JPH1164917A - Image blur correcting device and panning deciding device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To always obtain adequate image blur correcting effect by generating an image blur correcting signal based on the decision result of a panning deciding means and the information of a panning direction detected by a panning direction detecting means. SOLUTION: A yaw angular velocity sensor 12 detects the horizontal vibration of an image pickup device. A pitch angular velocity sensor 13 detects vertical vibration. A correcting optical system 14 corrects image blur by performing sliding or rotational displacement. A yaw correcting actuator 15 horizontally drives the system 14. A pitch correcting actuator 16 vertically drives the system 14. A horizontal position sensor 17 detects the horizontal position of the system 14. A vertical position sensor 18 detects the position of the system 14. A microcomputer 11 outputs the image blur correcting signal based on decision result by the panning deciding means and the information of the panning direction detected by the panning direction detecting means, and the system 14 performs the blur correction of the photographic image of the image pickup device.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to improvements in an image blur correction device and a panning determination device which are mounted on an imaging device such as a camera, a camcorder, and a digital still camera, and correct image blur due to camera shake or the like.

[0002]

2. Description of the Related Art At present, in an imaging device such as a camera or a camcorder, as a means for suppressing image shake caused by camera shake, a camera shake or a direct image shake is detected, and a photographing optical axis is set so as to cancel the shake. 2. Description of the Related Art An image blur correction system that drives a bending correction optical system or moves an imaging range has been put to practical use and has been adopted in some products.

FIG. 12 is a block diagram showing a schematic configuration of an example of the image blur correction system.

[0004] The main components are a microcomputer 1101 for performing signal processing, a yaw angular velocity sensor 1102 for detecting horizontal vibration of the image pickup apparatus, a pitch angular velocity sensor 1103 for detecting vertical vibration, and an image obtained by performing sliding or rotational displacement. A correction optical system 1104 for correcting shake, a yaw correction actuator 1105 for driving the correction optical system 1104 in the horizontal direction, a pitch correction actuator 1106 for driving the correction optical system 1104 in the vertical direction, and the correction optical system 11
The horizontal position sensor 1107 detects the horizontal position of the correction optical system 1104, and the vertical position sensor 1108 detects the horizontal position of the correction optical system 1104.

In FIG. 1, a yaw angular velocity sensor 1102
The signal detected by the high-pass filter 1109
, The high-frequency noise is removed by the low-pass filter 1110, the digital signal is digitized by the A / D converter 1111, and the microcomputer 110 is activated by the ON signal of the switch SW 1 as the yaw angular velocity signal Ω _Y.
It is taken into 1.

[0006] The signal detected by the pitch angular velocity sensor 1103 is subjected to the same processing, and is taken into the microcomputer 1101 as a pitch angular velocity signal Ω _P.

The detection signal of the horizontal position sensor 1107 is A /
It is digitized by the D converter 1117 and is taken into the microcomputer 1101 as a horizontal position signal Y _OUT . The detection signal of the vertical position sensor 1108 is also digitized in the same manner, and is taken into the microcomputer 1101 as a vertical position signal P _OUT .

The microcomputer 1101 generates angular velocity signals Ω _Y and Ω _P
, The horizontal position Y _IN and the vertical position P _IN to be displaced by the correction optical system 1104 are obtained, and the actual position signal Y
_OUT , _POUT and amplify it by taking the differential with the correction optical system 11
04 are output as drive signals D _Y and D _P. Alternatively, there is a method in which the drive signals D _Y and D _P are pulse width modulated and output as PWM signals.

The horizontal drive signal _DY is supplied to the D / A converter 11
The analog signal is converted into an analog signal by the controller 15 and input to the yaw correction actuator 1105. If the horizontal drive signal is a PWM signal, it is directly input to the yaw correction actuator 1105 without passing through the D / A converter 1115. Also vertical drive signals D _P, the pitch correction actuator 1 in the same manner
It is input to 106.

The angular velocity sensors 1102 and 1103
Detects the angular velocity caused by the panning of the image pickup device in addition to the angular velocity caused by the camera shake, so that the correction optical system 1104 can be used without changing the angular velocity signals Ω _Y and Ω _P.
Are converted into the drive signals D _Y and D _P of the correction optical system 110.
Although within the range of the stroke No. 4, the image shake due to the panning is also corrected, and the followability to the subject in the panning direction is reduced particularly after the start of the panning or at the turning point of the panning.

The main band of the angular velocity signal by panning is
Normally, since it is located at a position lower than the main band of the angular velocity signal due to camera shake, a method of extracting only the angular velocity signal due to camera shake by passing through a high-pass filter can be considered, but high frequency components are included in the angular velocity signal due to panning at the start and end of panning. Since it is included in a relatively large amount and is buried in the band of the angular velocity signal due to camera shake, the angular velocity signal due to panning and the angular velocity signal due to camera shake cannot be completely separated, and the effect of improving the followability to the subject cannot be expected.

In order to give priority to the ability to follow the subject, there is a method of providing some kind of panning determination means, and if it is determined that panning is being performed, there is a method of locking image blur correction. , The effect of the shake correction during panning cannot be expected.

Therefore, conventionally, panning judging means for each of the yaw and pitch directions of the image pickup apparatus (specifically, if the yaw and pitch direction shake signals are larger than predetermined values, panning is performed in that direction). Is determined, and if it is determined that panning is being performed, image shake correction in the determined direction is locked, and image shake correction in the direction not determined is continued. A method has been adopted in which image blur correction is performed as much as possible within a range that does not impair the followability of the image.

A sequence of image blur correction in which such a panning measure is taken will be described with reference to a flowchart shown in FIG. Each step number shown in the figure represents each processing unit.

First, the sequence is started by the input of the ON signal of the switch SW1, and in step 1201, the yaw swing angular displacement Ψ and the pitch swing angular displacement Θ are initialized. Next, in step 1202, the angular velocity signals Ω _Y , Ω _P ,
It is confirmed whether or not the position signals Y _OUT and P _{OUT of the} correction optical system 1104 have been updated. If the position signals have been updated, the process proceeds to step 1203 which is a panning determination routine.

In step 1203, first, a panning judgment in the yaw direction is performed. Here, when the absolute value of the yaw angular velocity Ω _Y exceeds a predetermined limit value Ω _LMT , it is determined that panning is being performed, and in step 1204, the yaw angular velocity Ω is determined.
Force _Y to zero. Next, in step 1205, panning judgment in the pitch direction is similarly performed, and if the absolute value of the pitch angular velocity Ω _P exceeds the limit value Ω _LMT , the pitch angular velocity Ω _P is fixed to zero in step 1206.

The angular velocity signals Ω _Y , Ω _P thus modified by the panning judgment are applied to a digital integrator in step 1207 to obtain deflection angular displacements Ψ, Θ. In this step, a and b are time constants determined by the integrator and constants determined by the sampling period.

The deflection angular displacements Ψ and Θ are calculated in step 12
08, the horizontal position Y _IN at which the correction optical system 1104 should be displaced,
The position signals are converted to the vertical position P _IN , respectively, and in step 1209, the difference between the actual position signal Y _OUT and the actual position signal P _OUT is obtained.
_Y, as D _P, the microcomputer 1101 in step 1210
Is output from the output terminal.

On the other hand, in step 1211, the switch SW
1 is OFF, and in step 1212,
A flag that is set when an unillustrated exposure sequence independent of the present sequence is activated is checked. If this flag is not set, the sequence ends. Otherwise, the process returns to step 1202 and returns to step 1202. Is repeatedly executed.

In the above sequence, in order to understand the behavior of the correction optical system 1104 during panning, consider, for example, the behavior of the shake angular displacements Ψ, 後 after the start of panning in the yaw direction.

Immediately after the start of the panning in the yaw direction, until the absolute value of the yaw angular velocity Ω _Y suddenly increases and exceeds the limit value Ω _LMT , the yaw angular velocity Ω _Y is determined as the vibration to be corrected and becomes the yaw vibration angular displacement Ψ in step 1207. It is added for each sampling with the coefficient a. During this time, the image blur due to the fluctuation of the yaw angle displacement Ψ is corrected by the correction optical system, so that the followability to the subject in the yaw direction is impaired, but a slight safety factor is expected in the maximum value of the imaginable camera shake angular velocity. Step 1 by setting the limit value as the limit value Ω _LMT
The panning determination of 203 can be performed as strictly as possible, and the loss of followability can be suppressed to a level that is hardly noticeable.

At the moment when the absolute value of the yaw angular velocity Ω _Y exceeds the limit value Ω _LMT , the yaw angular velocity Ω _Y is fixed to zero in step 1204. At that time, in step 1207, the calculation of the yaw shake angular displacement Ψ is processed as Ψ = bΨ. Since the constant b is usually set to 1 or less,
Thereafter, the yaw deflection angular displacement Ψ gradually approaches zero until the panning is completed and the absolute value of the yaw angular velocity Ω _Y falls below the limit value Ω _LMT .

[0023] That is, until below the absolute value of limit value Omega again limits since the moment of exceeding the _LMT Omega _LMT of the yaw angular velocity Omega _Y, the correction optical system 1104 stops the yaw direction of the image blur correction operation, horizontal A so-called centering operation of gently moving toward the center position is performed. During this time, the ability to follow the subject in the yaw direction is maintained. On the other hand, during this period, the image shake correction in the pitch direction is continuously performed without hindering the ability to follow the subject in the yaw direction, so that the maximum image shake correction effect is maintained while maintaining the ability to follow the subject in the yaw direction. Can be expected.

The same applies to panning in the pitch direction.

[0025]

However, in the above conventional example, since the panning judgment is performed only in the two directions of yaw and pitch, and the panning direction itself is not detected, there are the following problems. . (1) When performing panning in an oblique direction such that it is determined that panning is being performed in both the yaw and pitch directions,
The image shake correction in both the yaw and pitch directions is locked, and the image shake correction effect cannot be obtained. (2) If panning of the same size as panning determined to be panning in the yaw or pitch direction is performed in an oblique direction, it may not be determined that panning is being performed. Followability slightly decreases in the yaw or pitch direction.

(Purpose of the Invention) A first object of the present invention is to provide an image blur correction apparatus which can obtain an appropriate image blur correction effect even if the imaging apparatus is panned in any direction. It is.

A second object of the present invention is to provide an image shake correcting apparatus capable of detecting a panning direction without providing a dedicated shake detecting means for detecting a panning direction.

A third object of the present invention is to provide an image blur correction device capable of improving the reliability of a detected panning direction.

A fourth object of the present invention is to obtain the maximum image blur correction effect while keeping the followability to the object in the panning direction as much as possible, regardless of the direction in which the imaging apparatus is panned. It is intended to provide an image blur correction device.

A fifth object of the present invention is to provide an image blur correction effect in a direction orthogonal to the panning direction while maintaining the ability to follow the subject in the panning direction, regardless of the direction in which the imaging apparatus is panned. It is an object of the present invention to provide an image blur correction device that can obtain the following.

A sixth object of the present invention is to provide an image shake correcting apparatus which can eliminate the direction dependency of the panning judgment and can accurately perform the panning judgment in any direction.

A seventh object of the present invention is to provide an image shake correcting apparatus which can perform image shake correction only in a direction in which image shake correction is preferentially performed.

An eighth object of the present invention is to provide a panning judging device which can accurately judge panning in any direction.

[0034]

In order to achieve the first object, the present invention according to the first and second aspects of the present invention is directed to an image pickup apparatus which is arranged so that the image pick-up around two different axes orthogonal to the image pickup optical axis of the image pickup apparatus. First and second shake detecting means for detecting shake of the apparatus, panning determining means for determining whether or not the imaging apparatus is panning, and panning direction detecting means for detecting a panning direction of the imaging apparatus; The first and second shake signals detected by the first and second shake detection means, the determination result by the panning determination means, and
Based on information on the panning direction detected by the panning direction detection means, an image shake correction signal generation means for generating an image shake correction signal, and the image shake correction signal An image shake correction apparatus includes an image shake correction unit that performs shake correction.

In the above configuration, a panning direction detecting means for detecting a panning direction is provided, and image blur correction is performed by taking into account not only the first and second shake signals and the result of panning determination but also information on the detected panning direction. A signal is generated.

In order to achieve the second object,
According to a third aspect of the present invention, the first and second low-pass filters for removing high-frequency components from the first and second shake signals and outputting first and second low-pass shake signals, 1st, 2nd
And a determining means for determining a shake direction obtained from the low-frequency shake signal as the panning direction.

In the above configuration, the first and second shake detecting means are means for detecting a shake signal of the image pickup apparatus and also means for detecting a signal which is a base of the image pickup apparatus in the panning direction.

In order to achieve the third object,
According to a fourth aspect of the present invention, when the first and second low-pass filters determine that the panning is being performed by the panning determination means, the high-frequency component including the main band of the camera shake from the first and second shake signals. The time constant is determined so that is substantially removed, and if the panning determination unit does not determine that panning is being performed, the time constant is determined so that the cutoff frequency is higher than when the panning is determined to be performed. This is an image blur correction device having a low-pass filter of No. 2.

In the above configuration, focusing on the fact that the band of the shake signal component due to the panning is different between the start time of the panning and the subsequent panning, the panning determination means determines that the panning is not being performed and the panning is being performed. The time constants of the first and second low-pass filters in the panning direction detecting means are switched between when the determination is made.

In order to achieve the fourth object,
According to a fifth aspect of the present invention, the image shake correction signal generation means generates an image shake signal based on only the first and second shake signals unless the panning determination means determines that panning is being performed. If it is determined that panning is being performed by the means, a different process is performed on the shake signal component in the panning direction and the shake signal component in the direction orthogonal to the panning direction, and an image shake correction signal is generated based on the obtained signal. An image shake correction apparatus having an image shake correction signal generation unit to generate the image shake correction signal.

In the above configuration, if it is determined that the panning is being performed, different processes are performed on the shake signal component in the panning direction and the shake signal component in the direction orthogonal to the panning direction, specifically, the shake in the panning direction. For the signal component and the shake signal component in the direction orthogonal to the panning direction, processing is performed so that only the shake signal component due to camera shake or the like to be corrected is extracted as much as possible.

In order to achieve the fifth object,
According to a sixth aspect of the present invention, when the panning determination unit determines that panning is being performed, an image shake correction signal generation unit generates an image shake correction signal based only on a shake signal component in a direction orthogonal to the panning direction. This is an image blur correction device having

In the above configuration, since the shake signal component in the panning direction is mostly a shake signal component due to panning which is not to be corrected, the shake signal component due to panning is not used when generating the image shake correction signal. The image shake correction signal is generated based only on the shake signal component in the direction orthogonal to the panning direction.

In order to achieve the sixth object,
According to a seventh aspect of the present invention, there is provided a shake speed amount calculating means for obtaining a magnitude of a shake speed vector of the imaging apparatus from the first and second shake signals detected by the first and second shake detecting means, An image shake correction device comprising: a panning determination unit configured to determine that the imaging apparatus is performing panning when the magnitude of the shake speed vector obtained by the shake speed amount calculation unit exceeds a preset limit value. It is assumed that.

In the above arrangement, the magnitude of the shake velocity vector of the image pickup device is obtained from the first and second shake signals, and the magnitude of the shake velocity exceeds a preset limit value, Even if the panning is performed in the direction, the panning can be determined under the same condition (the preset limit value).

In order to achieve the seventh object,
The present invention according to claim 8, wherein the first and second shake detecting means for detecting shake of the imaging device around two different axes orthogonal to the photographing optical axis of the imaging device, Signal combining means for forming a signal corresponding to a shake in a direction different from the first and second directions by combining the first and second shake signals detected by the second shake detection means; Image shake correction signal generation means for generating an image shake correction signal based on the shake signal component in the predetermined direction from the synthesizing means, and shake correction of a photographed image of the imaging device by inputting the image shake correction signal However, the present invention is directed to an image blur correction apparatus including an image blur correction unit and a variable unit that changes the predetermined direction.

In the above arrangement, a shake signal component in a predetermined direction, that is, a direction in which image shake correction is to be preferentially performed, is calculated from the first and second shake signals, and image shake correction is performed using the shake signal. Like that. Note that the predetermined one direction may be arbitrarily designated from the outside.

In order to achieve the eighth object,
According to a ninth aspect of the present invention, there are provided first and second shake detecting means for detecting shake of the image pickup device around two different axes orthogonal to a photographing optical axis of the image pickup device. The first and second shake signals detected by the second shake detection means are combined to obtain a shake speed amount calculation means for obtaining the magnitude of the shake speed vector of the image pickup apparatus, and a shake speed amount calculation means for obtaining the magnitude of the shake speed vector. When the magnitude of the shake velocity vector exceeds a preset limit value, the imaging apparatus is a panning determination device that includes a determination unit that determines that panning is being performed.

In the above arrangement, the magnitude of the shake speed of the imaging device is obtained from the first and second shake signals, and the magnitude of the shake speed exceeds a preset limit value, that is, which direction Even if panning is performed, panning determination can be performed under the same conditions.

[0050]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, the present invention will be described in detail based on illustrated embodiments.

1 to 3 are views for explaining a first embodiment of the present invention, and FIG. 1 is a block diagram showing a schematic configuration of an image blur correction system according to the present embodiment.

The main components are a microcomputer 11 which is an image shake correction signal generating means for performing signal processing, a yaw angular velocity sensor 12 which is a first shake detecting means for detecting horizontal shake of the image pickup apparatus, and Pitch angular velocity sensor 13 as second shake detection means for detecting vertical shake, correction optical system 14 as correction means for correcting image shake by performing sliding or rotational displacement, and correction optical system A yaw correction actuator 15 for driving the correction optical system in the horizontal direction, a pitch correction actuator 16 for driving the correction optical system in the vertical direction, a horizontal position sensor 17 for detecting the horizontal position of the correction optical system, and a vertical for detecting the vertical position of the correction optical system. The position sensor 18 is exemplified.

In the same figure, the signal detected by the yaw angular velocity sensor 12 is converted into an angular velocity and a main band of 0.1 to 10 Hz by a camera shake according to the following procedure.
Angular velocity signal Ω _Y in which angular velocities due to panning of 〜1H _Z are mixed, and the main band of angular velocity due to panning (0.1 to
1 Hz) is converted into a low-frequency angular velocity signal Ω _YL mainly taken out, and is taken into the microcomputer 11 activated by the ON signal of the switch SW1.

First, a signal detected by the yaw angular velocity sensor 12 is subjected to a high-pass filter 19 to remove a DC component, and is output to a low-pass filter 110 and a low-pass filter 112 which is a first low-pass filter. The low-pass filter 110 removes high-frequency noise, is digitized by the A / D converter 111, and turns on the switch SW1 as the yaw angular velocity signal Ω _Y.
It is taken into the microcomputer 11 started by a signal. On the other hand, the low-pass filter 112 removes signals in the main band (1 to 10 Hz) of camera shake in addition to high-frequency noise.
The signal is digitized by the A / D converter 113 and is taken into the microcomputer 11 as a low-range yaw angular velocity signal _ΩYL .

The signal detected by the pitch angular velocity sensor 13 is subjected to the same processing, and the pitch angular velocity signal Ω
_P and the low-frequency pitch angular velocity signal Ω _PL are taken into the microcomputer 11.

The low-range angular velocity signals Ω _YL and Ω _PL received by the microcomputer 11 are used as a parameter indicating a panning direction in a sequence described later.

The detection signal of the horizontal position sensor 17 is A / D
It is digitized by the converter 121 and is taken into the microcomputer 11 as a horizontal position signal Y _OUT . The detection signal of the vertical position sensor 18 is also digitized in the same manner, and is taken into the microcomputer 11 as a vertical position signal P _OUT .

The microcomputer 11 obtains a horizontal position Y _IN and a vertical position P _IN to which the correction optical system 14 is to be displaced based on the angular velocity signals Ω _Y and Ω _P, and obtains actual position signals Y _OUT and P _IN respectively.
The differential signal from _OUT is amplified and output as drive signals D _Y and D _P for the correction optical system 14. Alternatively, the driving signals D _Y ,
There is also a method of performing D _P pulse width modulation and outputting as a PWM signal.

The horizontal drive signal _DY is supplied to the D / A converter 11
The signal is converted into an analog signal by the controller 9 and input to the yaw correction actuator 15. If the horizontal drive signal is a PWM signal, it is directly input to the yaw correction actuator 15 without going through the D / A converter 119. Vertical drive signal D _P
Are also input to the pitch correction actuator 16 in the same manner.

The sequence of the image blur correction system having the above configuration will be described with reference to the flowchart shown in FIG. Each step number shown in the figure represents each processing unit.

First, the sequence is started by the input of the ON signal of the switch SW1, and in step 21, the yaw swing angular displacement Ψ and the pitch swing angular displacement Θ are initialized. Next, at step 22, the angular velocity signals Ω _Y and Ω _P , the low-range angular velocity signals Ω _YL and Ω _PL , and the position signals Y _OUT and P of the correction optical system 14
It is checked whether _OUT has been updated, and if it has been updated, the process proceeds to step S23.

Steps 23 and 24 are a panning judgment routine which is a panning judgment means. First, in step 23, which is the shake velocity amount calculation means, the magnitude Ω of the angular velocity vector (Ω _Y , Ω _P ) is determined by the following equation.

Ω = (Ω _Y ² + Ω _P ² ) ^1/2 (1.1) Next, in step 24, a comparison between Ω represented by the equation (1.1) and a predetermined limit value Ω _LMT is performed. Ω is the limit value Ω
If it exceeds _LMT , it is determined that panning is being performed, and the process _proceeds to step S25. Note that the limit value Ω _LMT is generally
The maximum value of the imaginary camera shake angular velocity is set to a value that allows for a certain safety factor.

Here, the reason why the Ω expressed by the equation (1.1) is used for the panning determination is that the high-frequency component which appears at the start of the panning and indicates a sharp increase in the angular velocity due to the panning is used as the low-frequency angular velocity signals Ω _YL , Ω. Angular velocity signal Ω not in _PL
Y, because contained in the Omega _P. Therefore, it is possible to perform the panning determination while minimizing the time lag from the start of the panning to the panning determination.

In step 25, the direction of the low-range angular velocity vector (Ω _YL , Ω _PL ) is regarded as the panning direction, and the panning direction component of the angular velocity vector (Ω _Y , Ω _P ), which is the shake signal, is fixed to zero. In order to separate the components Ω _N in the direction orthogonal to the panning direction in the yaw and pitch directions and obtain the angular velocity signals Ω _YN and Ω _PN to be corrected, calculations are performed according to the following equations.

Ω _L = (Ω _YL ² + Ω _PL ² ) ^1/2 (1.2) Ω _N = (Ω _YL Ω _P −Ω _PL Ω _Y ) / Ω _L (1.3) Ω _YN = −Ω _N Ω _PL / Ω _L (1.4) Ω _PN = Ω _N Ω _YL / Ω _L (1.5) In the above equation, Ω _L is the low-range angular velocity vector (Ω _YL , Ω
_PL ), and Ω _N is the angular velocity vector (Ω _Y , Ω
_P ) represents a component in a direction orthogonal to the panning direction, and Ω _YN and Ω _PN represent a yaw component and a pitch component of Ω _N , respectively. The derivation method of Expressions (1.3) to (1.5) will be described later in detail.

When the arithmetic processing is actually performed, the equations (1.2) to (1.5) are expressed as follows: Ω _L2 = Ω _YL ² + Ω _PL ² (1.6) Ω _{N / L} = (Ω _YL Ω _P −Ω _PL Ω _Y ) / Ω _L2 (1.7) Ω _YN = −Ω _{N / L} Ω _PL (1.8) Ω _PN = Ω _{N / L} Ω _YL (1.9) It is desirable to change the description to reduce the division to reduce the load on the microcomputer 11. In the above equation, Ω _L2 represents Ω _L ² and Ω _{N / L} is a parameter representing Ω _N / Ω _L.

In step 24, Ω expressed by the equation (1.1)
Is _smaller than the limit value Ω _LMT , it is determined that panning has not been performed, and in step 26, the angular velocity signals Ω _Y , Ω
_P is stored as it is as the angular velocity signals Ω _YN and Ω _PN to be corrected.

The angular velocity signals Ω _YN , Ω _PN thus obtained
Is applied to a digital integrator in step 27 to determine deflection angular displacements Ψ and Θ. In this step, a and b are time constants determined by the integrator and constants determined by the sampling period.

The deflection angular displacements Ψ and Θ are calculated in step 28.
Then, the correction optical system 14, which is an image shake correction signal, is converted into a horizontal position Y _IN and a vertical position P _IN to be displaced. In step 29, actual position signals Y _OUT , P _OUT
Are amplified by the loop gains G _Y and G _P , and output from the output terminal of the microcomputer 11 in step 210 as drive signals D _Y and D _P for the correction optical system 14, respectively. It should be noted that since the displacement amount of the correction optical system 14 in the yaw / pitch direction is usually equal for the same shake angle displacement Ψ, Θ,
The conversion coefficient K _Y in step 28 is usually equal to K _P. Furthermore, if the correction actuators 15 and 16 are equivalent, the loop gains G _{Y and} G _P will match. In addition, when the arithmetic processing is actually performed, step 28 is combined with step 27 in one processing unit, and the input position signals Y _IN , _PIN are used without using the swing angular displacements Ψ, 中間 as intermediate parameters. There is no waste in unifying.

On the other hand, at step 211, the switch SW1
Is OFF, and at step 212, a flag that is set when an exposure sequence (not shown) independent of the present sequence is activated is confirmed. If this flag is not set, the sequence is terminated. Otherwise, the process returns to step 22 to repeatedly execute the sequence following this step. If the image blur correction device employing the above sequence is a camcorder, step 21 is executed.
No determination of 2 is required.

Next, the equation (1.
An example of the derivation method of 3) to (1.5) will be described below with reference to FIG.

FIG. 3 shows the angular velocity vectors (Ω _Y , Ω _P ),
The low-frequency angular velocity vectors (Ω _YL , Ω _PL ) and the modified angular velocity vectors (Ω _YN , Ω _PN ) projected on a plane orthogonal to the panning direction with the direction of the low-frequency angular velocity vector regarded as the panning direction Two vector yaw
It is the figure represented in the pitch plane.

In the figure, θ represents the angle of the low-range angular velocity vector (Ω _YL , Ω _PL ) with respect to the yaw axis, and φ represents the angle of the angular velocity vector (Ω _Y , Ω _P ) with the panning direction. For both θ and φ, the illustrated direction is positive.

Now, considering the vector product of the low-range angular velocity vector (Ω _YL , Ω _PL ) and the angular velocity vector (Ω _Y , Ω _P ), the unit vector in the vertical direction from the paper is defined as n, and from the definition: Ω _YL , Ω _PL ) × (Ω _Y , Ω _P ) = Ω _L Ω sinφ · n (1.10) where Ω = (Ω _Y ² + Ω _P ² ) ^1/2 (1.1) Ω _L = (Ω _YL ² + Ω _PL ² ) ^1/2 (1.2) In the above equation, Ω sinφ is the angular velocity vector (Ω
_Y, the Omega _P), since the component in the direction perpendicular to the panning direction (in the drawing Omega _N direction), placing it with Omega _N,
Equation (1.10) can be expressed as (Ω _YL , Ω _PL ) × (Ω _Y , Ω _P ) = Ω _L Ω _N · n (1.11) where Ω _N = Ω sin φ (1.12). Note that the sign of Ω _N changes with the sign of sin φ, and the direction shown is positive.

On the other hand, the low-range angular velocity vector (Ω _YL , Ω _PL )
The vector product of the angular velocity vector (Ω _Y , Ω _P ) can be expressed by the following equation.

(Ω _YL , Ω _PL ) × (Ω _Y , Ω _P ) = (Ω _YL Ω _P −Ω _PL Ω _Y ) · n (1.13) The above equations (1.11) and (1.13) ) Leads to the following equation:

Ω _L Ω _N = Ω _YL Ω _P −Ω _PL Ω _Y (1.14) By dividing both sides of the expression (1.14) by Ω _L , the expression (1.3) Ω _N = (Ω _YL Ω _P −Ω _PL Ω _Y ) / Ω _L (1.3) is obtained.

From FIG. 3, it can be understood that the angle formed by the direction of Ω _N with respect to the yaw axis is θ + π / 2, so that the corrected angular velocity signals Ω _YN and Ω _PN are expressed as follows using Ω _N. Is done.

Ω _YN = Ω _N cos (θ + π / 2) = − Ω _N sin θ (1.15) Ω _PN = Ω _N sin (θ + π / 2) = Ω _N cos θ (1.16) where sin θ, cos θ Can be understood from FIG. 3 as follows.

Sin θ = Ω _PL / Ω _L (1.17) cos θ = Ω _YL / Ω _L (1.18) Equations (1.17) and (1.18) are each converted into the equation (1.1
5) and (1.16), the expression (1.
4), (1.5) Ω _YN = −Ω _N Ω _PL / Ω _L (1.4) Ω _PN = Ω _N Ω _YL / Ω _L (1.5)

Finally, in order to understand the behavior of the correction optical system 14 during panning in the sequence shown in FIG. 2, consider the behavior of the shake angular displacements Ψ, 後 after the start of panning.

Immediately after the start of panning, the angular velocity vector (Ω
Until the magnitude Ω of _Y , Ω _P ) suddenly increases and exceeds the limit value Ω _LMT , the angular velocity signals Ω _Y , Ω _P are regarded as the vibrations to be corrected, and the coefficient a is added to the vibration angular displacements Ψ, で in step 27, respectively. Is added for each sampling. During this time,
Image shake due to fluctuation of shake angular displacements Ψ and Θ is corrected by optical system 1
Therefore, it is determined that the panning is in progress from the start of panning by performing the setting of the limit value Ω _LMT in the panning determination in step 24 as strictly as possible. Time lag until the following, and the loss of followability can be suppressed to a level that is hardly noticeable.

The magnitude Ω of the angular velocity vector (Ω _Y , Ω _P )
Instantaneously exceeds the limit value Ω _LMT , in step 25, the panning direction components of the angular velocity vectors (Ω _Y , Ω _P ) are fixed to zero as shown in FIG. The component Ω _N is separated as it is into the yaw / pitch direction components Ω _YN and Ω _PN as a shake to be continuously corrected.

The digital integration of the angular velocity signal to be corrected is performed in a coordinate system whose axis is the panning direction and a direction orthogonal to the panning direction. Ψ _pan ,
The deflection angle displacement in the direction perpendicular to the panning direction
_{As pan} , it is expressed as follows.

[0086] _{Ψ pan = bΨ pan (1.19)} Θ pan = aΩ N + bΘ pan (1.20) Detailed description is omitted, but represents the above formula in vector form, for the both sides, as shown in FIG. 3 When the rotational conversion of θ is performed on the yaw / pitch plane, the yaw /
It corresponds to the difference equation of digital integration in two directions of pitch. Therefore, the correction optical system 1 is obtained from the behavior of the shake angular displacement Ψ _pan , Θ _pan based on the equations (1.19) and (1.20).
The behavior of No. 4 may be judged.

First, in the equation (1.19), the constant b is usually set to 1 or less, and thereafter, the panning is completed and the magnitude Ω of the angular velocity vector (Ω _Y , Ω _P ) becomes the limit value Ω _LMT. , The panning direction deflection angular displacement Ψ
_pan will gradually _approach zero. That is, the angular velocity vector (Ω _Y, Ω _P) until below the size Omega is a limit value Omega again limits since the moment of exceeding the _LMT Omega _LMT, the correction optical system 14 is panning direction of the image blur correction operation is stopped Then, the centering operation of gently moving toward the center position in the panning direction is performed while correcting the panning direction every time the panning direction changes. During this time, the ability to follow the subject in the panning direction is maintained.

On the other hand, the equation (1.20) is orthogonal to the panning direction because there is no change before and after the moment when the magnitude Ω of the angular velocity vector (Ω _Y , Ω _P ) exceeds the limit value Ω _LMT. The image shake correction in the direction is continuously performed without hindering the ability to follow the subject in the panning direction. Therefore, the maximum image blur correction effect can be expected while maintaining the ability to follow the subject in the panning direction.

In this embodiment, as described above, the panning direction is detected by the angular velocity sensor 12,
Although the detection is performed using the detection signal of No. 13, the detection may be performed by a separately provided dedicated sensor for detecting the panning direction.

If the panning is being performed, the panning direction does not need to be updated every sampling. For example, every 0.1 second (if the sampling frequency is 1 KHz, the panning direction is 1).
The update may be performed every 00 samplings).

Further, the shake detecting means and the panning direction detecting means are not limited to an angular velocity sensor for detecting an angular velocity signal, but may be means for detecting an angular displacement signal or an image shake signal.

Further, in this embodiment, the correction optical system as the image shake correcting means by the position control has been described as an example. However, the present invention is not limited to this, and the control method of the image shake correcting means is speed control, position control and speed control. A combined control method may be used.

The image blur correction means is not limited to a correction optical system that bends the photographing optical axis in accordance with the image blur, but is a means for displacing the image capturing means or changing the recording range of the image signal in accordance with the image blur. There may be.

(Second Embodiment) In the first embodiment, a low-frequency component in which the main band (0.1 to 1 Hz) of the angular velocity due to pan is mainly extracted from the detection signals of the angular velocity sensors 12 and 13 is used. The direction of the area angular velocity vector (Ω _YL , Ω _PL ) is regarded as the panning direction, and an image shake correction signal is generated according to the direction.

However, in this method, since the high-frequency components which appear at the start of the panning and indicate a sharp increase in the angular velocity due to the panning are removed by the low-pass filters 112 and 117, the low-range angular velocity signals Ω _YL and Ω _PL are included in the panning start signals. An angular velocity signal due to panning at the time is hardly included, and particularly at the start of panning, there occurs a problem that the reliability of the panning direction obtained from the low-range angular velocity vectors (Ω _YL , Ω _PL ) is reduced.

This is illustrated by the typical values of the magnitude Ω of the angular velocity vector (Ω _Y , Ω _P ) and the magnitude Ω _L of the low-range angular velocity vector (Ω _YL , Ω _PL ) shown in FIG. This will be described below based on the time series change.

In the figure, the time at the moment when the panning is started is set to zero, and the magnitude Ω of the angular velocity vector (Ω _Y , Ω _P ) is set.
There time t ₁ the moment of exceeding the limit value Omega _LMT, the angular velocity vector (Ω _Y, Ω _P) size of Omega is the time t ₂ the moment falls below the limit value Omega _LMT. Accordingly, during the period from time t ₁ to time t ₂ is the fact that the section is determined that the panning.

In the figure, the sharp increase in the magnitude Ω of the angular velocity vector (Ω _Y , Ω _P ) from time zero to time t ₁ is caused by the influence of the low-pass filters 112 and 117 and the low-frequency angular velocity vector (Ω _YL) , Ω _PL ) is not much reflected in the magnitude Ω _L. That is, at time t ₁ , when the magnitude Ω of the angular velocity vector (Ω _Y , Ω _P ) reaches the limit value Ω _LMT , the low-range angular velocity signals Ω _YL , Ω _PL are both still close to zero and the main band ( (1 to 10 Hz). Therefore,
It is highly probable that the direction of the low-range angular velocity vector (Ω _YL , Ω _PL ) at time t ₁ is considerably deviated from the actual panning direction. In an extreme example, in FIG. 3, the angular velocity vectors (Ω _Y , Ω _P) ) Is the angle φ to the panning direction
Is π / 2 radians (90 °), the corrected angular velocity vector (Ω _YN , Ω _PN ) to be corrected matches the angular velocity vector (Ω _Y , Ω _P ), and follows the object in the actual panning direction. Not only is the image quality not improved, but also the image blur in the direction orthogonal to the panning direction is not corrected.

Therefore, in the present embodiment, at time t ₁ , the direction of the angular velocity vector (Ω _Y , Ω _P ) rather than the direction of the low-range angular velocity vector (Ω _YL , Ω _PL ) is set in the actual panning direction. focusing on close to the angular velocity vector (Ω _y, Ω _P) at time t ₁ so that the orientation of the panning initial direction of the start panning, is prior to time t _1, the low-frequency angle signal Omega _YL, the Omega _PL The time constant of the low-pass filter for obtaining is determined so as to match the time constant of the low-pass filter for obtaining the angular velocity signals Ω _Y and Ω _P, and after time t ₁ ,
_Switching the time constant of the low filter for obtaining the low-band angular velocity signals Ω _YL and Ω _PL as shown in the modified curve of Ω _L in FIG. 4 so as to remove high-frequency components above the main band of camera shake. Gives the low-range angular velocity vector (Ω
_YL , Ω _PL ) so as to be as close as possible to the actual panning direction.

FIGS. 5 to 10 are diagrams for explaining a second embodiment of the present invention, and FIG. 5 is a block diagram showing a schematic configuration of an image blur correction system according to the present embodiment.

The main components are a microcomputer 51 serving as image shake correction signal generating means for performing signal processing, a yaw angular velocity sensor 52 serving as first shake detecting means for detecting horizontal shake of the imaging apparatus, and Pitch angular velocity sensor 53 as second shake detecting means for detecting vertical shake, correction optical system 54 as correction means for correcting image shake by performing sliding or rotational displacement, and correction optical system A yaw correction actuator 55 for driving the correction optical system in the horizontal direction, a pitch correction actuator 56 for driving the correction optical system in the vertical direction, a horizontal position sensor 57 for detecting the horizontal position of the correction optical system, and a vertical for detecting the horizontal position of the correction optical system. A position sensor 58;

In the figure, the components other than the low-pass filters 512 and 517, which are the first and second low-pass filters, are the same as those in the first embodiment, so that the description is omitted. The functions of the filters 512 and 517 will be described below.

In the low-pass filters 512 and 517, when it is determined that panning is being performed in a panning determination routine in a sequence to be described later, a signal of about 1 Hz or less, which contains almost no angular velocity signal mainly due to camera shake, passes. Otherwise, low-pass filters 510,5
Similarly to 15, the microcomputer 51 sends a time constant switching signal S for switching the time constant (cutoff frequency) from the microcomputer 51 so that a signal of about 10 to 30 Hz or less including an angular velocity signal due to both camera shake and panning passes. Is entered.

Such low-pass filters 512, 51
FIG. 6 shows an example (known) of the internal configuration of FIG.

The low-pass filters 512 and 5 in FIG.
Reference numeral 17 is a circuit in which switching elements 66 and 67 are respectively incorporated in series with resistors 64 and 65 having a resistance value R, which are circuit components of a known low-pass filter.

Time constant switching signal S input from terminal 68
Is a PWM signal in which the ratio of the ON state (duty ratio) in one cycle is represented by r _d , and if the panning is in progress, the time constant switching signal S converted to PWM with the duty ratio r _d = r ₂ is the same. Otherwise, the PWM-converted time constant switching signal S with the duty ratio r _d = r ₁ is input to the switching elements 66 and 67 through the terminal 68.

[0107] The effective value of the current flowing at this time resistor 64 or the resistor 65, the switching element 66, 67 is r _d times the current in an ON state at all times, the resistance 64 and the switching element 66, or resistor the combination of 65 and the switching element 67, the resistance value can be regarded to be equivalent to the resistance of R / r _d.

Therefore, the low-pass filters 512 and 517
The cutoff frequency f _c, if not panning, if a _{f c = r 1 / (2πRC} ) (2.1) in _{panning, f c = r 2 / (} 2πRC) and (2.2) Become. In the above, C represents the capacitance of the capacitors 62 and 63 in FIG.

[0109] Thus, switching the constant is performed when by changing the duty ratio r _d by the results of the panning determination. That is, in equation (2.1), f _c =
R ₁ is determined to be 10 to 30 Hz, and the equation (2.
In 2), r ₂ is determined so that f _c ≒ 1 Hz.

The low-pass filters 512 and 517
Can realize the same function as the configuration shown in FIG. 6 also by configuring as shown in FIG.

The low-pass filters 512 and 5 in FIG.
17, the two resistors which are components of a circuit of a known low-pass filter of the resistor 76 and 77 the resistance value R ₂ if panning, otherwise the resistance of the resistance value R ₁ 7
It incorporates switching elements 78 and 79 for switching to 4,75. The switching signal S of the switching elements 78 and 79 is a simple signal that is maintained for two switching positions unless the panning determination is changed, and is input from the microcomputer 51 through the terminal 68.

Therefore, the low-pass filters 512 and 517
The cutoff frequency f _c, if not being _{panned, f c = 1 / (2πR} 1 C) (2.3) if it is being _{panned, f c = 1 / (2πR} 2 C) (2.4 ). In the above equation, C represents the capacitance of the capacitors 72 and 73 in FIG.

As described above, the time constant is switched by switching the resistance in the low-pass filters 512 and 517 according to the result of the panning determination. That is, in equation (2.3), R ₁ is determined so that f _c = 10 to 30 Hz, and in equation (2,4), f _c ≒
R ₂ is determined to be 1 Hz.

The sequence of the image blur correction system having the configuration shown in FIGS. 5 and 6 will be described with reference to the flowchart shown in FIG. Each step number shown in the figure represents each processing unit.

Steps 81 to 83 at the beginning of the sequence are as follows:
Steps 21 to 23 in FIG. 2 are the same as those in FIG.

[0116] At step 84, performs a comparison between the angular velocity vector (Ω _Y, Ω _P) in size Omega with a predetermined limit value Omega _LMT obtained in step 83, Omega exceeds the limit value Omega _LMT If it is determined that panning is in progress, step 8
Move to 5. It should be noted that the limit value Ω _LMT is generally set to a value that allows for a slight safety factor to the maximum value of the imaginary camera shake angular velocity.

In step 85, in order to obtain an accurate panning direction from the low-frequency angular velocity vector (Ω _YL , Ω _PL ),
As described in the description of FIG. 6, the duty ratio r _d of the time constant switching signal S is set to r ₂ and the PWM of the duty ratio r _d
A signal S is generated and output from an output terminal of the microcomputer 51.

In step 86, the direction of the low-frequency angular velocity vector (Ω _YL , Ω _PL ) is regarded as panning, and the panning direction component of the angular velocity vector (Ω _Y , Ω _P ), which is the shake signal, is fixed to zero. In order to separate the components Ω _N in the direction perpendicular to the panning direction in the yaw and pitch directions and obtain the angular velocity signals Ω _YN and Ω _PN to be corrected, the equation (1) described in the first embodiment is used. .6)-
The calculation is performed according to (1.9).

As described above, even if the time constants of the low-pass filters 512 and 517 are switched depending on whether panning is in progress or not, the direction of the low-frequency angular velocity vector (Ω _YL , Ω _PL ) and the actual panning direction Assuming that the deviation is large,
In steps 88, 89, and 810, a final reliability check in the panning direction is performed. As a criterion of reliability,
If the corrected angular velocity vector (Ω _YN , Ω _PN ) starting from the origin is within the circle of the radius Ω _LMT , if not, the reliability of the panning direction is insufficient. And the angular velocity signal Ω _YN ,
Do not perform image shake correction based on Ω _PN .

[0120] As the method, first step 87, the angular velocity vector (Ω _Y, Ω _P) seek direction component Omega _N perpendicular to the panning direction, the Omega _N in step 89
If the absolute value of exceeds Ω _LMT , it is determined that the reliability in the panning direction is insufficient, and in step 810 Ω _YN and Ω _PN are fixed to zero.

On the other hand, in step 84, if the magnitude Ω of the angular velocity vector (Ω _Y , Ω _P ) does not exceed the limit value Ω _LMT , it is determined that panning is not performed, and in step 810, FIG. As described above, the duty ratio r _d of the time constant switching signal S is set to r ₁ to generate the PWM signal S having the duty ratio r _d , output from the output terminal of the microcomputer 51, and , the angular velocity signal Omega _Y, angular velocity signal to be directly corrected Ω _P Ω _YN, Ω _PN
Store as

Steps 812 to 812 in the subsequent sequence
818 corresponds to steps 27 to 29 and 210 to 212 in FIG.
Therefore, the description is omitted.

The low-pass filters 512 and 51 described above
7, the low-pass filters 512, 5
17, a switching element is provided, and the time constant is switched by inputting the switching signal from the microcomputer 51. According to this method, the low-pass filter 51 is provided.
The structure of 2,517 is complicated, which causes an increase in cost.

Therefore, as a modification of the second embodiment, the low-pass filters 512 and 517 are taken in the sequence as digital filters, and
2, 517 and a method of eliminating the output terminal of the time constant switching S of the microcomputer 51 can be considered. FIG. 9 is a block diagram illustrating a schematic configuration of an image shake correction system according to the modification.

The configuration shown in the figure is different from the configuration shown in FIG. 5 in that low-pass filters 510, 515, 512, 517 and A /
D converters 513 and 518 and a time constant switching signal S
Has the same configuration as that from which the output terminal has been removed.

Therefore, the output signals of the high-pass filters 99 and 911 are directly supplied to the A / D converters 910 and 9 respectively.
12 and digitized angular velocity signals Ω _Y0 and Ω _P0 before passing through a low-pass filter.
It is taken into 1.

A sequence of the image blur correction system having the above configuration will be described with reference to a flowchart shown in FIG. Each step number shown in the figure represents each processing unit.

First, the sequence is started by the input of the ON signal of the switch SW1, and in order to use it as the initial value of the output signal of the digital filter, step 1001
Samples the angular velocity signals Ω _Y0 and Ω _P0 .

Using the values of Ω _Y0 and Ω _P0 obtained in the same step, in step 1002, steps 1004, 1
Initialization of the output signals Ω _Y , Ω _P , Ω _YL , Ω _PL of the digital low-pass filter in 007 and 1012, and the initial values of the yaw deflection angular displacement ψ and the pitch deflection angular displacement あ る which are the output signals of the digital integrator in step 1014. Is performed.

Next, in step 1003, the angular velocity signals Ω _Y0 , Ω _P0 , the position signals Y _OUT , P _{OUT of the} correction optical system 94
Is checked to see if it has been updated, and if it has been updated, the process proceeds to step 1004.

Step 1004 represents a well-known digital filter that executes processing in place of the low-pass filters 510 and 515 in FIG. 5, and performs filtering for removing high-frequency noise from the angular velocity signals Ω _Y0 and Ω _P0 by the following equation. .

Ω _y = A ₁ Ω _Y0 + B ₁ Ω _Y (2.5) Ω _p = A ₁ Ω _p0 + B ₁ Ω _p (2.6) In the above equation, A ₁ B ₁ is determined in the same filter. It is a constant determined by a time constant and a sampling period.

Steps 1005 and 1006 are a panning judgment routine which is panning judgment means.

First, in step 1005, the magnitude Ω of the angular velocity vector (Ω _Y , Ω _P ) is determined by equation (1.1), as in step 23 of FIG.

Next, at step 1006, the magnitude Ω of the angular velocity vector (Ω _Y , Ω _P ) and the predetermined limit value Ω
Comparison with _LMT is performed. If Ω exceeds the limit value Ω _LMT , it is determined that panning is being performed, and the process _proceeds to step 1007. Note that the limit value Ω _LMT is generally set to a value that allows for a slight safety factor to the maximum value of the imaginary camera shake angular velocity.

Step 1007 represents a well-known digital filter that performs processing in place of the low-pass filters 512 and 517, which are the first and second low-pass filters in FIG. 5, and the angular velocity is represented by the following equation. Signal Ω
Filtering is performed to remove high frequency components from _Y0 and Ω _{P0 that} are higher than the main band of angular velocity due to camera shake.

Ω _YL = A ₂ Ω _Y0 + A ₂ Ω _YL (2.7) Ω _PL = A ₂ Ω _P0 + B ₂ Ω _PL (2.8) In the above equation, A ₂ . B ₂ is a time constant determined for the filter and a constant determined by the sampling period.

Steps 1008 to 1011 and 1013
Are the same as steps 87 to 89 and 811 in FIG.

On the other hand, in step 1006, it is determined that panning is not performed unless the magnitude Ω of the angular velocity vector (Ω _Y , Ω _P ) exceeds the limit value Ω _LMT .
In Step 1012, as shown in the following equation, Step 10
The angular velocity signals Ω _Y0 and Ω _P0 are filtered with the same time constant as the digital filter shown in FIG. 04 to obtain low-range angular velocity signals Ω _YL and Ω _PL .

Ω _YL = A ₁ Ω _Y0 + B ₁ Ω _YL (2.9) Ω _PL = A ₁ Ω _P0 + B ₁ Ω _PL (2.10) That is, in this sequence, the panning judgment in step 1006 is based on the judgment result. The time constant of the digital filter for obtaining the low-range angular velocity signals Ω _YL and Ω _PL is switched by branching.

Step 1014 as a subsequent sequence
Steps 10 to 1019 correspond to steps 27 to 29 and 210 to 2 in FIG.
12, the description is omitted.

In the second embodiment described above, if it is not determined that panning is being performed, a low-band angular velocity vector (Ω _YL , Ω _PL ) regarded as a panning direction contains a component in a band higher than usual. In addition, since the time constant of the low-pass filter is switched, the effect of improving the reliability of the panning direction is obtained, particularly at the start of panning or at the turning point of the panning, and the followability to the subject in the panning direction is further improved. The image blur correction effect in the direction orthogonal to the panning direction is also enhanced.

In this embodiment, the panning direction is detected by the angular velocity sensor 5 serving as the shake detecting means as described above.
Although the detection is performed using the detection signals 2 and 53, the detection may be performed by a separately provided sensor dedicated to panning direction detection.

Also, during panning, the panning direction does not need to be updated every sampling, for example, every 0.1 seconds (if the sampling frequency is 1 KHz,
The update may be performed every 00 samplings).

Further, the shake detecting means and the panning direction detecting means are not limited to an angular velocity sensor for detecting an angular velocity signal, but may be means for detecting an angular displacement signal or an image shake signal.

Further, in the present embodiment, the correction optical system as the image shake means by position control has been described as an example. However, the present invention is not limited to this, and the control method of the image shake correction means is speed control, position control and speed control. A combined control method may be used.

The image shake correcting means is not limited to a correction optical system that bends the photographing optical axis in accordance with the image shake. There may be.

Here, the effects of each of the above embodiments will be enumerated below.

(1) Since the image blur correction is performed in consideration of the panning direction which has not been used in the related art, an appropriate image blur correction effect can be obtained regardless of the panning direction.

(2) If it is determined that panning is being performed, an image shake correction signal is generated based only on a shake signal component in a direction orthogonal to the panning direction. Even if the panning operation is performed, an image blur correction effect in a direction orthogonal to the panning direction can be obtained without impairing the panning operation.

(3) Since the swing direction of the low-frequency component extracted from the pitch and yaw direction shake signals is set as the panning direction, a new shake detection sensor is not required for detecting the panning direction. In addition, an appropriate image blur correction effect can be obtained by slightly changing the control algorithm for image blur correction.

(4) When it is determined that panning is being performed, the cutoff frequency of each low-pass filter that extracts low-frequency components from the pitch and yaw direction shake signals is lower than when it is not determined that panning is being performed. As described above, since the time constant is switched, the effect of improving the reliability of the panning direction is obtained, especially after the start of panning or at the turning point of the panning, and the followability to the subject in the panning direction is further improved, and the panning is performed. The image blur correction effect in the direction orthogonal to the direction is also enhanced.

(5) Since the camera is determined to be panning when the magnitude of the camera shake speed obtained from the shake signals in the pitch and yaw directions exceeds a preset limit value, it depends on the panning direction. No, fair panning determination can be realized, and even when the panning size is the same, it is possible to prevent a phenomenon that the followability to the subject changes depending on the panning direction.

(Correspondence between the Invention and the Embodiments) In each of the above embodiments, the microcomputers 11, 51, and 91 use the image blur correction signal generating means, the panning determining means, the panning direction detecting means, the signal combining means, Or yaw angular velocity sensors 12, 5.
2. The pitch angular velocity sensors 13 and 53 correspond to the first and second shake detecting means of the present invention, and the correction optical systems 14, 54 and 9
Reference numeral 4 corresponds to the image blur correcting means of the present invention, and the low-pass filters 112, 117, 512, and 517 correspond to low-pass filters.

The correspondence between the components of the embodiment and the components of the present invention has been described above. However, the present invention is not limited to the configuration of the embodiment, and the functions and features described in the claims are not limited.
Needless to say, any configuration may be used as long as the functions of the embodiment can be achieved.

(Modifications) The present invention relates to various types of cameras such as a single-lens reflex camera, a video camera, and an electronic still camera, as well as optical devices and other devices other than cameras, and furthermore, these cameras, optical devices and other devices. The present invention can be applied to a device applied to the device described above, or an element constituting the device.

Further, the present invention may have a configuration in which the above embodiments or their techniques are appropriately combined.

[0158]

As described above, according to the present invention,
It is possible to provide an image blur correction device that can obtain an appropriate image blur correction effect even if the imaging device is panned in any direction.

Further, according to the present invention, it is possible to provide an image shake correcting apparatus capable of detecting a panning direction without providing a dedicated shake detecting means for detecting a panning direction.

Further, according to the present invention, it is possible to provide an image blur correction device capable of improving the reliability of the detected panning direction.

Further, according to the present invention, even if the imaging apparatus is panned in any direction, an image which can obtain the maximum image blur correction effect while keeping the followability to the subject in the panning direction as much as possible. It is possible to provide a shake correction device.

Further, according to the present invention, even if the imaging apparatus is panned in any direction, the image blur correction effect in the direction orthogonal to the panning direction can be obtained while maintaining the ability to follow the subject in the panning direction. An image blur correction device that can be obtained can be provided.

Further, according to the present invention, it is possible to provide an image shake correcting apparatus which can eliminate the direction dependency of the panning determination and can accurately perform the panning determination in any direction.

Further, according to the present invention, it is possible to provide an image shake correcting apparatus capable of performing image shake correction only in a direction in which image shake correction is preferentially performed.

Further, according to the present invention, it is an object of the present invention to provide a panning determination device capable of accurately performing panning determination in any direction.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a schematic configuration of an image shake correction system according to a first embodiment of the present invention.

FIG. 2 is a flowchart illustrating an operation of the image blur correction system according to the first embodiment of the present invention.

FIG. 3 is a diagram illustrating a shake signal component in a panning direction and a direction orthogonal to the panning direction in the first embodiment of the present invention.

FIG. 4 is a diagram showing typical time-series changes in the magnitude of an angular velocity vector and the magnitude of a low-frequency angular velocity vector during panning.

FIG. 5 is a block diagram illustrating a schematic configuration of an image shake correction system according to a second embodiment of the present invention.

FIG. 6 is a circuit diagram showing an example of an internal configuration of the low-pass filter in FIG.

FIG. 7 is a circuit diagram showing another example of the internal configuration of the low-pass filter in FIG.

FIG. 8 is a flowchart showing an operation of the image blur correction system according to the second embodiment of the present invention.

FIG. 9 is a block diagram showing an example in which a part of FIG. 5 is modified.

FIG. 10 is a flowchart showing a part of the operation in the case of the configuration shown in FIG. 9;

FIG. 11 is a flowchart showing a continuation of the operation in FIG. 10;

FIG. 12 is a block diagram illustrating a schematic configuration of a conventional image blur correction system.

13 is a flowchart showing the operation of the image shake correction system of FIG.

[Explanation of symbols]

11, 51, 91 Microcomputer 12, 52 Yaw angular velocity sensor 13, 53 Pitch angular velocity sensor 14, 54, 94 Correction optical system (image blur correction means) 15, 55, 95 Yaw correction actuator 16, 56, 96 Pitch correction actuator

Claims (9)

[Claims] 1. A first and a second shake detecting means for detecting a shake of the imaging device around two different axes orthogonal to a photographing optical axis of the imaging device, and whether the imaging device is panning. Panning determination means for determining whether or not the panning direction is detected, and panning direction detection means for detecting a panning direction of the imaging apparatus based on each of the first and second shake signals detected by the first and second shake detection means. , The first and second detected by the first and second shake detecting means.
Shake signal, the determination result by the panning determination means,
And an image shake correction signal generation unit that generates an image shake correction signal based on information on a panning direction detected by the panning direction detection unit, and captures an image of the imaging apparatus by inputting the image shake correction signal. An image shake correction apparatus, comprising: an image shake correction unit that performs image shake correction. 2. The panning direction detecting means according to claim 1, wherein said first and second shake detecting means detect said first and second shake detecting means.
The apparatus according to claim 1, wherein a panning direction of the imaging device is detected based on each of the shake signals. 3. The panning direction detecting means removes a high-frequency component from the first and second shake signals and outputs the first and second shake signals.
A first low-pass filter that outputs a second low-pass shake signal; and a determination unit that determines a shake direction obtained from the first and second low-pass shake signals as a panning direction. 3. The image blur correction device according to claim 2, wherein: 4. The first and second low-pass filters include:
If it is determined that the panning is being performed by the panning determination unit, the time constant is determined so that high-frequency components including the main band of camera shake are substantially removed from the first and second shake signals, and the panning is determined by the panning determination unit. 4. The image blur correction device according to claim 3, wherein, if it is not determined, the time constant is determined so that the cutoff frequency becomes higher than that in the case where it is determined that panning is being performed. 5. The image blur correction signal generating means, if the panning determination means does not determine that panning is being performed, generates an image blur signal based only on the first and second shake signals. If it is determined by the means that panning is being performed, different processing is performed on the shake signal component in the panning direction and the shake signal component in a direction orthogonal to the panning direction, and image shake correction is performed based on the obtained signal. 5. The image blur correction device according to claim 1, wherein a signal is generated. 6. The image shake correction signal generation means generates an image shake correction signal based only on a shake signal component in a direction orthogonal to the panning direction when the panning determination means determines that panning is being performed. The image blur correction device according to claim 5, wherein 7. A shake speed amount calculating means for obtaining a magnitude of a shake speed vector of the imaging apparatus from first and second shake signals detected by the first and second shake detecting means, The panning determination unit determines that the imaging apparatus is performing panning when the magnitude of the shake speed vector obtained by the shake speed amount calculation unit exceeds a preset limit value. The image blur correction device according to claim 1, 2, 3, 4, or 5. 8. A first and second shake detecting means for detecting a shake of the imaging device around two different axes orthogonal to a photographing optical axis of the imaging device, and the first and second shake detection. Signal synthesizing means for forming a signal corresponding to a shake in a direction different from the first and second directions by synthesizing the first and second shake signals detected by the means; Image shake correction signal generation means for generating an image shake correction signal based on the shake signal component in the predetermined direction,
When the image shake correction signal is input, performs a shake correction of a captured image of the imaging device, but an image shake correction unit,
An image shake correcting apparatus, comprising: a variable unit configured to change the predetermined direction. 9. A first and second shake detecting means for detecting shake of the imaging device around two different axes orthogonal to a photographing optical axis of the imaging device, and the first and second shake detection. Means for calculating the magnitude of the shake velocity vector of the imaging device by combining the first and second shake signals detected by the means, and a shake velocity vector obtained by the shake speed amount calculation means A determining unit that determines that the imaging apparatus is performing panning when the size of the image exceeds a preset limit value.