changed according to the ratio of the contrast value obtained by summing the differences between the data of the adjacent photoelectric conversion elements and the best correlation value of the correlation values obtained by the mutual comparison of the pair of optical images, so that the subject is Even if there are multiple,
Accurate focus detection can be performed without erroneously selecting the subject that is the focus detection target.

Further, since the first determination level of the focus detection operation is relaxed compared to the case where the previous focus detection determination of the second and subsequent focus detection operations was made possible last time, the target object of focus detection may be mistaken. Can be prevented.

[Brief description of the drawings]

FIG. 1 is a schematic configuration diagram of a camera provided with an automatic focus detection device according to the present invention, FIG. 2 is a configuration diagram of a focus detection unit, FIG. 3 is a diagram showing a photoelectric conversion element array, and FIG. FIG. 5 shows a focus detection area of a photographing screen in FIG.
FIGS. 6 and 7 are diagrams illustrating the defocus range of each block, and FIG. 8 is a diagram illustrating an automatic focus detection device according to the present invention using a microcomputer. Circuit block diagram of the camera configured
9 and 10 are flowcharts of the interrupt operation of the microcomputer, FIGS. 11 to 14 are flowcharts showing a subroutine of the defocus amount calculation of the first to third islands, and FIG. 15 is a subroutine of the exposure calculation. FIG. 16 is a diagram for explaining a procedure for selecting a calculating means of the defocus amount, FIG. 17 is a flowchart showing a subroutine for obtaining an average processing width, and FIG. 18 (a) is a defocus amount of the third block. FIG. 18 (b) is a diagram for explaining the calculation and the focus detection impossible determination, FIG. 18 (b) is a diagram for explaining erroneous determination when a person is lined up, and FIG. FIG.
20 to 22 are flowcharts showing a subroutine for determining the contrast limit, FIGS. 23 to 25 are flowcharts showing an averaging process routine, FIG. 26 is a flowchart showing a defocus amount calculation subroutine, and FIG. 27 is an interpolation FIG. 4 is a diagram for explaining means of calculation. DESCRIPTION OF SYMBOLS 1 ... Automatic focus detection device, 2 ... Photographing lens, 3 ... Focus detection means, 4 ... Defocus amount determination means, 5 ... Lens position detection means, 6 ... Lens drive means, 7 ... Control means , 8 subject, 21 ... first island, 22 ... second island, 23 ... third island, 26 ... microcomputer, 27 ... lens circuit, 28 ... focus detection data output circuit, 34 ... Lens control circuit, 35 ... Power battery, 36 ... Power supply circuit, 141a, 142a, 143a ... Reference part, 141b, 142b, 143b ...
Reference section, BL1 First block, BL2 Second block, BL3 Third block, BL4 Fourth block, BL5
... 5th block, BL9 ... 9th block, BL10 ... 1st
0 block, SW ₁ ... Power switch, SW ₂ ... Shooting preparation switch, SW ₄ ... Selection switch.

Front Page Continuation (72) Kenji Ishibashi, Inventor Kenji Ishibashi 2-3-13 Azuchi-cho, Chuo-ku, Osaka Prefecture Osaka International Building Minolta Camera Co., Ltd. (72) Inventor, Hisao Morita 2 Azuchi-cho, Chuo-ku, Osaka-shi, Osaka Chome 3-13 Osaka International Building Minolta Camera Co., Ltd. (72) Inventor Hiroshi Ueda 2-33 Azuchicho, Chuo-ku, Osaka-shi, Osaka Osaka International Building Minolta Camera Co., Ltd. (72) Inventor Noriyuki Okisu 2-13-3 Azuchicho, Chuo-ku, Osaka City, Osaka Prefecture Osaka International Building Minolta Camera Co., Ltd. (56) References JP 63-17415 (JP, A) JP 60-247210 (JP, A) Special Kai 62-150310 (JP, A)

Claims (4)

(57) [Claims] 1. A focus detection device for detecting a focus state based on a light image of a subject incident on a focus detection region, wherein a determination level is set for each of a plurality of regions into which the focus detection region is divided. Level setting means, a judgment means for judging whether or not focus detection is possible by comparing the judgment level set for each area and a calculation result based on the data of each area, and the judgment result A focus detection device comprising: a storage unit that stores the change and a change unit that changes the determination level according to the stored previous determination result. 2. The focus detection device includes a light receiving element array including a plurality of photoelectric conversion elements, and the determination level is changed according to a contrast value obtained by summing up differences between data of adjacent photoelectric conversion elements. The focus detection device according to claim 1, wherein 3. The focus detection device includes a light receiving element array including a plurality of photoelectric conversion elements that receive a pair of optical images obtained from an optical image of an incident object, and the determination levels are adjacent photoelectric conversion elements. The focus detection device according to claim 1, wherein the focus detection device is changed in accordance with a ratio between a contrast value obtained by summing up the differences of the data of (1) and a best correlation value of correlation values obtained by mutual comparison of the pair of optical images. . 4. The determination level at the first focus detection operation is set to a lower level than when the focus detection operation is determined to be impossible last time after the second focus detection operation. Focus detection device.