Vibration-proof controller and method

Abstract

To more stably alleviate the influence of gravitational acceleration on shake correction, while reducing the cost of an angular velocity meter and an acceleration meter.SOLUTION: A vibration-proof controller has: first input means that inputs a translation shake signal indicating a translation shake in a first direction; second input means that inputs a rotational shake signal indicating a rotational shake around a first axis intersecting a direction of gravity and the first direction; shake amount acquisition means that, on the basis of the rotational shake signal, adjusts and corrects a gain applied to the translation shake signal to determine a shake amount in the first direction; and target value acquisition means that determines a target value for correcting a shake in the first direction on the basis of the shake amount determined by the shake amount acquisition means.SELECTED DRAWING: Figure 22

Description

[Technical Field]

[0001] The present invention relates to an image stabilization control device and method, and more particularly to a technique for reducing the effect of gravitational acceleration on image stabilization. [Background technology]

[0002] Conventionally, in an imaging device, a technology has been used for navigation purposes, for example, in which acceleration in three axial directions and angular velocity around three axes are detected, and a matrix operation is performed on these signals to determine the amount of variation in gravitational acceleration added to acceleration in the X direction (see Patent Document 1).

[0003] Furthermore, Patent Document 2 discloses a method for reducing image degradation caused by gravitational acceleration using an accelerometer. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP2014-021464 Public Relations [Patent Document 2] Patent No. 5675179 Summary of the Invention [Problem to be solved by the invention]

[0005] However, the technology disclosed in Patent Document 2 requires a highly accurate angular velocity meter and accelerometer, and also requires sufficient time for stabilization before calculation results can be obtained, making it unsuitable for devices such as cameras that are often carried around and take photographs frequently.

[0006] The present invention has been made in consideration of the above problems, and has as its object to more stably mitigate the effect of gravitational acceleration on shake correction while reducing the costs of angular velocity sensors and accelerometers. [Means for solving the problem]

[0007] In order to achieve the above object, the vibration isolation control device of the present invention has a first input means for inputting a translational shake signal indicative of translational shake in a first direction, a second input means for inputting a rotational shake signal indicative of rotational shake about a first axis intersecting the direction of gravity and the first direction, a shake amount acquisition means for determining the amount of shake in the first direction by adjusting and correcting a gain to be applied to the translational shake signal based on the rotational shake signal, and a target value acquisition means for determining a target value for correcting shake in the first direction based on the amount of shake determined by the shake amount acquisition means. [Effects of the Invention]

[0008] According to the present invention, it is possible to reduce the cost of the angular velocity sensor and the accelerometer, and to more stably reduce the effect of gravitational acceleration on shake correction. [Brief explanation of the drawings]

[0009] [Figure 1] FIG. 1 is a side view of a camera according to a first embodiment of the present invention. [Figure 2] FIG. 2 is a top view of the camera according to the first embodiment. [Figure 3] FIG. 2 is a front view of the camera according to the first embodiment. [Figure 4] FIG. 10 is a diagram illustrating the influence of gravitational acceleration on an X acceleration signal. [Figure 5] FIG. 4 is a block diagram showing a functional configuration for calculating an X shake correction target value by reducing a gravitational acceleration component superimposed on an X acceleration signal in the first embodiment. [Figure 6] 10 is a graph for explaining a method for determining an X-shake acceleration in which the influence of gravitational acceleration is alleviated in the first embodiment. [Figure 7] 5 is a flowchart showing a method for correcting X-shake in the first embodiment. [Figure 8] FIG. 10 is a side view of the camera according to the second embodiment. [Figure 9] FIG. 10 is a top view of the camera according to the second embodiment. [Figure 10]FIG. 11 is a block diagram showing a functional configuration for calculating an X shake correction target value by reducing a gravitational acceleration component superimposed on an X acceleration signal in the second embodiment. [Figure 11] 10 is a graph illustrating a method for determining an X-shake velocity in which the influence of gravitational acceleration is alleviated in the second embodiment. [Figure 12] 10 is a flowchart showing a method for correcting X-shake in the second embodiment. [Figure 13] FIG. 11 is a block diagram showing a functional configuration for calculating an X shake correction target value by reducing a gravitational acceleration component superimposed on an X acceleration signal in the third embodiment. [Figure 14] FIG. 11 is a block diagram showing a functional configuration for calculating an X shake correction target value by reducing a gravitational acceleration component superimposed on an X acceleration signal in another form of the third embodiment. [Figure 15] FIG. 13 is a block diagram showing a functional configuration for calculating an X shake correction target value by reducing a gravitational acceleration component superimposed on an X acceleration signal in the fourth embodiment. [Figure 16] FIG. 13 is a block diagram showing a functional configuration for calculating an X shake correction target value by reducing a gravitational acceleration component superimposed on an X acceleration signal in the fifth embodiment. [Figure 17] 13 is a graph illustrating a method for removing a gravitational acceleration component from an X acceleration signal in the fifth embodiment. [Figure 18] FIG. 13 is a block diagram showing the functional configuration of a phase determination unit in the fifth embodiment. [Figure 19] FIG. 13 is a front view of a camera according to a sixth embodiment. [Figure 20] FIG. 23 is a block diagram showing a functional configuration for calculating an X shake correction target value by reducing the gravitational acceleration component superimposed on the X acceleration signal in the sixth embodiment. [Figure 21] FIG. 13 is a block diagram showing a functional configuration for calculating an X shake correction target value by reducing a gravitational acceleration component superimposed on an X acceleration signal in another form of the sixth embodiment. [Figure 22] FIG. 23 is a block diagram showing a functional configuration for calculating an X shake correction target value by reducing the gravitational acceleration component superimposed on the X acceleration signal in the seventh embodiment. [Figure 23] FIG. 13 is a block diagram showing a functional configuration for calculating an X shake correction target value by reducing a gravitational acceleration component superimposed on an X acceleration signal in another form of the seventh embodiment. [Figure 24] FIG. 13 is a block diagram showing a functional configuration for calculating an X shake correction target value by reducing a gravitational acceleration component superimposed on an X acceleration signal in another form of the seventh embodiment. [Figure 25] 13 is a flowchart showing a method for correcting X-shake in the seventh embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0010] Hereinafter, embodiments will be described in detail with reference to the accompanying drawings. Note that the following embodiments do not limit the scope of the invention claimed. Although multiple features are described in the embodiments, not all of these multiple features are necessarily essential to the invention, and multiple features may be combined arbitrarily. Furthermore, in the accompanying drawings, the same reference numerals are used to designate the same or similar components, and redundant explanations will be omitted.

[0011] First Embodiment A first embodiment of the present invention will be described below. FIG. 1 is a side view showing a simplified functional configuration of an anti-shake control system in a camera 11 consisting of a camera body 11a and an interchangeable lens 11b that can be attached to and detached from the camera body 11a, FIG. 2 is a top view, and FIG. 3 is a front view.

[0012] A camera CPU 12 provided in the camera body 11a controls the shooting operation and vibration isolation control operation within the camera 11 in response to shooting instruction operations from the photographer. When a light beam from a subject along the optical axis 10 passes through a photographing optical system 13 provided in the interchangeable lens 11b and enters the image sensor 14, the image sensor 14 photoelectrically converts the incident light beam and outputs an image signal.

[0013] In FIG. 1, the third angular velocity meter 15pg detects the angular velocity of rotational shake applied to the camera 11 in the direction indicated by the arrow 15ps (pitch direction) and outputs an angular velocity signal (hereinafter referred to as the "pitch angular velocity signal"). The pitch angular velocity signal is input to the camera CPU 12, which performs calculations using this pitch angular velocity signal to determine a pitch shake correction target value for correcting angular shake indicated by the arrow 15p (hereinafter referred to as "pitch shake") and outputs this to the driver 14b. The driver 14b moves the image sensor 14 in the direction indicated by the arrow 14y based on the pitch shake correction target value, thereby reducing the deviation of the image plane due to pitch shake. In the first embodiment, the shake correction unit 14a is configured by the image sensor 14 and a mechanism (not shown) that movably supports the image sensor 14.

[0014] Third accelerometer 16ya detects the acceleration of translational shake applied to camera 11 in the direction indicated by arrow 16ys (Y direction) and outputs an acceleration signal (hereinafter referred to as the "Y acceleration signal"). The Y acceleration signal is input to camera CPU 12, which performs calculations using this Y acceleration signal to determine a Y shake correction target value for correcting translational shake indicated by arrow 16y (hereinafter referred to as "Y shake") and outputs this value to driver 14b. Driver 14b moves image sensor 14 in the direction indicated by arrow 14y based on the Y shake correction target value, thereby reducing the shift in the image plane caused by Y shake.

[0015] 2, second angular velocity meter 15yg detects the angular velocity of rotational shake applied to camera 11 in the direction indicated by arrow 15ys (yaw direction) and outputs an angular velocity signal (hereinafter referred to as the "yaw angular velocity signal"). The yaw angular velocity signal (rotational shake signal) is input to camera CPU 12, which performs calculations using this yaw angular velocity signal to determine a yaw shake correction target value for correcting angular shake indicated by arrow 15y (hereinafter referred to as "yaw shake") and outputs this value to driver 14b. Driver 14b moves image sensor 14 in the direction indicated by arrow 14x based on the yaw shake correction target value, thereby reducing the shift in the image plane caused by yaw shake.

[0016] Second accelerometer 16xa detects the acceleration of translational shake applied to camera 11 in the direction indicated by arrow 16xs (X direction), and outputs an acceleration signal (hereinafter referred to as the "X acceleration signal"). The X acceleration signal (translational shake signal) is input to camera CPU 12, which performs calculations using this X acceleration signal to determine an X shake correction target value for correcting translational shake indicated by arrow 16x (hereinafter referred to as "X shake") and outputs this value to driver 14b. Driver 14b moves image sensor 14 in the direction indicated by arrow 14x based on the X shake correction target value, thereby reducing the shift in the image plane caused by X shake.

[0017] 3, first angular velocity meter 15rg detects the angular velocity of rotational shake applied to camera 11 in the direction indicated by arrow 15rs (roll direction) and outputs an angular velocity signal (hereinafter referred to as the "roll angular velocity signal"). The roll angular velocity signal (rotational shake signal) is input to camera CPU 12, which performs calculations using this roll angular velocity signal to determine a roll shake correction target value for correcting rotational shake around optical axis 10 indicated by arrow 15r (hereinafter referred to as "roll shake") and outputs this value to driver 14b. Driver 14b reduces the shift in the image plane caused by roll shake by rotating image sensor 14 in the direction indicated by arrow 14r based on the roll shake correction target value.

[0018] Next, with reference to FIG. 4, the influence of gravitational acceleration acting on the second accelerometer 16xa due to roll shake will be described.

[0019] 4(a) shows the case where camera 11 is in an upright state (reference posture). In this case, acceleration detection direction 16xs (horizontal direction) of second accelerometer 16xa is perpendicular to gravity direction 51, and the X acceleration signal output from second accelerometer 16xa is not affected by gravitational acceleration.

[0020] In contrast, FIG. 4(b) shows a case where camera 11 is rotating around optical axis 10 due to roll shake. In this case, acceleration detection direction 16xs of second accelerometer 16xa and gravity direction 51 are no longer perpendicular. Here, if X shake occurs in the direction of arrow 52, ​​the X acceleration signal output from second accelerometer 16xa is a signal in which the gravitational acceleration component is added to the X shake acceleration. Note that if X shake occurs in the direction opposite to arrow 52, ​​the signal is a signal in which the gravitational acceleration component is subtracted from the X shake acceleration. Which of the above states exists is determined by the phase relationship between first gyro 15rg and second gyro 15yg. If the two are roughly out of phase, the gravitational acceleration component is added to the X acceleration signal from second accelerometer 16xa. If they are roughly in phase, the gravitational acceleration component is subtracted from the X acceleration signal.

[0021] FIG. 5 is a block diagram showing a functional configuration for removing the gravitational acceleration component superimposed on the X acceleration signal output from the second accelerometer 16xa and calculating the X shake correction target value in the first embodiment, which is realized by the camera CPU 12 executing a program.

[0022] Based on the roll angular velocity signal from first angular velocity meter 15rg and the initial attitude of the camera, gravitational acceleration fluctuation calculation unit 12a calculates the gravitational acceleration component applied to second accelerometer 16xa. First fluctuation range calculation unit 12b calculates the fluctuation range of the gravitational acceleration component calculated by gravitational acceleration fluctuation calculation unit 12a. The fluctuation range will be described later. Then, based on the fluctuation range of the gravitational acceleration component calculated by first fluctuation range calculation unit 12b, fluctuation range correction unit 12c reduces the influence of the gravitational acceleration component superimposed on the X acceleration signal output from second accelerometer 16xa. The X acceleration signal from second accelerometer 16xa, from which the influence of the gravitational acceleration component has been reduced, is output to target value calculation unit 12e, where it is converted into an X shake correction target value based on the sensitivity of the optical system of lens 11b and the imaging magnification.

[0023] FIG. 6 is a graph illustrating a method for reducing the influence of the gravitational acceleration component (error signal) superimposed on the X acceleration signal in this embodiment, where the horizontal axis represents time and the vertical axis represents acceleration. Waveform 61a shown in Figure 6(a) represents the X acceleration signal output from second accelerometer 16xa when roll shake is occurring. As explained using Figure 4(b), when the signals from first gyroscope 15rg and second gyroscope 15yg are roughly in phase, the acceleration due to X shake and the acceleration due to gravity acting on second accelerometer 16xa are in opposite directions. Therefore, the amplitude of the output X acceleration signal is smaller than when there is no effect of gravitational acceleration.

[0024] 6(b) shows the gravitational acceleration component output from the gravitational acceleration fluctuation calculation unit 12a, and is a waveform obtained by calculating the gravitational acceleration component acting on the second accelerometer 16xa based on the roll angular velocity signal output from the first angular velocity meter 15rg. For example, the gravitational acceleration component shown in the waveform 61b is obtained by calculating the angle between the acceleration detection direction 16xs of the second accelerometer 16xa and the direction of gravity 51 from the roll shake rotation angle obtained by time-integrating the roll angular velocity signal output from the first angular velocity meter 15rg and the initial attitude of the camera.

[0025] This embodiment is characterized in that the influence of gravitational acceleration is calculated using the fluctuation range of the acceleration. Because waveform 61b representing the gravitational acceleration component applied to second accelerometer 16xa is an alternating waveform, first fluctuation range calculator 12b calculates effective value B (e.g., root mean square) 62b of the gravitational acceleration component of waveform 61b over a predetermined period (e.g., 1 to 5 seconds). Similarly, because waveform 61a of the X acceleration signal output from second accelerometer 16xa is also an alternating waveform, second fluctuation range calculator 12d calculates effective value A 62a of the X acceleration signal over a predetermined period (e.g., 1 to 5 seconds) of waveform 61a.

[0026] Effective value A62a is the amount of alternating fluctuation resulting from the combination of the X-shake acceleration and the gravitational acceleration component, and effective value B62b is the amount of alternating fluctuation in the gravitational acceleration component. Therefore, fluctuation range correction unit 12c multiplies signal waveform 61a of second accelerometer 16xa by effective value B / effective value A (hereinafter referred to as "B / A"), which is the ratio of the two, to obtain X-shake acceleration waveform 61c corresponding to the output of second accelerometer 16xa when no roll shake is occurring. While this waveform is not an ideal X-shake acceleration obtained by subtracting the gravitational acceleration component from the X acceleration signal, it is a reasonable value for use in calculating a target value for correcting X-shake.

[0027] In conventional methods, if the accuracy of the angular velocity meter and accelerometer is low and the phases of waveforms 61a and 61b are not aligned, there is a risk that the gravitational acceleration fluctuations cannot be accurately removed even if they are subtracted from the accelerometer signal. In contrast, in the method of this embodiment, the X acceleration signal output from second accelerometer 16xa is multiplied by the ratio of the fluctuation range of the gravitational acceleration component to the fluctuation range of the X acceleration signal, so phase misalignment with the angular velocity meter is not a problem. Therefore, a stable X shake acceleration waveform 61c can be obtained. The signal obtained in this manner is input to target value calculation unit 12e.

[0028] In this embodiment, the fluctuation ranges of the X acceleration signal and the gravitational acceleration component are calculated using root-mean-square effective values ​​A and B, but other methods may be used. For example, the fluctuation ranges may be calculated using the maximum and minimum values ​​of waveforms 61a and 61b within a predetermined period in FIG. 6, the area of ​​waveforms 61a and 61b within a predetermined period, or the discrete Fourier transform value of a predetermined frequency. The gravitational acceleration component superimposed on the X acceleration signal output from second accelerometer 16xa can be removed by multiplying the X acceleration signal by the ratio of the fluctuation range of the gravitational acceleration component to the obtained fluctuation range of the X acceleration signal.

[0029] Target value calculation unit 12e converts the X-shake acceleration input from fluctuation range correction unit 12c into an X-shake displacement by, for example, second-order integration, and calculates an X-shake correction target value based on the sensitivity and imaging magnification of imaging optical system 13. Then, driver 14b moves image sensor 14 in the direction of arrow 14x based on the calculated X-shake correction target value, thereby reducing the deviation of the image plane due to X-shake.

[0030] FIG. 7 is a flowchart showing the method for correcting X-shake in the first embodiment, which starts when the power of the camera 11 is turned on.

[0031] In S101, the gravitational acceleration fluctuation calculation unit 12a calculates the gravitational acceleration component acting on the second accelerometer 16xa from the roll angular velocity signal output by the first angular velocity meter 15rg, and outputs the calculated component to the first fluctuation range calculation unit 12b. At the same time, the X acceleration signal output by the second accelerometer 16xa is input to the second fluctuation range calculation unit 12d.

[0032] In S102, the first fluctuation range calculation unit 12b accumulates the gravitational acceleration component, and the second fluctuation range calculation unit 12d accumulates the X acceleration signal for a predetermined time (for example, one second). In S103, the first fluctuation range calculation unit 12b and the second fluctuation range calculation unit 12d calculate the effective value B of the accumulated gravitational acceleration component and the effective value A of the X acceleration signal, respectively.

[0033] In S104, the process returns to S101 and continues calculating the effective values ​​A and B until a shooting instruction is given by the photographer. Note that when the loop from S101 to S104 is repeated, the accuracy of each effective value may be improved by calculating a moving average of the effective values ​​A and B calculated, for example, every second before exposure. Furthermore, if the time from when the camera is turned on to the start of exposure is less than one second, the accuracy of the effective values ​​A and B cannot be improved, so X-shake correction in S106, which will be described later, may not be performed.

[0034] When exposure starts in S104, the process proceeds to S105. In S105, the fluctuation range correction unit 12c multiplies the X acceleration signal from the second accelerometer 16xa by the ratio B / A of the effective value A and the effective value B calculated in S103, thereby correcting the signal to an X shake acceleration equivalent to a signal from which the gravitational acceleration component has been removed, and outputs the corrected signal.

[0035] In S106, target value calculation unit 12e converts the X-shake acceleration signal output from fluctuation range correction unit 12c into X-shake displacement, etc., and calculates an X-shake correction target value using the sensitivity of the photographing optical system and the photographing magnification. The calculated X-shake correction target value is then output to drive unit 14b, which drives image sensor 14 in the direction of arrow 14x, thereby reducing the deviation of the image plane due to X-shake.

[0036] In S107, it is determined whether exposure has ended, and the process returns to S105 to continue X-shake correction until exposure has ended, and when exposure has ended, the process returns to S101.

[0037] As described above, according to the first embodiment, even if the phases of the signals output from the angular velocity meter that detects roll shake and the accelerometer that detects translational shake in the X direction are not aligned, it is possible to quickly correct the gravitational acceleration component that is caused by roll shake and applied to the accelerometer. This makes it possible to stably mitigate the influence of the gravitational acceleration component that is superimposed on translational shake in the X direction.

[0038] Second Embodiment Next, a second embodiment of the present invention will be described. Fig. 8 is a side view of camera 11 in the second embodiment, and Fig. 9 is a top view of camera 11. The configuration shown in Figs. 8 and 9 differs from the configuration shown in Figs. 1 and 2 in that it has a blur correction unit 13a that drives lens 13c, which is part of photographing optical system 13, in the directions of arrows 13x and 13y by driver 13b, instead of blur correction unit 14a that drives image sensor 14 by driver 14b. The other configuration is the same as in Figs. 1 and 2, so the same reference numerals are used and their description will be omitted.

[0039] The front view of the camera 11 is the same as that shown in FIG. 3, and therefore will not be described here.

[0040] The method of correcting Y-shake in the camera 11 having the configuration shown in Fig. 8 is different from the correction method described in the first embodiment. The method of Y-shake correction in the second embodiment will be described below.

[0041] First, the units of the Y acceleration signal in the direction of the arrow 16ys obtained from the third accelerometer 16ya and the pitch angular velocity signal in the direction of the arrow 15ps obtained from the third angular velocity sensor 15pg are aligned to determine the ratio. This determines the radius of rotation 91y from the third accelerometer 16ya to the center of rotation 91yc of the Y shake. Next, the determined radius of rotation 91y is added to the predetermined radius of rotation 92y from the third accelerometer 16ya to the optical system principal point to determine the true radius of rotation 93y. Finally, the pitch angular velocity signal output from the third angular velocity sensor 15pg is multiplied by the true radius of rotation 93y to determine the Y shake in the direction of the arrow 16y.

[0042] The method for correcting X-shake is the same as the method for correcting Y-shake. First, the units of the X acceleration signal in the direction of arrow 16xs obtained from second accelerometer 16xa and the yaw angular velocity signal in the direction of arrow 15ys obtained from second angular velocity meter 15yg are aligned and a ratio is calculated. This determines the radius of rotation 91x from second accelerometer 16xa to the shake rotation center 91xc. Next, the calculated radius of rotation 91x is added to the preset radius of rotation 92x from second accelerometer 16xa to the optical system principal point to calculate the true radius of rotation 93x. Finally, the yaw angular velocity signal output from second angular velocity meter 15yg is multiplied by the true radius of rotation 93x to calculate the X-shake in the direction of arrow 16x.

[0043] In this way, once the true rotation radii 93x and 93y are known, the stable X-shake amount and Y-shake amount can be obtained using only the angular velocity signal from the angular velocity meter, without using the signal from the accelerometer.

[0044] Fig. 10 is a block diagram showing a functional configuration for mitigating the gravitational acceleration component superimposed on the X acceleration signal output from second accelerometer 16xa and calculating an X shake correction target value in the second embodiment, and is realized by the camera CPU 12 executing a program. Note that in Fig. 10, components having functions similar to those shown in Fig. 5 are assigned the same reference numerals. Fig. 11 is a graph illustrating a method for determining an X shake velocity in this embodiment in which the influence of gravitational acceleration due to roll shake has been mitigated.

[0045] FIG. 11(b) shows waveform 1301b of the gravitational acceleration component in the X direction calculated by gravitational acceleration fluctuation calculation unit 12a from first angular velocity meter 15rg. Based on this gravitational acceleration component, first fluctuation range calculation unit 12b calculates effective value B1302b of the gravitational acceleration component superimposed on the X acceleration signal due to roll shake. Note that the method for calculating effective value B1302b is the same as in the first embodiment. FIG. 11(a) shows waveform 1301a of the X acceleration signal output from second accelerometer 16xa. Based on this X acceleration signal, second fluctuation range calculation unit 12d calculates effective value A1302a of the X acceleration signal including the gravitational acceleration component due to roll shake. Note that the method for calculating effective value A1302a is the same as in the first embodiment.

[0046] FIG. 11(c) shows the effective value C1302c of the X-shake acceleration calculated by the effective value correction unit 12j using the effective value A1302a and the effective value B1302b when there is no influence of fluctuations in gravitational acceleration. As explained with reference to FIG. 4(b), whether the gravitational acceleration component is added to or subtracted from the X-shake acceleration depends on the phase relationship between the first and second angular velocity sensors 15rg and 15yg. The phase determination unit 12i determines whether the signals from the first and second angular velocity sensors 15rg and 15yg are generally in phase or out of phase, and outputs the result to the fluctuation range correction unit 12c. For example, the phase determination is made such that if the phase difference between the signals from the first and second angular velocity sensors 15rg and 15yg is within a predetermined range, the signals are in phase; otherwise, the signals are out of phase.

[0047] If the phase determination unit 12i determines that the phases are in phase, the effective value correction unit 12j sets the absolute value of the difference between the effective value A1302a and the effective value B1302b as the effective value C1302c, and if the phase determination unit 12i determines that the phases are out of phase, the effective value correction unit 12j sets the sum of the effective value A1302a and the effective value B1302b as the effective value C1302c.

[0048] Meanwhile, the third fluctuation range calculation unit 12f differentiates the yaw angular velocity signal output from the second angular velocity meter 15yg, converts it into a waveform 1301d indicating yaw angular acceleration as shown in Figure 11(d), and then calculates an effective value D 1302d of yaw shake. The method for calculating effective value D 1302d is the same as the method for calculating effective value A 1302a and effective value B 1302b. The yaw angular velocity signal is differentiated to obtain the yaw angular acceleration so that the units are consistent with those of the X acceleration signal.

[0049] The radius of gyration calculation unit 12g obtains the radius of gyration 91x shown in Fig. 9 by calculating the ratio between the effective value C1302c of the X-shake acceleration input from the effective value correction unit 12j and the effective value D1302d of the yaw angular acceleration input from the third fluctuation range calculation unit 12f. Next, the known radius of gyration 92x is added to the radius of gyration 91x to obtain the true radius of gyration 93x.

[0050] Multiplication unit 12h multiplies the yaw angular velocity signal output from second angular velocity meter 15yg by true radius of rotation 93x calculated by radius of rotation calculation unit 12g to calculate X-shake velocity 1301e shown in Fig. 11(e). Target value calculation unit 12e calculates an X-shake correction target value based on the signal from multiplication unit 12h and the sensitivity and imaging magnification of the optical system of lens 11b, and outputs the result to drive unit 13b.

[0051] 12 is a flowchart showing a method for correcting X-shake in the second embodiment, and starts when the power of the camera 11 is turned on. Note that the same steps as those in the flowchart of FIG. 7 are given the same step numbers, and descriptions thereof will be omitted where appropriate.

[0052] In the second embodiment, in S201, the gravitational acceleration fluctuation calculation unit 12a calculates the gravitational acceleration component acting on the second accelerometer 16xa from the roll angular velocity signal output by the first angular velocity meter 15rg, and outputs the calculated component to the first fluctuation range calculation unit 12b. At the same time, the X acceleration signal output by the second accelerometer 16xa is input to the second fluctuation range calculation unit 12d, and the roll angular velocity signal output by the first angular velocity meter 15rg and the yaw angular velocity signal output by the second angular velocity meter 15yg are input to the phase determination unit 12i.

[0053] Thereafter, the X acceleration signal and the gravitational acceleration component signal are accumulated in S102, and the effective values ​​A and B are calculated in S103.

[0054] When exposure is started in S104, the process proceeds to S202. In S202, the phase determination unit 12i determines whether the roll angular velocity signal and yaw angular velocity signal input in S201 are roughly in phase or out of phase. If they are roughly in phase, the process proceeds to S203, and if they are out of phase, the process proceeds to S204.

[0055] In S203, the effective value correction unit 12j calculates the absolute value of the difference between the effective value A and the effective value B, and sets the calculated value as the in-phase effective value C. On the other hand, in S204, the sum of the effective value A and the effective value B is calculated, and sets the calculated value as the out-of-phase effective value C.

[0056] Next, in S205, the third fluctuation range calculation unit 12f calculates the effective value D of the yaw angular acceleration from the yaw angular velocity signal output from the second angular velocity meter 15yg, and the radius of rotation calculation unit 12g calculates the radius of rotation of the yaw shake from the ratio of the effective values ​​C and D. Then, in S206, the multiplication unit 12h multiplies the yaw angular velocity signal output from the second angular velocity meter 15yg by the radius of rotation calculated in S205 to generate an X shake velocity equivalent to a signal from which the gravitational acceleration component has been removed, and outputs the signal.

[0057] In S106, target value calculation unit 12e converts the X-shake velocity signal output from multiplication unit 12h into X-shake displacement, etc., and calculates an X-shake correction target value using the sensitivity of the photographing optical system and the photographing magnification. The calculated X-shake correction target value is then output to drive unit 13b, which drives lens 13c in the direction of arrow 13x, thereby reducing the deviation of the image plane due to X-shake.

[0058] In S107, it is determined whether exposure has ended, and the process returns to S202 to continue X-shake correction until exposure has ended, and when exposure has ended, the process returns to S101.

[0059] As described above, according to the second embodiment, when correcting X-shake using a yaw angular velocity signal, it is possible to stably mitigate the influence of gravitational acceleration caused by roll shake acting on the second accelerometer 16xa.

[0060] <Third embodiment> Next, a third embodiment of the present invention will be described. Fig. 13 is a block diagram showing a functional configuration for reducing the gravitational acceleration component superimposed on the X acceleration signal output from second accelerometer 16xa and calculating an X shake correction target value in the third embodiment, and is realized by camera CPU 12 executing a program. The configuration shown in Fig. 13 is the same as the configuration shown in Fig. 10, except that first integrator 12k, second integrator 12l, and third integrator 12m are further provided. With this configuration, in the third embodiment, the effective value is calculated from the velocity.

[0061] The configuration of the camera 11 is the same as that described with reference to FIGS. 8, 9, and 3 in the second embodiment, and therefore will not be described here.

[0062] The first integrator 12k integrates the gravitational acceleration component applied to the second accelerometer 16xa output from the gravitational acceleration fluctuation calculator 12a and converts it into a gravitational velocity component. That is, it calculates the velocity error of the fluctuation of the gravitational acceleration component caused by roll shake. The first fluctuation range calculator 12b then calculates the effective value B of the fluctuation range of the gravitational velocity component, which is the velocity error.

[0063] The second integrator 12l integrates the X acceleration signal output from the second accelerometer 16xa and converts it into an X velocity signal (translational shake signal). The converted X velocity signal is a signal in which a velocity error due to fluctuations in the gravitational acceleration component is superimposed on the X shake velocity. The second fluctuation range calculator 12d then calculates the effective value A of the fluctuation range of the X shake velocity signal on which the velocity error is superimposed. In this way, by integrating the gravitational acceleration component and the X acceleration signal and then calculating the effective value, the effective values ​​A and B can be calculated with high accuracy and with less influence of noise.

[0064] The radius of gyration calculation unit 12g obtains the radius of gyration 91x shown in Fig. 9 by calculating the ratio between the effective value C output from the effective value correction unit 12j and the effective value D of the yaw angular velocity signal output from the third fluctuation range calculation unit 12f. Here, in the third embodiment, unlike the second embodiment, the third fluctuation range calculation unit 12f obtains the effective value D of the yaw angular velocity signal without differentiating the yaw angular velocity signal.

[0065] The third integrator 12m integrates the yaw angular velocity signal from the second angular velocity meter 15yg and converts it into an angle signal (rotational shake signal).Then, the multiplier 12h multiplies the angle signal input from the third integrator 12m by the radius of rotation obtained from the radius of rotation calculation unit 12g to obtain an X-shake displacement and outputs it to the target value calculation unit 12e.

[0066] In addition, the accuracy of the effective values ​​A and B can be further improved by adding a high-pass filter to the first integrating section 12k, the second integrating section 12l, and the third integrating section 12m to remove extremely low frequency noise.

[0067] Fig. 14 is a block diagram showing another configuration for calculating an X-shake correction target value in the third embodiment. The configuration shown in Fig. 14 differs from the configuration shown in Fig. 13 in that the first integrator 12k and the second integrator 12l are replaced with fourth integrator 12n and fifth integrator 12p, respectively, and that the fourth integrator 12n converts the gravitational acceleration component applied to the second accelerometer 16xa, output from the gravitational acceleration fluctuation calculator 12a, into a displacement fluctuation by performing a second integration. That is, the displacement error of the fluctuation of the gravitational acceleration component caused by roll shake is calculated. The first fluctuation range calculator 12b then calculates an effective value B of the fluctuation range of the displacement component, which is the displacement error.

[0068] Furthermore, fifth integrator 12p double-integrates the X acceleration signal from second accelerometer 16xa to convert it into a displacement signal (translational shake amount). The converted displacement is a signal in which a displacement error due to fluctuations in the gravitational acceleration component is superimposed on the X shake displacement. Second fluctuation range calculator 12d then calculates effective value A of the fluctuation range of the X shake displacement on which the displacement error is superimposed. In this way, by double-integrating the gravitational acceleration component and the X acceleration signal and then calculating the effective value in terms of displacement, effective values ​​A and B can be calculated with high accuracy and with less noise.

[0069] The third integrator 12m integrates the yaw angular velocity signal from the second angular velocity meter 15yg and converts it into an angle signal (rotational shake signal), and the third fluctuation range calculator 12f calculates the effective value D of the angle signal. The radius of gyration calculator 12g calculates the ratio between the effective value C input from the effective value corrector 12j and the effective value D input from the third fluctuation range calculator 12f to obtain the radius of gyration 91x shown in FIG. 9. The multiplier 12h then multiplies the angle signal input from the third integrator 12m by the radius of gyration to convert it into an X-shake displacement, and outputs it to the target value calculator 12e.

[0070] As described above, according to the third embodiment, the signals obtained from each angular velocity meter and accelerometer are converted into velocity signals or displacement signals before calculating the effective values, thereby making it possible to obtain effective values ​​with higher accuracy and less influence of noise.

[0071] <Fourth embodiment> Next, a fourth embodiment of the present invention will be described. Fig. 15 is a block diagram showing a functional configuration for mitigating the gravitational acceleration component superimposed on the X acceleration signal output from second accelerometer 16xa and calculating an X shake correction target value in the fourth embodiment, which is realized by executing a program by camera CPU 12. The configuration shown in Fig. 15 is a configuration in which band-pass filters 12q, 12r, and 12s that extract specific frequencies of the input signal are added to the configuration shown in Fig. 10.

[0072] The configuration of the camera 11 is the same as that described with reference to FIGS. 8, 9, and 3 in the second embodiment, and therefore will not be described here.

[0073] First band-pass filter 12q extracts a signal of a predetermined frequency (e.g., 2 Hz) from the gravitational acceleration component applied to first angular velocity meter 15rg output from gravitational acceleration variation calculation unit 12a. Similarly, second band-pass filter 12r extracts a signal of the same frequency as first band-pass filter 12q from the X acceleration signal output from second accelerometer 16xa. Furthermore, third band-pass filter 12s extracts a signal of the same frequency as first band-pass filter 12q from the yaw angular velocity signal output from second angular velocity meter 15yg.

[0074] The processing after extracting the signal of the predetermined frequency from each signal is the same as the processing described in the second embodiment with reference to FIG. 10, and therefore the description thereof will be omitted here.

[0075] In this way, by passing the signals through the first and second band-pass filters 12q and 12r, it is possible to attenuate noise superimposed on the roll angular velocity signal and the X acceleration signal. This makes it possible to stably calculate the effective values ​​A and B at a frequency (e.g., 2 Hz) at which X shake is likely to occur. Furthermore, the third band-pass filter 12s also attenuates noise superimposed on the yaw angular velocity signal, thereby enabling the third fluctuation range calculation unit 12f to stably calculate the effective value D. This enables the gyration radius calculation unit 12g to calculate the gyration radius with high accuracy.

[0076] In the configurations described in FIGS. 13 and 14, as in FIG. 15, the effective values ​​A, B, and D can be stably determined by extracting only predetermined frequencies from the signals of the first gyroscope 15rg, the second accelerometer 16xa, and the second gyroscope 15yg using a band-pass filter.

[0077] Note that each bandpass filter may not extract a single frequency (e.g., 2 Hz), but may extract signals of multiple frequencies (e.g., 0.5 Hz, 2 Hz, 5 Hz) and determine the effective values ​​A, B, and D for each frequency. In this case, highly accurate shake correction can be performed by using the average or largest effective value of the effective values ​​determined for each frequency.

[0078] As described above, according to the fourth embodiment, the effective value can be found more stably by using a band-pass filter that extracts a predetermined frequency.

[0079] Fifth Embodiment Next, a fifth embodiment of the present invention will be described. FIG. 16 is a block diagram showing a functional configuration for mitigating the gravitational acceleration component superimposed on the X acceleration signal output from the second accelerometer 16xa and calculating the X shake correction target value in the fifth embodiment, which is realized by the camera CPU 12 executing a program.

[0080] The configuration of the camera 11 is the same as that described with reference to FIGS. 8, 9, and 3 in the second embodiment, and therefore will not be described here.

[0081] The difference between FIG. 16 and FIG. 15 lies in the phase determination method used by the phase determination unit 12i. In FIG. 15 , the phase determination unit 12i determines whether the roll angular velocity signal output from the first angular velocity meter 15rg and the yaw angular velocity signal output from the second angular velocity meter 15yg are approximately in phase or out of phase. In contrast, in FIG. 16 , the phase determination unit 12i determines whether the phase of the gravitational acceleration component applied to the second accelerometer 16xa, output from the gravitational acceleration fluctuation calculation unit 12a, and the phase of the X acceleration signal output from the second accelerometer 16xa are approximately in phase or out of phase. The effective value correction unit 12j changes the method for calculating the effective value C depending on the determination result, and the reason for this will be explained using FIG. 17.

[0082] First, consider the case where a fluctuation in the gravitational acceleration component is added to the X acceleration signal 1901 from the second accelerometer 16xa when an X shake is applied with no fluctuation in the gravitational acceleration component (no roll shake). At this time, there are two directions in which the fluctuation in the gravitational acceleration component is added to the X shake acceleration. One is waveform 1902 when the fluctuation in the gravitational acceleration component and the X shake acceleration are added to the gravitational acceleration, and the other is waveform 1903 when they are subtracted.

[0083] Comparing the waveform 1904 of the gravitational acceleration component output from the gravitational acceleration variation calculation unit 12a with the above-mentioned waveforms 1902 and 1903, it is found that the waveforms 1902 and 1904 are in phase when added together, and that the waveforms 1903 and 1904 are out of phase when subtracted. Therefore, the effective value correction unit 12j determines the absolute value of the difference between the effective values ​​A and B as the effective value C when they are in phase, and determines the absolute value of the sum of the effective values ​​A and B as the effective value C when they are out of phase.

[0084] 18 is a block diagram showing the functional configuration of the phase determination unit 12i in the fifth embodiment. A specific frequency (e.g., 2 Hz) is extracted by the fourth and fifth band-pass filters 12t and 12u from the gravitational acceleration component from the gravitational acceleration variation calculation unit 12a and the X acceleration signal from the second accelerometer 16xa, respectively, and the phases of the extracted signals are compared by the phase comparison unit 12v. The phase comparison can be performed by comparing sine wave integrals and cosine wave integrals in the well-known discrete Fourier transform. The comparison result is then output to the effective value correction unit 12j.

[0085] Sixth Embodiment Next, a sixth embodiment of the present invention will be described. FIG. 19 is a front view of the camera 11 in the sixth embodiment, and since it has the same configuration as FIG. 3, the same reference numerals are used and the description thereof will be omitted. However, since the X-shake correction method is different from that in the first embodiment, the method of correcting the gravitational acceleration component, which will be described later, is also different. Therefore, FIG. 19 shows the center of rotation 91rc and the radius of rotation 93r, etc., which are necessary for explaining the method of correcting the gravitational acceleration component in the sixth embodiment. Note that the side view and top view of the camera 11 in the sixth embodiment are the same as those shown in FIGS. 8 and 9, and therefore the description thereof will be omitted.

[0086] Fig. 20 is a block diagram showing the functional configuration for removing the gravitational acceleration component superimposed on the X acceleration signal output from the second accelerometer 16xa and calculating the X shake correction target value in the sixth embodiment, which is realized by the camera CPU 12 executing a program. The differences between the block diagrams shown up to Fig. 16 and the block diagram shown in Fig. 20 are as follows.

[0087] One is that the radius of gyration calculation unit 12g calculates the radius of gyration 93r in Fig. 19 from the ratio between the effective value D of the roll angular velocity signal output from the first angular velocity meter 15rg and the effective value C calculated by the effective value correction unit 12j. Then, the multiplication unit 12h calculates the product of the radius of gyration calculated by the radius of gyration calculation unit 12g and the roll angular velocity signal output from the first angular velocity meter 15rg, thereby converting the roll angular velocity signal into an X-shake velocity in the direction of the arrow 16x in Fig. 1, and inputs the converted value to the target value calculation unit 12e.

[0088] Note that X-shake may occur due to yaw shake, as explained in the second to fifth embodiments, or due to roll shake, as in the sixth embodiment. Therefore, image degradation due to X-shake can be alleviated by using either method, or by adding up the X-shakes of both types.

[0089] The method of obtaining X-shake using roll shake is not limited to the above method, and can also be applied to the functional configurations shown in the block diagrams from Fig. 10 onwards which have been explained so far. For example, the functional configuration shown in Fig. 16 and the functional configuration shown in Fig. 14 may be modified to obtain the functional configuration shown in Fig. 21. In Fig. 21, X-shake is obtained by the product of the angle obtained by integrating the roll angular velocity signal from the first angular velocity meter 15rg and the radius of rotation, and accuracy is improved by suppressing noise through integration and bandpass filtering in error calculation and fluctuation range correction due to gravitational acceleration.

[0090] Seventh Embodiment Next, a seventh embodiment of the present invention will be described. Note that the configuration of the camera 11 in the seventh embodiment is the same as that described with reference to Figures 1 to 3 or 9, 10, and 19, and therefore description thereof will be omitted here.

[0091] FIG. 22 is a block diagram showing a functional configuration for removing the gravitational acceleration component superimposed on the X acceleration signal output from the second accelerometer 16xa and calculating the X shake correction target value in the seventh embodiment, which is realized by the camera CPU 12 executing a program.

[0092] The configuration shown in Fig. 22 differs from the configuration shown in Fig. 5 in that it is a simpler configuration that does not use the X acceleration signal from second accelerometer 16xa to correct the fluctuation range of the gravitational acceleration component. In Fig. 22, the gravitational acceleration component is calculated from the roll angular velocity signal from first angular velocity meter 15rg, and the fluctuation range is used, just like in Fig. 5. Then, fluctuation range corrector 12c changes the gain applied to the X acceleration signal from second accelerometer 16xa according to the fluctuation range. In the seventh embodiment, since there is no need to calculate effective value A of the X acceleration signal, the influence of the gravitational acceleration component can be mitigated earlier.

[0093] Furthermore, to further simplify the configuration, as shown in FIG. 23, the fluctuation range correction unit 12c may adjust the gain applied to the X acceleration signal from the second accelerometer 16xa in accordance with the magnitude of the roll angular velocity signal from the first angular velocity meter 15rg.

[0094] This may also be applied to the method of calculating the radius of gyration, as explained in the second to sixth embodiments. For example, as shown in Fig. 24, when the roll angular velocity signal output from the first angular velocity meter 15rg increases, the fluctuation range corrector 12c reduces the effective value A output from the second fluctuation range calculator 12d. Then, the radius of gyration calculator 12g calculates the radius of gyration 93r based on the effective value D of the roll shake angle obtained from the third fluctuation range calculator 12f and the reduced effective value A.

[0095] 25 is a flowchart showing an X-shake correction method in the seventh embodiment, which starts when the power of the camera 11 is turned on. Note that the same steps as those in the flowchart of FIG. 7 are assigned the same step numbers.

[0096] In S101, the gravitational acceleration fluctuation calculation unit 12a calculates the gravitational acceleration component acting on the second accelerometer 16xa from the roll angular velocity signal output by the first angular velocity meter 15rg, and outputs the calculated gravitational acceleration component to the first fluctuation range calculation unit 12b. At the same time, the X acceleration signal output by the second accelerometer 16xa is input to the second fluctuation range calculation unit 12d.

[0097] Next, in S701, it is determined whether gravitational acceleration is being applied to the second accelerometer 16xa. Here, as described with reference to FIG. 4, when the camera is oriented such that the direction of detection by the second accelerometer 16xa is the same as the direction of gravity (for example, when the camera 11 is held vertically), fluctuations in the gravitational acceleration component due to roll shake are small, and therefore the gain adjustment described above is not performed. On the other hand, when the camera is oriented such that the direction of detection by the second accelerometer 16xa is different from the direction of gravity (for example, when the camera 11 is held horizontally), fluctuations in the gravitational acceleration component due to roll shake are large, and therefore gain adjustment is performed. Therefore, if it is determined in S701 that fluctuations in the gravitational acceleration component due to roll shake for the second accelerometer 16xa are small, the process proceeds to S104, skipping S702 to S704 (described later). On the other hand, if fluctuations in the gravitational acceleration component for the second accelerometer 16xa are large, the process proceeds to S702.

[0098] In addition to holding the camera 11 vertically, even if the camera 11 is pointed upward to photograph the sky or pointed downward to photograph the ground, fluctuations in the gravitational acceleration component due to roll shake are small, so the process proceeds to S104.

[0099] In S702, the magnitude of the roll angular velocity signal output from the first angular velocity meter 15rg is detected, and if it indicates that roll shake is large, the process proceeds to S703 to issue a warning to the photographer and then returns to S702, so that the flow does not proceed to the exposure operation.

[0100] On the other hand, if the roll shake is within the allowable range, the process proceeds to S704, where the gain applied to the X acceleration signal is changed according to the magnitude of the tilt obtained by integrating the roll angular velocity signal output from the first angular velocity meter 15rg. For example, if the tilt of the camera 11 due to roll shake exceeds 0.3 degrees, the gain is reduced to one-fourth, and if it exceeds 0.6 degrees, the gain is reduced to one-half.

[0101] In S104, the process returns to S101 until the photographer performs an exposure operation, and when exposure starts in S104, the process proceeds to S106.

[0102] In S106, target value calculation unit 12e calculates an X-shake correction target value based on the corrected X-acceleration signal output from fluctuation range correction unit 12c. The X-shake correction target value is then output to drive unit 13b or 14b, which drives lens 13c in the direction of arrow 13x or image sensor 14 in the direction of arrow 14x to reduce the shift of the image plane due to X-shake.

[0103] In S107, the process returns to S106 to continue X-shake correction until exposure is complete, and then returns to S101 upon completion of exposure.

[0104] In the first embodiment, the present invention has been described assuming that shake correction is performed by driving the image sensor 14, and in the second and subsequent embodiments, shake correction is performed by driving the lens 13c, but the present invention is not limited to either one, and shake correction may be performed by cooperative control of both. For example, pitch shake and yaw shake may be corrected by the lens 13c, and X shake, Y shake, and roll shake may be corrected by the image sensor.

[0105] Furthermore, in the above-described embodiment, the case where the camera is held in the normal position (a position where the up-down direction on the paper surface of FIG. 3 coincides with the direction of gravity) has been described. However, the direction of rotational shake that causes gravitational acceleration in the direction detected by the accelerometer differs depending on the camera's attitude (its orientation relative to the direction of gravity). For example, if the lens is held facing directly downward or directly upward so that the optical axis coincides with the direction of gravity, rotational shake (yaw shake) about the Y axis will cause gravitational acceleration to be superimposed on the X acceleration signal. Therefore, the detection result of yaw shake may be used instead of the detection result of roll shake in the above-described embodiment.

[0106] Specifically, in the first to seventh embodiments, the angular velocity signal input to the gravitational acceleration fluctuation calculation unit 12a may be replaced with a yaw angular velocity signal from the second angular velocity meter 15yg, and gravitational acceleration fluctuation may be calculated based on the yaw angular velocity signal. If only the angular velocity signal input to the gravitational acceleration fluctuation calculation unit 12a is replaced from the roll angular velocity signal to the yaw angular velocity signal, in the second to fifth embodiments, both the effective value B and the effective value D will be values ​​based on the second angular velocity meter 15yg, but this is not a problem. Alternatively, by replacing the first angular velocity meter 15rg with the second angular velocity meter 15yg and then replacing the second angular velocity meter 15yg with the first angular velocity meter 15rg, the X-shake amount due to roll shake may be obtained, as in the sixth embodiment. Alternatively, the X-shake amount due to roll shake and the X-shake amount due to yaw shake may be added together.

[0107] It is more preferable that the angular velocity signal input to the gravitational acceleration fluctuation calculation unit 12a be selectable based on the relationship between the direction of gravity and the Y-axis and Z-axis. When the direction of gravity and the Y-axis coincide, a roll angular velocity signal is input to the gravitational acceleration fluctuation calculation unit 12a, and the gravitational acceleration component applied to the second accelerometer 16xa is calculated based on the roll angular velocity signal. On the other hand, when the direction of gravity and the Z-axis coincide, a yaw angular velocity signal is input to the gravitational acceleration fluctuation calculation unit 12a, and the gravitational acceleration component applied to the second accelerometer 16xa is calculated based on the yaw angular velocity signal. When the direction of gravity intersects with both the Y-axis direction and the Z-axis direction, it is preferable to calculate the gravitational acceleration component using the angular velocity signal of a rotational motion about the axis whose angle with the direction of gravity is closest to 90 degrees. For example, when the imaging device is held in an attitude tilted 10 degrees from the normal position, the gravitational acceleration component applied to the second accelerometer 16xa is calculated based on the roll angular velocity signal. Furthermore, although the amount of calculation increases slightly, when the direction of gravity intersects with both the Y-axis direction and the Z-axis direction, the gravitational acceleration component due to roll shake and the gravitational acceleration component due to yaw shake can be calculated and added together to calculate the gravitational acceleration component.

[0108] Furthermore, in the above-described embodiment, the explanation has been given for vibration reduction control in an imaging device, but the device to which the present invention can be applied is not limited to an imaging device, and the present invention can be applied to various devices.

[0109] <Other embodiments> The present invention may be applied to a system made up of a plurality of devices, or to an apparatus made up of a single device.

[0110] The present invention can also be realized by supplying a program that realizes one or more functions of the above-described embodiments to a system or device via a network or a storage medium, and having one or more processors in the computer of the system or device read and execute the program.The present invention can also be realized by a circuit (e.g., ASIC) that realizes one or more functions.

[0111] The invention is not limited to the above-described embodiments, and various changes and modifications can be made without departing from the spirit and scope of the invention. Accordingly, the following claims are appended to apprise the public of the scope of the invention. [Explanation of symbols]

[0112] 10: optical axis, 11: camera, 11a: camera body, 11b: interchangeable lens, 12: camera CPU, 13: photographing optical system, 14: image sensor, 14a: shake correction unit, 14b: drive unit, 15rg: first angular velocity meter, 15yg: second angular velocity meter, 15pg: third angular velocity meter, 16xa: second accelerometer, 16ya: third accelerometer, 12a: gravitational acceleration fluctuation calculation unit, 12b: first fluctuation range calculation unit, 12c: fluctuation range correction unit, 12d: second fluctuation range calculation unit, 12e: target value calculation unit output unit, 12f: third fluctuation width calculation unit, 12g: rotation radius calculation unit, 12h: multiplication unit, 12i: phase determination unit, 12j: effective value correction unit, 12k: first integration unit, 12l: second integration unit, 12m: third integration unit, 12n: fourth integration unit, 12p: fifth integration unit, 12q: first band pass filter, 12r: second band pass filter, 12s: third band pass filter, 12t: fourth band pass filter, 12u: fifth band pass filter, 12v: phase comparison unit

Claims

1. a first input means for inputting a translational shake signal indicative of a translational shake in a first direction; a second input means for inputting a rotational shake signal indicating a direction of gravity and a rotational shake about a first axis intersecting the first direction; a shake amount acquisition means for adjusting and correcting a gain applied to the translational shake signal based on the rotational shake signal, thereby obtaining the amount of shake in the first direction; a target value obtaining means for obtaining a target value for correcting the shake in the first direction based on the amount of shake obtained by the amount of shake obtaining means; An anti-vibration control device comprising:

2. 2. The image stabilization control device according to claim 1, wherein the shake amount acquisition means adjusts the gain based on the magnitude of the rotational shake signal.

3. 3. The vibration damping control device according to claim 2, wherein the shake amount acquisition means adjusts the gain so that when the rotational shake signal has a first value, the gain is smaller than when the rotational shake signal has a second value smaller than the first value.

4. a fluctuation range acquisition means for determining a fluctuation range of a component of the gravity applied to the rotational shake in the first direction based on the rotational shake signal, 2. The image stabilization control device according to claim 1, wherein the shake amount acquisition means adjusts the gain based on the rotational shake signal by adjusting the gain based on the fluctuation range.

5. the second input means is capable of inputting signals indicating a direction of gravity and rotational shake around a plurality of axes intersecting the first direction, The shake amount acquisition means selects a signal to be used as the rotational shake signal from the signals indicating the rotational shake around the plurality of axes.

5. The vibration isolation control device according to claim 1, wherein the vibration isolation control device is a vibration isolation control device.

6. a first input step of inputting a translational shake signal indicative of translational shake in a first direction; a second input step of inputting a rotational shake signal indicating a direction of gravity and a rotational shake about a first axis intersecting the first direction; a first calculation step of adjusting and correcting a gain applied to the translational shake signal based on the rotational shake signal to obtain the amount of shake in the first direction; a second calculation step of calculating a target value for correcting the shake in the first direction based on the amount of shake calculated in the first calculation step; 10. A vibration isolation control method comprising:

7. A program for causing a computer to function as each of the means of the vibration isolation control device according to any one of claims 1 to 5.

8. A computer-readable storage medium storing the program according to claim 7.

Images