JPH1010416A - Focusing device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a focusing device flexibly adaptable to the change of various kinds of object conditions and capable of adequately and accurately making the suppression of hunting and the performance of focusing compatible in the focusing device performing the focusing of a photographing optical system. SOLUTION: This device is provided with a focus detecting means 2 for detecting the focusing information of the photographing optical system 1, a focusing controlling means 3 for driving the system 1 to a focusing condition, based on the focusing information detected by the means 2, and a statistical arithmetic calculation means 4 for performing statistical arithmetic calculation of the focusing information at a plurality of points of time and calculating the variance of the focusing information. The means 3 resumes the driving of the system 1 when the focusing information exceeds a threshold predetermined corresponding to the variance during the drawing of the system 1 is stopped or during the updating of the target driving position for the means 3 is stopped.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a focus adjusting device for adjusting the focus of a photographing optical system, and more particularly, to a threshold value for starting re-driving of the photographing optical system in accordance with fluctuations in focus information of the photographing optical system. Focus adjustment device.

[0002]

2. Description of the Related Art Conventionally, optical devices such as cameras are equipped with a focus adjusting device for automatically adjusting the focus of a photographing optical system. In these focus adjustment apparatuses, focus information such as a defocus amount and a distance measurement value is sequentially detected using a known focus detection method (a phase difference detection method, an external light passive method, or the like).

The focus adjusting device sequentially updates the target driving position of the photographing optical system based on the defocus amount and the like, and controls the driving of the photographing optical system so as to reach the target driving position. Ideally, when the photographing optical system reaches the target driving position, the defocus amount becomes zero, and the photographing optical system stops at the target driving position.

However, in practice, the focus detection CC
Due to noise or the like occurring in D, even if the photographing optical system reaches the target drive position, the detected value of the defocus amount does not become zero and slight fluctuation occurs. Therefore, the target drive position is updated in small increments, and small fluctuations (hunting) occur in the photographing optical system. Such a hunting phenomenon is recognized by the photographer as "strange or instability of the focus adjustment operation", and therefore suppression of the hunting phenomenon is strongly desired.

In this type of focus adjusting device, FIG.
A focus width as shown in FIG. 2A is set in advance. In this setting, when the magnitude of the defocus amount is less than 100 μm, it is uniformly determined as “focused state”. The focus adjustment device stops updating the target drive position in accordance with the determination of the in-focus state.

Therefore, even if the defocus amount fluctuates within the range of the focusing width, the target drive position is not updated at all, and the photographing optical system stops at the point when the target drive position is reached. As described above, when the fluctuation of the defocus amount is limited within the in-focus range, no hunting occurs in the imaging optical system. On the other hand, if the defocus amount changes beyond the in-focus range, the focus adjustment device uniformly determines that the state is "out of focus". The focus adjustment device starts updating the target drive position again in response to the determination of the out-of-focus state, and drives the imaging optical system toward the new target drive position.

Further, as shown in FIG. 12B, there is also known an apparatus in which a focus width is provided with hysteresis. In such a setting, once the defocus amount falls within the range of the focus width, the focus width is widened. Therefore, the fluctuation of the defocus amount is less likely to be out of the range of the focusing width, and hunting of the photographing optical system can be strongly prevented.

[0008]

Normally, for a subject having low luminance or low contrast, the accuracy of focus detection is reduced, and fluctuations in focus information (a defocus amount and a distance measurement value) are largely generated. However, FIG. 12 (a) and FIG.
In the conventional example shown in FIG. 2B, the focus width or the hysteresis margin is set uniformly regardless of the type of the subject.

For this reason, when the focus is adjusted for an extremely low-contrast or low-brightness subject, the focus information fluctuates beyond the margin of the focus width, and hunting occurs. Was.

[0010] Such hunting is improved by extremely widening the focus width or hysteresis margin.
However, in this state, there is a problem that the re-start of the photographing optical system due to the change or movement of the subject is delayed by an amount corresponding to an increase in the focus width or the hysteresis margin.

As a result, there has been a problem that the responsiveness and the followability of the focus adjustment are remarkably impaired, and a precious shutter chance is missed at the moment of photographing. Furthermore, for a subject that is slightly out of focus, the focus information does not change beyond the in-focus width or the hysteresis margin, so that there is a problem that the photographing optical system is not driven again and the focus adjustment does not operate.

As a result, even if the photographer captures a subject whose focus position is slightly shifted, the focus adjusting device does not detect the subject, and there is a problem that the subject cannot be finely and finely adjusted. . As described above, it is very difficult to achieve both “suppression of hunting” and “performance of focus adjustment” according to various object situations.

Therefore, a method of sequentially changing the focus width or the hysteresis margin according to the luminance and the contrast amount of the subject can be considered. However, the fluctuation of the focus information changes unpredictably due to various image patterns of the subject image. For this reason, the above-described method has a problem that the fluctuation of the focus information cannot be accurately determined, and the focus width or the hysteresis allowance cannot be adequately and flexibly changed.

According to the first and second aspects of the present invention, in order to solve such a problem, it is possible to flexibly adapt to a variety of changes in the situation of a subject and to suppress "hunting" and "focus adjustment performance". It is an object of the present invention to provide a focus adjustment device capable of accurately achieving compatibility with the above.

According to the third aspect of the present invention, in addition to the object of the first aspect, there is provided a focus adjusting device capable of achieving both "suppression of hunting" and "performance of focus adjustment" more quickly and accurately. The purpose is to provide. Claims 4 and 5
It is another object of the present invention to provide a focus adjustment device capable of immediately changing the threshold value of the re-drive start in response to a sudden change in the subject state, in addition to the object of the first aspect.

[0016]

FIG. 1 is a block diagram for explaining the first to third aspects of the present invention. Hereinafter, the solution of the present invention will be described with reference to FIG. According to the first aspect of the present invention, a focus detecting unit for detecting focus information of the photographing optical system, and the photographing optical system is driven to a focused state based on the focus information detected by the focus detecting unit. A focus control unit 3 and a statistical calculation unit 4 that performs a statistical operation on the focus information at a plurality of time points and calculates a variance of the focus information are provided. Alternatively, when the focus information exceeds the threshold value corresponding to the variance during “the update of the target drive position is stopped by the focus control unit 3”, the driving of the imaging optical system 1 is restarted.

According to a second aspect of the present invention, in the focus adjustment apparatus according to the first aspect, the statistical operation means 4 calculates the variance of the focus information with an average value of the focus information as a center value. . According to a third aspect of the present invention, in the focus adjustment apparatus according to the first aspect, the statistical calculation means 4 calculates a variance of the focus information with a trend center of the focus information as a center value.

FIG. 2 is a block diagram for explaining the fourth and fifth aspects of the present invention. Hereinafter, the solution of the present invention will be described with reference to FIG. According to a fourth aspect of the present invention, there is provided the focus adjustment apparatus according to the first aspect, further comprising a reliability handling unit 5 for selecting or correcting the threshold value according to reliability information relating to reliability of focus information. It is characterized by the following.

According to a fifth aspect of the present invention, in the focus adjustment apparatus according to the fourth aspect, the reliability information includes brightness of the subject image, spatial frequency distribution of the subject image, or contrast of the subject image. Or a steepness of a correlation curve or a minimum correlation amount when a correlation operation is performed on a set of light images obtained by dividing and forming a subject light flux, or a moving speed of a subject.

(Function) In the focus adjusting device according to the first aspect, the defocus amount, the distance measurement value, and other focus information are detected via the focus detecting means 2. Focus control means 3
The photographing optical system 1 is driven to a focused state based on the focus information.

The statistical operation means 4 performs a statistical operation on the focus information at a plurality of points in time to calculate the variance of the focus information. The focus control unit 3 obtains a predetermined threshold value corresponding to the “calculated value of the variance of the focus information”. Generally, when it is determined that “the imaging optical system 1 has reached the focused state”,
The focus control unit 3 stops driving the photographing optical system 1 or
Alternatively, the update of the target drive position is stopped.

In such a stopped state, the focus control means 3 determines whether or not the focus information exceeds the above-mentioned threshold value, and restarts the driving of the photographing optical system 1 when the focus information is exceeded. Here, the variance calculated by the statistical calculation is a value indicating the degree of fluctuation of the focus information.

In particular, unlike the subject brightness and the contrast amount, which are only one of the causes of the fluctuation of the focus information, the variance is a value directly indicating the degree of the fluctuation of the focus information. The "occurrence margin" can be grasped more accurately. That is, since a probability density distribution (eg, a normal distribution) indicating the degree of fluctuation of the focus information is known in advance, the “range in which the fluctuation of the focus information falls” can be stochastically specified from the calculated variance. Correlation with such a range, by determining the threshold of the re-drive start, "whether the change of the focus information is due to fluctuations, or a significant change" to stochastically accurately determine,
It is possible to appropriately determine whether or not to re-drive.

For example, when the probability density distribution of the focus information is a normal distribution, the probability that the fluctuation of the focus information is located outside the following threshold with respect to the variance σ ² is as follows. Threshold fluctuation probability outside the threshold ± σ 31.74% ± 2σ 4.55% ± 3σ 0.27% ± 4σ 0.006% Therefore, the focus information outside the threshold is detected at a frequency exceeding the fluctuation probability described above, If focus information outside the threshold is detected despite a sufficiently low fluctuation probability,
It can be accurately determined that "the change of the focus information is not due to the fluctuation but a significant change". Re-driving can be started accurately based on such an accurate determination.

Therefore, unlike the conventional example, it is not necessary to increase the threshold value in anticipation of various changes in the object situation, and it is possible to minimize the deterioration in the performance of the focus adjustment.

As described above, by determining the threshold of the re-drive start based on the variance, the "margin of hunting" in the current object situation is accurately grasped based on the stochastic support, and the "hunting margin" is obtained. The balance between the "suppression effect" and the "performance of focus adjustment" can be properly achieved. In the focus adjustment device according to the second aspect, the statistical calculation means 4 calculates the variance of the focus information with the average value of the focus information as a center value.

Normally, immediately before the focus state is determined, the moving speed of the photographing optical system 1 is sufficiently reduced because the photographing optical system 1 is approaching the target drive position. In such a state, the “change in focus information” caused by the movement of the imaging optical system 1 is small, and the average value of the focus information can be regarded as the fluctuation center. Therefore, when the sample section is limited immediately before the focus state is determined, the variance is calculated using the average value of the focus information as a center value, thereby easily and accurately detecting the degree of fluctuation of the focus information. can do.

After the in-focus state is determined, the moving speed is further reduced because the driving of the photographing optical system 1 is stopped or the updating of the target driving position is stopped. Therefore, even when a sample section is set immediately before and after the determination of the in-focus state, the variance is calculated with the average value of the focus information as the center value, so that the degree of fluctuation of the focus information can be easily and accurately determined. Can be detected.

In general, the arithmetic processing for obtaining the average value requires a much smaller amount of arithmetic processing than the arithmetic processing for calculating the trend center (moving average, regression analysis, etc.), and enables high-speed processing. Note that if the sample section of the focus information is set to be long, the number of samples increases, and the variance accuracy increases. On the other hand,
As the sample section becomes longer, the movement of the center of fluctuation becomes conspicuous, and the variance value is calculated to be larger than the value indicating “the degree of fluctuation of the focus information”.

With respect to such a point, the calculated value of the variance may be determined to be lower in anticipation of “movement of the fluctuation center”. In addition, when setting the sample section in advance, the optimum sample section can be set by weighting and evaluating two points of “accuracy of dispersion” and “movement of fluctuation center”. In the focus adjusting apparatus according to the third aspect, the statistical operation means 4 obtains the trend center of the focus information using, for example, a statistical operation (moving average, regression analysis, or the like).

The statistical calculation means 4 calculates the variance of the focus information using the center of the trend as a central value. In this way, by setting the trend center as the center value, the influence of “movement of the fluctuation center” included in the calculated variance value is reliably eliminated, and the variance value is accurately calculated. Therefore, even if the sample section of the focus information is set to be long, the correlation between the “variance value” and the “fluctuation degree of the focus information” is kept good, so that the “fluctuation degree of the focus information” can be determined more accurately. can do.

Since the variance can be calculated in advance retroactively before the determination of the in-focus state, it is possible to immediately start the "re-start threshold determination" immediately after the determination of the in-focus state. Becomes In the focus adjustment device according to the fourth aspect, the reliability handling means 5 selects or corrects the threshold value of the re-drive start according to the reliability information related to the reliability of the focus information.

Therefore, the threshold value for the start of re-driving is set based on a plurality of values such as "dispersion of focus information" and "reliability information of focus information". "Dispersion of focus information"
As described above, this is a value that directly indicates the degree of fluctuation of the focus information, and the threshold value for starting re-drive can be set more accurately. However, a change in the variance requires a change of the focus information at several points in time. Therefore, in a case where the subject situation changes suddenly, the change of the threshold value is slightly delayed.

On the other hand, the brightness of the subject, the amount of contrast, and other reliability information are only one of the causes of the fluctuation of the focus information, so that the threshold value cannot be set accurately. However, these pieces of reliability information are values that change in response to a sudden change in the subject situation. Therefore, by selecting or correcting the threshold for starting re-driving according to such reliability information, it becomes possible to immediately change the threshold in response to a sudden change in the subject situation.

As described above, since the threshold value change based on the reliability information is a quick change, the change in the threshold value may return to the original value as time passes.
With such a configuration, it is possible to smoothly shift to “more accurate threshold value setting” based on the variance of the focus information over time. If a large deviation occurs between the variance expected from the reliability information and the actual variance, it is expected that the fluctuation of the focus information is not on the measurement system but on the subject side. You. For example, a subject competing for perspective and near (a subject in which a near and far shooting object coexists in a close range) or a periodic subject (since the image pattern has periodicity, a comb-shaped characteristic is generated in the correlation curve of the phase difference detection, and false data is generated. (A subject in which the focus amount is erroneously detected) is expected.

Therefore, by performing the correlation analysis between the reliability information and the variance, it is possible to detect that the subject is a subject such as a subject competing for distance and near when the correlation between the two is extremely reduced. The shooting mode of the camera, parameters for focus adjustment, and the like can be appropriately changed in accordance with the detection of the type of the subject. In the focus adjusting device according to the fifth aspect, at least one of the following values is adopted as the reliability information relating to the reliability of the focus information.

(1) Brightness of the Subject Image When the brightness of the subject image is low, the S / N of the focus detecting light receiving element becomes low, and the fluctuation of the focus information becomes large. For example, such brightness can be detected based on the photometric value of the peripheral light amount, the photometric value of the subject, the accumulation time of the focus detection light receiving element, and the like.

(2) Spatial frequency distribution of the subject image If the spatial frequency distribution is distributed in the center of the low frequency band, for example, in the correlation calculation of the phase difference detection method, the residual of the image pattern does not change so much. , The detection accuracy of the phase difference becomes low. As a result, the fluctuation of the focus information increases. For example, such a spatial frequency distribution can be detected from the output level of a transversal filter or another spatial filter.

(3) Contrast of Subject Image When the contrast of the subject image is low, the image pattern does not include a high-frequency and large-amplitude frequency component, so that the residual of the image pattern does not change significantly. as a result,
The detection accuracy of the spatial phase difference is reduced, and the fluctuation of the focus information is increased. (4) Moving speed of the subject When the moving speed of the subject image is high, the exposure flow of the subject image becomes large, and the high-frequency spatial frequency component included in the image pattern is lost. Therefore, the residual of the image pattern does not change significantly, and the detection accuracy of the spatial phase difference is reduced. As a result, the fluctuation of the focus information increases.

(5) Steepness of Correlation Curve When the steepness of the correlation curve (shown in FIG. 3) is low, the detection accuracy of the spatial phase difference decreases, and the fluctuation of the focus information increases. (6) Minimum Correlation Amount of Correlation Curve When the minimum correlation amount (shown in FIG. 3) is large, it is a case where the image pattern has low correlation with respect to a set of divided and formed optical images. In such a case, the detection accuracy of the spatial phase difference is reduced, and the fluctuation of the focus information is increased.

By examining any one of the reliability information (1) to (6) as described above, it is possible to easily estimate "the degree of fluctuation of the focus information" accompanying a sudden change in the situation of the subject. Therefore, it is possible to immediately change the threshold value of the redrive start without delaying a sudden change in the subject situation.

[0042]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 4 is a diagram for explaining the configuration of the first embodiment (corresponding to claims 1 and 2). In FIG. 4, a photographing optical system 12 is mounted on the camera 11, and a quick return mirror 13 and a sub-mirror 14 are sequentially arranged along the optical axis of the photographing optical system 12.

A focus detecting section 15 is arranged in the direction of reflection of the sub-mirror 14, and an output of the focus detecting section 15 is connected to a microprocessor 16. Microprocessor 16
The lens information storage unit 17 is connected to the second input, and the encoder 18 arranged on the drive shaft of the motor 20 is connected to the third input. On the other hand, the microprocessor 16
Is connected to a motor 20 via a drive circuit 19, and the power of the motor 20 is transmitted to a lens drive mechanism 21 in the photographing optical system 12 via a camera mount.

In FIG. 4, the microprocessor 16
Is shown as a block diagram. That is, the output of the focus detection unit 15 described above is provided to the focus calculation unit 23, and the calculation result of the focus calculation unit 23 is taken into the target position update unit 24 and the memory 25. This memory 2
The output of 5 is given to the statistical operation unit 26.

The calculation result of the statistical calculation unit 26 is given to the target position update unit 24, and the update result of the target position update unit 24 is given to the motor operation amount calculation unit 28. The PWM output (pulse width modulation output) of the motor operation amount calculation unit 28 is provided to the motor 20 via the drive circuit 19 described above. The lens information in the lens information storage unit 17 is provided to the target position update unit 24, and the pulse output of the encoder 18 is output from the target position update unit 24 and the motor operation amount calculation unit 28.
Given to.

Here, the invention according to claims 1 and 2 and the first
The focus detection means 2 corresponds to the focus detection unit 15 and the focus calculation unit 23, and the focus control means 3 corresponds to the drive circuit 19, the target position update unit 24, and the motor operation amount calculation unit 28. Correspondingly, the statistical calculation means 4 corresponds to the statistical calculation unit 26. FIG. 5 is a flowchart illustrating the operation of the first embodiment.

The operation of the first embodiment will be described below with reference to these figures. First, when the release (not shown) of the camera 11 is half-pressed (S1 in FIG. 5), the focus detection unit 15
Starts accumulation of photocharges (S2 in FIG. 5). The focus calculation unit 23 captures the photoelectric output of the focus detection unit 15,
The defocus amount is calculated by a known phase difference detection calculation (S3 in FIG. 5).

The target position updating unit 24 calculates a target drive position to be reached by the photographing optical system 12 based on the lens position and the defocus amount obtained from the encoder 18 (S4 in FIG. 5). Here, when the target drive position was updated last time (S5 in FIG. 5), it is determined whether the magnitude of the defocus amount is equal to or less than a predetermined focusing width (S6 in FIG. 5) and whether the lens speed is equal to or less than a predetermined value. (S7 in FIG. 5).

If the result of this determination is true, the target position updating section 24 determines that the in-focus state has been reached, and stops updating the target drive position (S8 in FIG. 5). In such a case, the motor operation amount calculation unit 28 drives the motor 20 based on the positional deviation between the target driving position where updating is stopped and the current lens position, and drives the imaging optical system 12 to the target driving position. And stop. (FIG. 5S9).

If the result of the determination is false, it is determined that the camera is out of focus, and the target drive position is updated to the latest value (S12 in FIG. 5). In such a case, the motor operation amount calculation unit 28 drives the motor 20 based on the position deviation between the latest target drive position and the current lens position, and directs the imaging optical system 12 to the latest target drive position. Drive gradually (Fig. 5
S9).

On the other hand, if the target drive position has not been updated last time (S5 in FIG. 5), it is determined according to the following procedure whether the latest change in the defocus amount is due to fluctuation or significant change. First, the statistical calculation unit 26 fetches the defocus amount bf at the past N times from the memory 25,

(Equation 1) Is used to calculate an average value bfave for N times.

The difference from each defocus amount bf is determined with the average value bfave as the center value, and the mean square is determined for the difference.

(Equation 2) The variance σ ^{2 of the} defocus amount is calculated as in the above equation (S10 in FIG. 5). Based on such a calculation result, the target position updating unit 24 determines the threshold value of the redrive start to be 3σ (three times the standard deviation σ).

Next, the target position updating section 24 determines whether or not the latest defocus amount is within 3σ (S11 in FIG. 5). When the fluctuation of the defocus amount has a normal distribution, the probability of the fluctuation of the defocus amount exceeding 3σ is “about 0.27%”. Therefore, in spite of such a low probability, when the defocus amount changes beyond 3σ, it is not a change due to fluctuation, but a significant change to the subject (for example, movement of the subject).
Can be determined to have occurred.

Based on the result of this determination, the target position updating section 24 restarts updating the target drive position (S1 in FIG. 5).
2). As a result, the driving of the photographing optical system 12 is started again, and the driving is started toward a new focal point. When the latest magnitude of the defocus amount is within 3σ, the target position update unit 24 determines that the change is caused by the fluctuation of the defocus amount, and maintains the previous target drive position as it is (S8 in FIG. 5). . As a result, the driving of the imaging optical system 12 is not restarted, and hunting due to fluctuation of the defocus amount is prevented.

These operations are repeated by returning to step S1. By the operation as described above, in the first embodiment, the threshold value of the re-drive start is appropriately set according to the dispersion of the defocus amount. That is, as shown in FIG. 6A, when the fluctuation of the defocus amount is small, the variance σ ² becomes small, and the threshold value 3σ of the re-drive start becomes small. Therefore, a significant change in the defocus amount (point P in the drawing) is sensitively detected, and updating of the target drive position (re-driving of the imaging optical system 12) can be promptly restarted.

On the other hand, as shown in FIG. 6B, when the fluctuation of the defocus amount is large, the variance σ ² becomes large, and the threshold value 3σ for starting the re-drive becomes large. Therefore, the fluctuation of the defocus amount rarely exceeds the threshold value 3σ, and hunting can be strongly prevented. As described above, the threshold value of the redrive start is appropriately set based on the dispersion of the defocus amount, so that the occurrence of hunting is necessary and sufficiently and surely prevented, and the deterioration of the responsiveness and the followability of the focus adjustment is reduced. Can be minimized.

Further, with the average value of the focus information as the center value,
Since the variance of the defocus amount is calculated, the amount of arithmetic processing is much smaller than the arithmetic processing (moving average, regression analysis, etc.) for determining the trend of the defocus amount. Therefore, it is possible to easily and quickly calculate the variance σ ^{2 of the} defocus amount. Hereinafter, another embodiment will be described.

FIG. 7 is a diagram showing the configuration of the second embodiment (corresponding to claims 1 and 3). The feature of the configuration in the second embodiment is that a statistical operation unit 32 is provided as an internal function of the microprocessor 16. In addition,
The same components as those in the first embodiment are denoted by the same reference numerals and are shown in FIG. 7, and description thereof will be omitted.

Here, the invention according to claims 1 and 3 and the second invention
The focus detection means 2 corresponds to the focus detection unit 15 and the focus calculation unit 23, and the focus control means 3 corresponds to the drive circuit 19, the target position update unit 24, and the motor operation amount calculation unit 28. Correspondingly, the statistical calculation means 4 corresponds to the statistical calculation unit 32. FIG. 8 is a flowchart illustrating the operation of the second embodiment.

The operation of the second embodiment will be described below with reference to these figures. First, when the release (not shown) of the camera 11 is half-pressed (S1 in FIG. 8), the focus detection unit 1
5, the accumulation of the photocharge is started (S2 in FIG. 8). The focus calculation unit 23 captures the photoelectric output of the focus detection unit 15 and calculates the defocus amount bf by a known phase difference detection calculation (S3 in FIG. 8). This defocus amount bf is
The information is sequentially stored in the memory 25 together with the “lens position Y obtained from the encoder 18” and the “detection time t”.

The target position updating unit 24 calculates the defocus amount b
Based on f and the lens position Y, a target drive position to be reached by the photographing optical system 12 is calculated (S4 in FIG. 8). here,
The statistical operation unit 26 obtains the trend center of the defocus amount bf according to the following procedure. First, the statistical calculation unit 26 stores the defocus amount bf, the lens position Y (the same unit as the defocus amount bf) and the detection time t at the past N times in the memory 2.
Take in sequentially from 5,

(Equation 3)

(Equation 4) Is used to calculate the average image plane position Zave and the average detection time tave.

A product-sum operation is performed using these average values,

(Equation 5)

(Equation 6) The product sums Szt and Stt are calculated. Using the values of these sum of products Szt and Stt,

(Equation 7)

(Equation 8) Is calculated, and the slope a and the intercept b of the linear regression equation are obtained.

By the above statistical operation (regression analysis), the trend center X (t) of the defocus amount bf becomes X (t) = at · b + Y (9) (S5 in FIG. 8). By calculating a difference from each defocus amount bf with the trend center X (t) as a center value, and calculating a mean square of the difference,

(Equation 9) The variance σ ^{2 of the} defocus amount is calculated as in the above equation (S6 in FIG. 8).

If the target drive position has been updated last time (S7 in FIG. 8), the target position update unit 24 determines that the magnitude of the defocus amount is equal to or smaller than the predetermined focus width (S8 in FIG. 8).
It is determined whether the lens speed is equal to or less than a predetermined value (FIG. 8S
9). If the result of this determination is true, the target position update unit 24 determines that the in-focus state has been reached, and stops updating the target drive position (S10 in FIG. 8).

In such a case, the motor operation amount calculation unit 28
Drives the motor 20 based on the positional deviation between the target driving position for which updating has been stopped and the current lens position, drives the imaging optical system 12 to the target driving position, and stops. (FIG. 8S
11). If the above determination result is false, it is determined that the camera is not in focus, and the target drive position is updated to the latest value (S13 in FIG. 8).

In such a case, the motor operation amount calculation unit 28
Drives the motor 20 on the basis of the position deviation between the latest target drive position and the current lens position, etc.
Toward the latest target drive position (S1 in FIG. 8).
1). On the other hand, when the target drive position has not been updated last time (S7 in FIG. 8), it is determined by the following procedure whether the change in the latest value of the defocus amount is due to fluctuation or significant change.

First, the target position updating section 24 sets the threshold value of the redrive start to 3σ based on the variance σ ^{2 of the} defocus amount.
(3 times the standard deviation σ). Next, the target position updating unit 24 determines whether or not the latest defocus amount is within 3σ (S12 in FIG. 8). When the fluctuation of the defocus amount has a normal distribution, the probability of the fluctuation of the defocus amount exceeding 3σ is “about 0.27%”.

Therefore, despite the low probability, when the defocus amount exceeds 3σ, a significant change (for example, movement of the subject) is not caused by the fluctuation but the subject. It can be determined that it has occurred. Based on this determination result, the target position updating unit 24
The update of the target drive position is restarted (S13 in FIG. 8). As a result, re-driving of the photographing optical system 12 starts, and the photographing optical system 12 is driven toward a new focal point. When the latest magnitude of the defocus amount is within 3σ, the target position update unit 24 determines that the change is caused by the fluctuation of the defocus amount, and maintains the previous target drive position as it is (S10 in FIG. 8). . As a result, the driving of the imaging optical system 12 is not restarted, and hunting due to fluctuation of the defocus amount is prevented.

These operations are repeated by returning to step S1. By the operation as described above, in the second embodiment, the same effect as in the first embodiment can be obtained.

Further, as an effect peculiar to the second embodiment, the variance σ ^{2 of the} defocus amount is calculated with the center of the trend of the defocus amount as a center value, so that the influence of “movement of the fluctuation center” is completely eliminated. Can be eliminated. Further, since the variance σ ² is calculated with the trend center X (t) as the center value, even when the sample section of the defocus amount is set to be long, the variance σ ² and “the degree of fluctuation of the defocus amount” Is kept good, and "the degree of fluctuation of the defocus amount" can be determined more accurately. As a result, it is possible to more accurately set the threshold value 3σ for starting the re-drive.

That is, as shown in FIG. 9A, when the fluctuation of the defocus amount is small, the movement center X
Since the variance σ ² centered on (t) becomes small, the threshold value 3σ for starting the re-drive becomes small. Therefore, a significant change in the defocus amount (point R in the drawing) can be sensitively detected, and updating of the target driving position (re-driving of the imaging optical system 12) can be promptly restarted.

On the other hand, as shown in FIG. 9B, when the fluctuation of the defocus amount is large, the variance σ ² centered on the movement center X (t) becomes large, so 3σ increases. Therefore, the fluctuation of the defocus amount rarely exceeds the threshold value 3σ, and hunting can be strongly prevented. In addition, since the variance σ ² can be calculated in advance before entering the focusing width, the threshold value 3σ is determined immediately after entering the focusing width, and the “judgment of the threshold value for starting the re-drive” is promptly started. Can be.

Next, another embodiment will be described. FIG. 10 is a diagram for explaining the configuration of the third embodiment (corresponding to claims 1, 2, 4, and 5). Regarding the structural features in the third embodiment, the microprocessor 16
The internal function of (1) is that a reliability handling unit 42 that fetches reliability information (contrast amount) at the time of focus detection from the focus calculation unit 23 and corrects a threshold value for starting re-drive is provided.

The same components as those in the first embodiment are denoted by the same reference numerals and are shown in FIG. 10, and the description thereof will be omitted. Here, regarding the correspondence between the inventions according to the first and second aspects and the third embodiment, the focus detection unit 2 corresponds to the focus detection unit 15 and the focus calculation unit 23, and the focus control unit 3 corresponds to the drive circuit. 19. Target position update unit 2
4 and the motor operation amount calculation unit 28, and the statistical calculation unit 4 corresponds to the statistical calculation unit 26.

Regarding the correspondence between the fourth and fifth aspects of the present invention and the third embodiment, the reliability handling unit 5 corresponds to the reliability handling unit 42. FIG. 11 is a flowchart illustrating the operation of the third embodiment. Hereinafter, the operational features of the third embodiment will be described. First, when the release (not shown) of the camera 11 is half-pressed (S1 in FIG. 11),
The accumulation of photocharges is started in the focus detection unit 15 (S2 in FIG. 11).

The focus calculation section 23 takes in the photoelectric output of the focus detection section 15 and calculates the defocus amount by a known correlation calculation. On the other hand, the reliability handling unit 42 calculates the steepness of the correlation curve (FIG. 3) obtained by the correlation operation based on, for example,
The contrast amount of the subject image is calculated (S3 in FIG. 11).
The target position updating unit 24 is configured to control the photographing optical system 12 based on the lens position and the defocus amount obtained from the encoder 18.
Is calculated (S4 in FIG. 11).

If the target drive position has been updated last time (S5 in FIG. 11), the same operation as in the first embodiment is performed (S6 and S7 in FIG. 11). On the other hand, if the target drive position has not been updated last time (S5 in FIG. 11), it is determined whether or not a predetermined number of times have elapsed since the drive was stopped (S10 in FIG. 11).

If the predetermined number of times has not elapsed, the reliability handling section 42 obtains a threshold value corresponding to the contrast amount and outputs it to the target position updating section 24. On the other hand, when the predetermined number of times has elapsed, the statistical calculation unit 26 calculates the variance σ ² of the defocus amount using the average value of the defocus amount as a center value (S12 in FIG. 11). Based on the variance σ ² and the contrast amount, a threshold value is obtained by table processing (see FIG. 11).
S13).

The target position updating section 24 determines that the latest defocus amount exceeds the above-described threshold (S1 in FIG. 11).
4), it is determined that a significant change (for example, movement of the subject) has occurred in the subject, and the update of the target drive position is restarted (S15 in FIG. 11). As a result, the driving of the photographing optical system 12 is restarted, and the photographing optical system 12 is driven toward a new focal point. If the latest defocus amount is within the threshold value, the target position update unit 24 maintains the previous target drive position as it is to prevent hunting (FIG. 11S).
8).

These operations are repeated by returning to step S1. According to the operation described above, in the third embodiment, the same effect as in the first embodiment can be obtained.
Further, in the third embodiment, the threshold value for starting the re-drive is selected or corrected according to the contrast amount. Since this contrast amount changes immediately in response to the subject situation, the threshold value is immediately changed in response to the sudden change in the subject situation without waiting for a change in the variance value that gradually changes over several points of focus information. It is possible to do.

In the above-described embodiment, the threshold value for starting re-driving is set to 3σ (three times the standard deviation). However, the present invention is not limited to this. Should be set to. Further, in the above-described embodiment, the threshold value for the start of re-drive is always 3σ. However, when the variance σ ² is extremely small, 3σ takes an invalid value in view of the optical performance of the photographing optical system. There are cases. Therefore, a predetermined lower limit value fd
May be determined.

Further, in the above-described embodiment, the variance σ ^{2 of the} focus information is obtained. However, the present invention is not limited to this configuration, and a numerical value that can be converted into the variance of the focus information may be obtained. For example, the standard deviation σ, the variation coefficient, the frequency distribution, and the like may be obtained by using the focus information at several points as a sample. Also,
In the above-described embodiment, the defocus amount is used as the focus information. However, the present invention is not limited to this configuration. Any detected value or calculated value indicating the focus state can be used as the focus information. . For example, the distance measurement value to the subject, the position of the subject image plane, the focusing distance of the photographing optical system, the image interval of the external light passive type, the target driving position of the photographing optical system, the light receiving angle or light receiving position of the external light active type May be used as focus information.

In the above-described embodiment, the function X (t) using the time t as an input variable is adopted as the prediction function of the regression analysis, but the present invention is not limited to this. In general, it is possible to use a prediction function using a variable having a correlation with focus information as an input variable. Also, the input variables may be two or more types. In that case, a multiple regression analysis technique can be used.

Furthermore, in the above-described embodiment, a linear function is used as a prediction function for regression analysis, but the present invention is not limited to this. In general, it is sufficient to specify a parameter for obtaining the least square error by assuming a predetermined function, so that a quadratic function, an exponential function, and other various functions can be adopted as the prediction function. Further, in the above-described embodiment, the trend center of the focus information is obtained using the regression analysis, but the present invention is not limited to this. In general, the focus information at several points may be used as sample data to determine the trend of the focus information. For example, moving average (moving avera
ge) or auto-regressive.

[0085]

As described above, according to the first aspect of the present invention, the variance of the focus information is calculated by the statistical calculation, so that the "fluctuation degree of the focus information" in the current subject situation is stochastically and accurately. Can be detected. Therefore, it is possible to appropriately and accurately set the re-drive start threshold within a range in which the occurrence of the fluctuation of the focus information can be neglected stochastically in accordance with various subject situations.

As a result, it is possible to reliably prevent hunting from occurring in almost all of the various object situations. Further, since the probability that the fluctuation of the focus information occurs outside the threshold value can be accurately grasped, the hunting cycle can be precisely controlled regardless of various object situations. Therefore, the hunting cycle can be suppressed to such an extent that the photographer does not perceive hunting or does not feel uncomfortable.

Further, if focus information outside the threshold is detected at a frequency exceeding the fluctuation probability distribution, or if the focus information outside the threshold is detected despite the probability distribution being sufficiently low, the following message is displayed. The change is not due to fluctuations,
It is a significant change ". Therefore, it is possible to accurately and appropriately determine the start of re-driving in accordance with various object situations.

Further, the threshold value of the re-drive start can be accurately set in accordance with various situations of the subject, as far as the fluctuation of the focus information does not exceed the probability. Accordingly, it is possible to suppress the hunting to a necessary and sufficient level and to minimize the deterioration of the responsiveness and the followability of the focus adjustment. As a result, the delay in the operation of the focus adjustment and the like are eliminated, and the photographer can accurately capture a valuable photo opportunity.

In particular, for a subject that is slightly out of focus, the focus adjustment operation is started sensitively, so that fine and fine focus adjustment can be performed. According to the second aspect of the present invention, the variance of the focus information is calculated using the average value of the focus information as a center value. In general, the calculation processing for obtaining the average value is much smaller than the calculation processing for calculating the trend center or the like (moving average, regression analysis, etc.). Therefore, by using such an average value as the central value, it is possible to easily and quickly calculate the variance of the focus information. As a result, the configuration of claim 1 can be easily realized.

According to the third aspect of the invention, the variance of the focus information is calculated with the center of the movement as the center value, so that the influence of the "movement of the fluctuation center" is completely eliminated, and the value of the variance can be accurately obtained. . Therefore, even in a case where the sample section of the focus information is set to be long, the correlation between the “variance value” and the “fluctuation degree of the focus information” is favorably maintained, and the “fluctuation degree of the focus information” is more accurately determined. You can judge. As a result, it becomes possible to more accurately set the threshold value of the re-drive start.

Since the variance can be calculated in advance before the determination of the in-focus state, it is possible to immediately start the "re-start threshold determination" immediately after the determination of the in-focus state. . According to the fourth aspect of the present invention, the threshold value for starting the re-drive is selected or corrected according to the reliability information.
Since the reliability information is a value that changes in response to the subject situation, it is possible to immediately change the threshold value in response to a sudden change in the subject situation without waiting for a change in the variance that gradually changes over several points of the focus information. It becomes possible.

Further, by performing a correlation analysis between the reliability information and the variance, the variance expected from the reliability information is obtained.
If there is a large deviation from the actual dispersion,
The cause of the fluctuation of the focus information is not in the measurement system,
It can be determined that it is on the subject side. In other words, in such a case, the conflicting subject (the subject in which the photographic subject is located in the close range) or the periodic subject (since the image pattern has periodicity, the correlation curve at the time of detecting the phase difference has a comb tooth shape). (A subject that causes characteristics and erroneously detects a false defocus amount) is expected. Therefore, it is also possible to appropriately change the photographing mode of the camera or the parameters of the focus adjustment, etc., in accordance with the detection of the type of the subject.

According to the fifth aspect of the present invention, a value which has a strong correlation with the fluctuation of the focus information and changes in response to a change in the situation of the subject is selected and adopted as the reliability information. The degree of fluctuation can be easily and immediately corrected. Therefore, the threshold value can be easily and immediately changed according to a sudden change in the subject situation. As described above, in the focus adjustment device to which the present invention is applied, the “re-start threshold” can be set to a stochastically appropriate value in accordance with the degree of fluctuation of focus information flexibly.
It is possible to realize a focus adjustment device that appropriately balances "suppression of hunting" and "performance of focus adjustment".

In particular, at the time of low brightness or low contrast, the operation of the photographing optical system is remarkably stabilized, and a focus adjusting device which gives an extremely stable operation feeling can be realized.

[Brief description of the drawings]

FIG. 1 is a block diagram for explaining the invention according to claims 1 to 3;

FIG. 2 is a block diagram for explaining the invention according to claims 4 and 5;

FIG. 3 is a diagram illustrating a correlation curve.

FIG. 4 is a diagram illustrating a configuration of a first embodiment (corresponding to claims 1 and 2);

FIG. 5 is a flowchart illustrating the operation of the first embodiment.

FIG. 6 is an explanatory diagram illustrating a defocus amount according to the first embodiment.

FIG. 7 is a diagram showing a configuration of a second embodiment (corresponding to claims 1 and 3);

FIG. 8 is a flowchart illustrating the operation of the second embodiment.

FIG. 9 is an explanatory diagram illustrating a defocus amount according to the second embodiment.

FIG. 10 is a diagram illustrating a configuration of a third embodiment (corresponding to claims 1, 2, 4, and 5).

FIG. 11 is a flowchart illustrating the operation of the third embodiment.

FIG. 12 is a diagram showing a focus width of a conventional example.

[Explanation of symbols]

 REFERENCE SIGNS LIST 1 shooting optical system 2 focus detection means 3 focus control means 4 statistical calculation means 5 reliability handling means 11 camera 12 shooting optical system 13 quick return mirror 14 sub-mirror 15 focus detection unit 16 microprocessor 17 lens information storage unit 18 encoder 19 drive circuit Reference Signs List 20 motor 21 lens drive mechanism 23 focus calculation unit 24 target position update unit 25 memory 26 statistical calculation unit 28 motor operation amount calculation unit 32 statistical calculation unit 42 reliability handling unit

Claims (5)

[Claims] A focus detection unit configured to detect focus information of an imaging optical system; a focus control unit configured to drive the imaging optical system to a focused state based on the focus information detected by the focus detection unit; Statistical operation means for performing a statistical operation on the focus information at the time point and calculating a variance of the focus information, wherein the focus control means includes: "while the driving of the imaging optical system is stopped" or "in the focus control means, If the focus information exceeds a predetermined threshold value corresponding to the variance while the update of the target drive position is stopped, the drive of the imaging optical system is restarted. 2. The focus adjustment apparatus according to claim 1, wherein the statistical calculation means calculates a variance of the focus information using an average value of the focus information as a center value. 3. The focus adjustment apparatus according to claim 1, wherein the statistical calculation means calculates a variance of the focus information with a trend center of the focus information as a center value. 4. The focus adjustment device according to claim 1, wherein the focus information includes:
A focus adjusting device comprising a reliability handling means for selecting or correcting the threshold value. 5. The focus adjustment apparatus according to claim 4, wherein the reliability information includes brightness of a subject image, spatial frequency distribution of the subject image, contrast of the subject image, or moving speed of the subject, or A focus adjustment device characterized by a steepness of a correlation curve or a minimum correlation amount when a correlation operation is performed on a set of light images obtained by dividing and forming an object light beam.