JPH05204013A - Jiggle prevention device for photographing device - Google Patents

Abstract

PURPOSE:To enable a prediction time to be variably set to enhance a vibration proofing effect by accurately predicting jiggle in the midst of exposure. CONSTITUTION:When jiggling is detected by a jiggle signal detection part 1, at least one or more outputs among the outputs from the detection part 1 are stored in a jiggle signal storage part 2. Besides, the length of an exposure time is automatically or manually set by an exposure time set part 3. Based on the output of the set part 3, at least two or more kinds of predictive coefficients for predicting the jiggle signal of a photographing time are set by a jiggle predictive coefficient set part 4. Then, a predictive jiggle signal is calculated by multiplying and adding the output of the detection part 1 and/or the output of the storage part 2 and the output from the set part 4.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image stabilizing device for a photographing device, and particularly to image quality deterioration caused by relative movement of the photographer and the subject in an optical device for recording an image of the subject such as a camera or a video camera. BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a camera shake prevention device for an image pickup apparatus that corrects and prevents the effects of vibration, so-called camera shake vibration.

[0002]

2. Description of the Related Art Conventionally, there have been devised devices for preventing deterioration of photographed images due to camera shake. Among them, there is a method of detecting a camera shake, prohibiting the exposure when the value is large, and permitting the exposure when the camera shake is small to reduce the camera shake during the exposure. In this case, in order to increase the effect, it is necessary to know the size of the blurring during the exposure before the exposure, and the size of the camera shake that may occur during the exposure before the start of the exposure is predicted. .. If a low-pass filter is used to remove the high frequency noise of the blur sensor, the delay time must be predicted because the signal is delayed.

Further, in the case of a camera shake prevention device devised so as to eliminate the image blur on the image pickup system by actively moving the optical system or the like based on the camera shake signal, in order to drive the optical system. The actuator response causes a delay. This delay also needs to be predicted.

Further, in a camera shake correction mechanism which positively drives an optical system or an image pickup system to correct camera shake, exposure is performed in order to perform correction without being affected by rattling of a mechanical system. There is one that drives the correction member linearly or curvedly in only one direction during the time.
In this case, it is necessary to predict the camera shake that may occur during exposure before exposure and determine the driving direction and speed.

Here, the time to the future desired to be obtained by this prediction is classified into two types, that is, the exposure time and the signal processing and the response of the actuator as described above. When making a prediction, the accuracy of the prediction is improved by independently defining these terms and adding them to perform the prediction. That is, there is also known a method for preventing an image blur that is advantageous by treating the predicted time separately from the fixed value and the variable value and making the predicted time variable.

On the other hand, as a method for predicting the camera shake signal, conventionally, there is a method of approximating the camera shake vibration to a simple vibration and detecting its cycle or frequency to predict the camera shake, or predict the timing at which the camera shake is eliminated. It is considered.

A method has also been proposed in which time-series data of camera shake vibration is approximated to a straight line or a higher-order regression line of quadratic or higher using a least square method or the like and extrapolated. Since this method uses a statistical method, some accuracy is expected.

[0008]

However, the vibration of camera shake is not a clean simple vibration, and vibrations of 10 Hz or less are complicatedly superposed in the dimension of the position of the image, so that the prediction by the simple vibration is inaccurate. .. Further, for real-time detection of synchronization and frequency, complex calculation such as Fourier transform is required.

Further, in the conventional method using simple vibration or regression equation, the prediction operation is performed by replacing the camera shake with the function of time, so that the prediction time is easily variable, but the coefficient and constant of the function are real-time. To repeat the detection setting,
A large amount of calculation is required. As a result, it becomes difficult to predict real-time blurring, and it is necessary to install a high-performance computer for predictive calculation, leading to large size and high cost, and application to small popular products such as cameras. Will be difficult.

The present invention has been made in view of the above problems, and in order to realize more effective camera shake prevention, vibration of camera shake vibration in an optical device for recording an image of a subject such as a camera or a video camera. In the image stabilization device that prevents and corrects the effects, it provides a highly accurate image stabilization signal predictive calculation with a simple configuration and simple calculation, and provides an image stabilization device for the imaging device that allows variable setting of the prediction time. The purpose is to do.

The present invention further relates to a camera for photographing at a timing without camera shake, or a camera provided with a camera shake preventing device for correcting the movement of the object image on the image pickup system by moving the optical axis in one direction during exposure. In this regard, it is an object of the present invention to provide a camera shake prevention device for a photographing device that can further enhance the camera shake prevention effect.

[0012]

That is, according to the present invention, as shown in FIG. 1A, a shake signal for detecting shake vibration is detected in a shake preventing device in a photographing device for recording an image of a subject. A detection unit 1, a camera shake signal storage unit 2 that stores at least one output from the camera shake signal detection unit 1, an exposure time setting unit 3 that automatically or manually sets the length of the exposure time, and a shooting time Camera shake prediction coefficient setting unit 4 for setting at least two kinds of prediction coefficients for predicting the camera shake signal of the camera based on the output of the exposure time setting unit 3, and the output of the camera shake signal detection unit 1 and / or the camera shake The present invention is characterized by including an arithmetic unit 5 that multiplies and adds an output of the signal storage unit 2 and an output from the camera shake prediction coefficient setting unit 4 to calculate a predicted camera shake signal.

Further, as shown in FIG. 1 (b), the present invention, in addition to the apparatus of FIG. 1 (a), corrects the camera shake based on the output of the arithmetic unit 5 and detects the camera shake signal detecting means. Camera shake signal delay information storage unit 6 that stores at least one of the delay time of the image stabilization operation and the delay time of the camera shake correction operation, and an exposure permission control unit 7 that controls permission and prohibition of exposure start based on the predicted camera shake signal. The camera shake prediction count setting unit 4 sets the prediction coefficient based on the output signal of the exposure time setting unit 3 and the output signal of the camera shake signal delay information storage unit 6. To do.

Further, according to the present invention, as shown in FIG. 1 (c), the camera shake signal average calculator 8 for calculating the average of the camera shake signals based on the output of the exposure time setting unit 3 and the calculator 5 are provided. And a correction unit 9 for correcting camera shake based on the output, and by storing the output of the camera shake signal average calculation unit 8 in the camera shake signal storage unit 2, the driving direction and speed of the camera shake prevention unit before the start of exposure. And at least one of them is determined.

[0015]

As shown in FIG. 1 (a), in the image stabilizing apparatus for an image capturing apparatus according to the present invention, in the image stabilizing apparatus for an image capturing apparatus for recording an image of a subject, an image stabilization signal detecting section is provided. When the camera shake signal detection unit 1 detects the camera shake vibration, at least one output from the camera shake signal detection unit 1 is stored in the camera shake signal storage unit 2.
More than one is memorized. Further, the length of the exposure time is automatically or manually set by the exposure time setting unit 3, and based on the output of the exposure time setting unit 3, at least two or more kinds for predicting the camera shake signal at the time of photographing are set. The prediction coefficient is set by the camera shake prediction coefficient setting unit 4. Then, the calculation unit 5 multiplies and adds the output of the camera shake signal detection unit 1 and / or the output of the camera shake signal storage unit 2 and the output from the camera shake prediction coefficient setting unit 4 to calculate the predicted camera shake signal. To be done.

A second configuration example of the present invention is shown in FIG.
As shown in (b), information on the delay time of the signal of the camera shake signal detection unit 1 and the delay time of the operation of the camera shake correction means
The camera shake signal delay information storage unit 6 stores the output signal of the exposure time setting unit 3 and the camera shake signal delay information storage unit 6.
The prediction count is set based on the output signal of

Furthermore, in the third configuration example of the present invention, an exposure permission control unit 7 is further provided, and the exposure permission control unit 7 controls permission and prohibition of exposure start based on the camera shake signal. )

A fourth configuration example of the present invention is shown in FIG.
As shown in (c), the signal of the camera shake signal detection unit 1 is stored in the camera shake signal storage unit 2 which takes a time average based on the output of the exposure time setting unit 3. Based on this value
The driving direction and driving speed of the camera shake correction unit 9 that moves the optical axis of the subject image of the optical system in the camera shake correction direction are determined before the start of exposure.

[0019]

Embodiments of the present invention will be described below with reference to the drawings. Before going into the detailed description of the embodiment, first, the principle of the hand movement prediction in the present invention will be explained.

As shown in FIG. 3, for the camera,
Set three axes of x, y, and z. Although camera shake occurs in the horizontal and vertical directions of the screen, an example in which correction is performed for camera shake in the horizontal direction of the screen will be described in order to simplify the description. The same can be applied to the vertical direction of the screen.

In the case of controlling the prohibition and permission of exposure, it is expected that the blurring during the exposure becomes small when the blurring becomes sufficiently small in half the exposure time. Therefore, in this case, the time to the future that we want to predict is
Just before the start of exposure, the time becomes half the exposure time. Furthermore, when the amount of delay of the angular velocity sensor and the amount of mechanical delay for camera shake correction have values that require prediction, this amount of delay is set to half the exposure time. It is sufficient to add and set the time to the future to be predicted.

Further, camera shake vibration generated during exposure is approximated to a straight line unidirectionally so as not to be affected by rattling of the correction mechanism, and the optical axis of the photographing system is positively moved to prevent camera shake. When correcting, this straight line was found to be almost equal to the average value of the angular velocity of camera shake that occurs during exposure, by conducting a camera shake prevention effect simulation. Therefore, by calculating a time average corresponding to the exposure time and predicting half the exposure time immediately before the start of exposure, the driving direction and speed for correction to be controlled during exposure can be obtained. Be done. Also in this case, when the amount of delay of the angular velocity sensor and the amount of mechanical delay for camera shake correction have values that require prediction, the time until the future including this delay amount is predicted. Should be set.

Now, it is assumed that the camera shake signal is X (i) and the predicted camera shake signal resulting from the prediction is (Y). Here, the value of (i) takes at least values of 0 and 1, and is used up to a natural number (N) of 2 or more depending on the required accuracy of the data to be predicted. The shake signal at the present time is X (0), the shake signal at the past one time point is X (1), and the signal at the past two time points is X (1).
(2), X (i) if there is a camera shake signal at time i past
Represents.

The X (0) signal is output from the camera shake signal detecting section 1. Also, past camera shake signal X
(I) is sequentially stored in the camera shake signal storage unit 2,
Old and unnecessary data is discarded and updated with a new camera shake signal. In addition, the camera shake prediction coefficient setting unit 3
As the coefficients used for prediction, the above X (0), X
(1), ..., A (0), (1), corresponding to X (i)
..., A (i) is set. How to determine the prediction coefficient will be described later. The calculation unit 5 multiplies the camera shake signal X (0) and the prediction coefficient A (0) to obtain the intermediate data W
(0) is obtained by the relational expression of Equation 1.

[0025]

[Equation 1] Further, the arithmetic unit 5 similarly similarly outputs the intermediate data W (i).
Is calculated by the relational expression of Equation 2.

[0026]

[Equation 2]

However, (i) is a natural number from 1 to (N). Also, in the camera shake signal prediction unit (not shown),
From the intermediate data (W (i)) obtained above, the predicted camera shake signal (Y) is obtained from the relational expression of Equation 3.

[0028]

[Equation 3] In other words, it becomes like the relational expression of Expression 4.

[0029]

[Equation 4] Next, how to determine the prediction coefficient used above will be described.

Now, consider the difference between the detected hand-shake signals. The difference on the first floor is Δ1, the difference on the second floor is Δ2, and the difference on the third floor is Δ3. Then, the relational expressions of the equations 5, 6, and 7 are expressed, and thus the relational equations of the equations 8, 9, and 10 are obtained.

[0031]

[Equation 5]

[0032]

[Equation 6]

[0033]

[Equation 7]

[0034]

[Equation 8]

[0035]

[Equation 9]

[0036]

[Equation 10]

The difference between the predicted camera shake signal (Y) and the latest camera shake signal is Δ1 (Y). Then,
From equations (11), (12) and (13), equation (14) and equation (1)
The relational expressions 5 and 16 are obtained.

[0038]

[Equation 11]

[0039]

[Equation 12]

[0040]

[Equation 13]

[0041]

[Equation 14]

[0042]

[Equation 15]

[0043]

[Equation 16] Further, the time interval for detecting the camera shake signal (X) is Δt, and the time to be predicted is (S). Moreover, the ratio is set to (k). Then, the relational expression of Expression 17 holds.

[0044]

[Equation 17]

Regarding the temporal transition of the camera shake signal, it is assumed that the transition is smooth to some extent.
Assume the following derivative is a constant. Now, since a discrete camera shake signal is handled, in order for the camera shake signal to make a smooth transition, the third order difference should be proportional to the data interval time. In other words, the relational expression of Eq.

[0046]

[Equation 18] From the relational expression of the equation 18 and the relational expressions of the equations 16, 15, and 14, the relational expressions of the equations 19, 20 and 21 are obtained.

[0047]

[Formula 19]

[0048]

[Equation 20]

[0049]

[Equation 21] Then, from the relational expression of the equation 21 and the relational expressions of the equations 5, 8, and 10, the relational expression of the equation 22 is obtained, and thus the relational expression of the equation 23 is established.

[0050]

[Equation 22]

[0051]

[Equation 23]

As expressed by the relational expression (23),
By multiplying the past camera shake signal and the current camera shake signal by a predetermined coefficient, the future camera shake signal can be predicted.
This coefficient is the prediction coefficient. In this example, the ratio of the predicted time and the time interval of the data regarding the difference of the third floor is considered to be proportional. However, in order to improve the accuracy of the prediction of the camera shake signal, a higher order difference is further considered. Of course, it is possible.

Further, as can be seen from the relational expression of the equation (23), the coefficient stored in the camera shake signal prediction means for the prediction calculation differs depending on the time to the future to be predicted. Because the time changes in the prediction because it depends on the exposure time,
The camera shake prediction coefficient setting unit 4 sets the coefficient based on the output of the exposure time setting unit 3 and in some cases, including the signal from the camera shake signal delay information storage unit 6.

In this case, as shown in the relational expression of the above equation 23, the prediction coefficient may be obtained by performing an operation according to a predetermined expression, or a table of prediction coefficients corresponding to the prediction time may be stored. It is also possible to set the coefficient by referring to the table according to the prediction time. Further, in the case of the method of obtaining the prediction coefficient described below, this table reference method is appropriate.

By the way, there are cases where it is possible to judge that the accuracy of the prediction is insufficient with the coefficient for the prediction set by the method shown by the relational expression of the above equation 23. This is probably because the unique vibration due to camera shake and the response of the actuator cannot be described in the form of a simple linear function.

In order to reduce this error, the camera shake signal is measured at the designing and manufacturing stages of the camera shake preventing device,
It is also possible to set this coefficient so that the difference between the predicted camera shake signal and the actual camera shake signal is minimized. Specifically, by performing multiple regression analysis using the current and past camera shake signals, the weighting coefficient of each camera shake signal can be obtained by the least squares method.

Next, an example of optimizing the prediction coefficient when predicting the detection delay time of the camera shake detection sensor will be described. The number of data and the number of measurements can be increased as necessary.

Considering the camera shake signal obtained in the experiment, it is now assumed that 200 obtained in 2 seconds at a sampling interval of 1 msec.
One shake signal is B (j) (j is 0 to 2000). In addition, this camera shake detection sensor outputs 30
It is assumed that it has a delay of msec.

Now, for the data at the same time point, the camera shake signal sequence C0 (j) represented by the subscript (j) having the same value,
Consider C1 (j), C2 (j), C3 (j).
Here, C0 (j) is the latest camera shake signal at the time of prediction, and C1 (j), C2 (j), and C3 (j) are each 10 msec past camera shake signal, 20 mec past data, and 30 msec past. The data of
In addition, the actual amount of camera shake after the time to be predicted is D (j). When the above-mentioned actually measured camera shake signal is applied to this signal train, it becomes as shown in Table 1.

[0060]

[Table 1] The data set used for prediction and the reference data set is 1
There will be 941 pieces.

Here, the prediction coefficients are A0, A1, A2, A
Set to 3. These are 0 msec and 10 mse, respectively.
c is a coefficient for multiplying past data of 20 msec and 30 msec. An arithmetic expression for prediction is expressed as a relational expression of Expression 24, where Y (j) is a predicted value at time (j).

[0062]

[Equation 24] Error e between the actual camera shake signal and the predicted value at time (j)
(J) is expressed by the relational expression of Expression 25.

[0063]

[Equation 25] Then, the sum of squares (E) of the errors is expressed as the relational expression of Expression 26.

[0064]

[Equation 26]

Next, based on the least squares method, a combination of prediction coefficients that minimizes the sum of squares (E) of the errors is obtained. Therefore, consider the matrix (G) shown by the relational expressions of the equations 27 and 28. (J) is 0 to 194 here
It is a value of 1.

[0066]

[Equation 27]

[0067]

[Equation 28] Then, the inverse matrix (G) ⁻¹ of the matrix (G) is obtained from the relational expression of Expression 29.

[0068]

[Equation 29]

By performing such an operation and determining the prediction coefficient, the prediction coefficient can be optimized according to the actual occurrence of camera shake. This coefficient may be stored in the storage means in advance. Here, an example in which the delay of the output of the camera shake detection sensor is corrected has been shown, but if the delay of the actuator and the like are taken into consideration, a more practical prediction is performed.

Further, in the above-mentioned example, an example of predicting the value of the same camera-shake detecting sensor from the value of the camera-shake detecting sensor is shown, but optimization of these prediction coefficients is performed at the design stage or manufacturing stage of the camera. Considering that the image is pre-stored in the camera shake prediction coefficient storage means of the camera before being operated by the photographer, the camera shake signal from a more accurate camera shake detection device is used as a reference in the optimization step of the prediction coefficient. By optimizing the prediction coefficient for prediction by the camera shake detection sensor in the camera, it is possible to reduce the error of the camera shake detection sensor mounted in the camera. Next, the time interval of camera shake data used for prediction will be described.

In the above example, the interval time (Δt) of this data is 10 msec. But,
This time is not limited as long as it allows the sample to sufficiently trace the trajectory of the camera shake signal. In order to sufficiently represent the camera shake, a value that can be sufficiently detected with respect to the high frequency of the camera shake signal may be used. From the conventional analysis of the phenomenon of camera shake, the frequency of camera shake that causes image deterioration has been determined to be about 10 Hz at the maximum, so it is sufficient if the interval can accurately detect 10 Hz. According to the sample theorem, this may be detected at a frequency of 8 times or more. That is, it is considered that the time interval of the prediction data may be higher than 12 msec. When the prediction time is long, the interval is set long, and when the prediction time is short, the interval is set short to improve the prediction accuracy.

Further, when the data used for prediction includes noise, the noise can be reduced by a software method such as a low-pass filter or averaging. In this case, it is possible to prevent the signal delay from being corrected by also predicting the signal delay due to the low-pass filter. Further, in this case, this is achieved by defining the data stored in the camera shake signal delay information storage unit 6 including this value. Next, a first embodiment corresponding to the first and second configuration examples of the present invention will be described.

In the first embodiment, in the camera, a change in subject brightness is measured as a camera shake signal, and when the brightness change is large, it is determined that the camera shake is large, and the exposure of the image pickup system is prohibited to change the brightness. When the image blur is small, it is determined that the camera shake is small, and the exposure is permitted to obtain an image of a subject without deterioration of the image quality due to the camera shake.

The change in subject brightness should occur in proportion to the magnitude of the blur on the plane of the screen. That is, this signal is not limited to the horizontal and vertical directions of the screen, and provides information about blurring.

FIG. 2 is a block diagram showing the basic structure of the image stabilizing apparatus of the image pickup apparatus according to the first embodiment of the present invention. In FIG. 1, the camera shake signal detection unit 1 includes a subject brightness detection unit 10 and a subject brightness differential calculation unit 11 which will be described later.
And an A / D converter 12. The output of the camera shake signal detection unit 1 is sent to the camera shake signal storage unit 2 and the multiplication calculation unit 5a of the calculation unit 5.

The exposure time setting unit 3 is means for setting the exposure time (shutter speed) for exposing the film, and is manually operated by the photographer in the manual exposure mode and the shutter speed priority mode according to the exposure mode of the camera. The exposure time is set from the value instructed by, or from the information on the brightness of the subject in the aperture priority mode and the program priority mode, or from the information on the subject brightness information including the operation amount of the photographer's aperture and program shift. It is a thing.

Hand shake signal prediction calculation unit 5 in the calculation unit 5
In b, the sum of the outputs of the multiplication calculator 5a is calculated, and the predicted camera shake signal is output. An exposure permission control unit 7 that controls permission and prohibition of exposure start based on the predicted camera shake signal is provided. The exposure permission control unit 7 includes
An exposure start signal generator 15 is connected to the exposure unit 14 in response to a signal from the exposure command output unit 13 which outputs a signal to start exposure in response to the operation of the photographer.

FIG. 4 is a diagram showing a detailed structure of the hand-movement-blurring signal detecting section 1. In the figure, the subject brightness detection unit 1
0 is composed of an SPD 10a, diodes 10b and 10c, and a buffer amplifier 10d which are connected as shown. In the first embodiment, the subject light is received,
A relatively low cost SPD 10a is used as a light receiving means for detecting the brightness of the subject. SPD10a
Is a light receiving element for partial photometry used for camera photometry. The diodes 10b and 10c are the SPD 10
It is for logarithmically compressing the photocurrent of a, and the number thereof depends on the power supply voltage. The buffer amplifier 1 is connected to the anode of the SPD 10a and the diode 10b.
0d is connected and outputs a subject brightness signal.

The subject brightness differential operation unit 11 is a differential circuit 1
1a, a low-pass filter 11b, and an amplifier 11c, and converts the subject brightness signal into a subject brightness change signal. The differentiating circuit 11a converts the signal into a change signal, the low-pass filter 11b removes high-frequency noise components such as flicker noise of the light source, and the amplifier 11c standardizes the signal size to a required value. Then, the standardized signal is converted into a digital signal by the A / D converter 12.

Of the circuits described above, the low pass filter 1
b has a delay in the output signal with respect to the input signal.
The delay depends on the frequency of the input signal, the cutoff frequency of the filter, and the order of the filter. By setting the cutoff frequency to be sufficiently higher than the shake frequency,
Even at different frequencies of the input shake signal, there is a delay time of one rate.

Now, in order to remove the flicker noise of a fluorescent lamp or the like, it is assumed that a filter is used with a cutoff frequency of 25 Hz and an 8th order configuration. In this case, a delay of about 40 msec occurs. Information on the signal delay of the change in the brightness is stored in the camera shake signal delay information storage unit 6.

Returning to FIG. 2, the camera shake signal delay information storage section 6 stores information on the delay time of the camera shake signal detection section 1 and the actual exposure after the exposure section 14 receives an exposure start signal. Information about a time lag until the execution and information about a time required for various prediction calculations are stored.

The camera shake prediction coefficient setting unit 4 receives the signals from the exposure time setting unit 3 and the camera shake signal delay information storage unit 6 and sets the coefficient for prediction. For this reason, the camera shake prediction coefficient setting unit 4 stores a coefficient group for prediction, which is obtained by the method shown by the relational expression of Expression 29, according to the time to be predicted in a table form. Based on the received signal, the camera shake signal prediction coefficient required for the current prediction is set in a table reference format. Alternatively, the coefficient according to the prediction time may be calculated by the method shown by the relational expression of Expression 23.

The exposure unit 14 is provided with a shutter for performing exposure through an actuator. However, a normal mechanical structure has a mechanical time constant and has a delay at the time of starting. In addition, since a range of the run-up that does not contribute to the exposure is set in the shutter, it takes time to move in that section. Therefore, even if the exposure start instruction signal is received, it is necessary for the exposure to actually start at about 10 msec at the earliest. This delay information is also stored in the camera shake signal delay information storage unit 6.

In this way, the camera shake signal delay information storage unit 6
In addition to the information on the signal delay at the time of detecting the camera shake and the delay at the start of the exposure, the time required for other signal processing is also stored.

The exposure permission control unit 7 which controls permission and prohibition of exposure start based on the predicted camera shake signal outputs an exposure start permission signal when the predicted camera shake signal is equal to or less than a predetermined value, and otherwise. A signal for prohibiting exposure is output to.

The exposure command output section 13 is provided with a so-called second release switch of the camera. By the operation of the second release switch, a signal for instructing the exposure operation is issued to the exposure start signal generator 15 after the focusing is completed and the preparation for the exposure is completed.

The exposure start signal generating section 15 is for outputting a signal instructing the exposure operation from the exposure command output section 13. Further, when the signal from the exposure permission control section 7 permits the exposure, the exposure start signal generating section 15 outputs a signal for instructing the exposure section 14 to start the exposure.
With the configuration as described above, it is possible to obtain an image pickup apparatus that performs exposure only when camera shake in the actual exposure time of subject light is small.

In the present embodiment, it is assumed that prediction is performed from four camera shake signals at a time offset of every 10 msec, including the camera shake signal due to the latest subject luminance change. That is,
0msec past data, 10msec, 20mse
c and 30 msec past data are used for prediction.

The output of the camera shake signal detection unit 1 is 1 msec.
Each time, it is taken into the camera shake signal storage unit 2. The size of the storage area of the camera shake signal storage unit 2, including the storage area for the latest current camera shake data, is required to be 31. Thirty
Past data exceeding msec is updated by new data.

In the camera shake prediction coefficient setting unit 4, a coefficient (A (0)) for the current latest camera shake signal and a coefficient (A (1)) for multiplying the past data by 10 msec, 20 msec and 30 msec, respectively. , A (2), A
(3)) is set.

In the multiplication operation unit 5a, the current latest camera shake signal is X (0), and 10 msec and 20 mse of the past camera shake signals stored in the camera shake signal storage unit 2 are used.
c, 30msec past data, respectively X
Further, as (1), X (2), and X (3), when the calculation result of the data at each time point is represented by the intermediate data (W (i)), the same relational expression as in the above mathematical expression 2 W (i) = (X (i) × A (i)) (where (i) is a natural number from 0 to 3) is calculated.

With respect to the intermediate data (W), the camera shake signal predicting / calculating section 5b immediately carries out the operation represented by the relational expression (30) to obtain the predicted camera shake data (Y).

[0094]

[Equation 30] It should be possible to perform this operation directly by the operation represented by the relational expression of Equation 31 without using the intermediate data (W).

[0095]

[Equation 31]

The exposure permission control unit 7 compares the predicted absolute value of the camera shake signal (Y) with a predetermined value (Th) indicating the amount of camera shake to determine whether or not exposure is permitted. Is judged.

When the signal from the exposure command output unit 13 instructs the exposure operation and the output of the exposure permission control unit 7 permits the exposure, the exposure start signal generation unit 15 causes the exposure unit 14 to perform the exposure operation. Is instructed to start exposure. These series of operations are performed within 1 msec, and the hand movement is prevented based on the predicted hand movement signal every 1 msec.

Next, referring to the flow chart of FIG.
The operation of outputting the exposure start instruction signal to the exposure unit 14 will be described. Here, the camera shake signal storage unit 2, the camera shake prediction coefficient setting unit 4, the multiplication calculation unit 5a, and the camera shake signal prediction calculation unit 5 are provided.
The program flow when b, the exposure permission control unit 7, and the exposure start signal generation unit 15 are mainly configured by a microcomputer will be described. It is assumed that this subroutine is executed after operating the release switch of the camera. Further, in this embodiment, the time interval of the stored camera shake signal used for prediction is 10 msec, and the number of data used for prediction is four. FIG. 5 shows when the subroutine for this operation is called.

First, in step S1, the exposure time setting unit 3
The exposure time (S1) is read from. Then, step S2
Then, the camera shake signal delay information (S2) is read from the camera shake signal delay information storage unit 6. Then, in step S3, the predicted time to the future to be predicted (S) is calculated and set from the relational expression of Expression 32 from the read exposure time (S1) and camera shake signal delay time (S2).

[0100]

[Equation 32] In this embodiment, the upper limit of the prediction time is 100 msec.

Next, in step S4, four types of prediction coefficients (A (0), A
(1), A (2), A (3)) are set. for that reason,
As shown in Table 2, the data is stored in a table form and the value corresponding to the predicted time (S) is read.

[0102]

[Table 2]

The coefficients stored in the table are previously
It is a value that is optimized so that the timing of blurring can be predicted from the signal of the camera shake detection means used according to the prediction time.

Then, in step S5, a continuous RAM area for storing the camera shake signal is stored in the data pointer used for writing the camera shake signal in the free write / read storage area (RAM) of the microcomputer. The start address (# V0) of This is the start address of the area where 31 pieces of data including the latest camera shake signal are stored.

Then, in step S6, the camera shake signal necessary for the prediction is stored in the camera shake signal storage unit 2 (RA described above).
The pre-data counter as a counter for detecting that it is stored in the (M area) is cleared. This counter counts up until the first 31 pieces of data are stored.

Further, in step S7, subject brightness change data as a camera shake signal is read from the camera shake signal detector 1. Then, in step S8, the camera shake signal is stored in the RAM area indicated by the data pointer using a general indirect addressing method.

Next, in step S9, the value of the pre-data counter is checked to see if a sufficient number of data for prediction has already been stored in the camera shake signal storage unit 2. If sufficient data is not yet stored, the process proceeds to step S10, and if necessary data is available, the process proceeds to step S15 described later.

When the process proceeds to step S10, first, the pre-data counter is incremented and incremented by 1 in step S10. Then, in step S11, the value of the data pointer is incremented to indicate the area of the next data. Then, in step S12, it is checked whether or not the value of the data pointer exceeds the area for storing the camera shake signal. In the case of the embodiment, as described above, since 31 pieces of data including the latest camera shake signal are stored, when the storage area exceeds 30 from the start address,
In step S13, the value of the data pointer is reset to the head address (# V0).

Succeedingly, in a step S14, it is determined whether or not the processing of this routine is to be ended. The signal for termination is
It is determined by the switch operation of the photographer or the elapsed time after the control is started. If the termination condition is met, the processing of this subroutine is terminated. If the conditions for termination are not met, the process returns to step S7. The condition for ending the subroutine is applicable to the case where the exposure start instruction signal is issued, the case where the process is interrupted by the operation of the photographer, and the like.

If it is determined in the above step S9 that the past data sufficient for prediction has been accumulated, the process proceeds to step S15, and first, the latest camera shake signal indicated by the current data pointer value. The data is read from the storage address of and is assigned to the variable (X (0)). here,
If you still have the latest data captured, you may substitute that value.

Next, steps S16 to S18
At, the data of 10 msec past is read. For that purpose, first, it is checked whether the data address of 10 or less from the current value of the data pointer is larger than the head address (# V0) of the RAM area storing the camera shake signal.

Here, if it is larger than # V0, the data of 10 msec past is read out by using the address information having a value smaller by 10 data than the data pointer by the indirect address and the variable (X
(1)) is substituted. In step S16, when the value of the data pointer-10 is smaller than the start address (# V0), 10 msec past data is stored in the address 21 larger than the current data pointer value. .. Therefore, the process proceeds to step S18 and the value is assigned to the variable (X (1)).

Next, steps S19 to S21.
In the same manner as above, the data of 20 msec past is read and substituted into the variable (X (2)). Further, in steps S22 to S24, 30 msec past data is read by the same method as described above, and the variable (X
(3)) is substituted.

The present and the past 10 thus obtained
The predicted camera shake signal (Y) is calculated in steps S25 to S28 using the four camera shake signals of msec, 20 msec, and 30 msec and the prediction coefficient set in step S4 described above. Here, the operations of the multiplication calculation unit 5a and the camera shake signal prediction unit 5b are developed in software in a complex manner.

First, in step S25, the value of the result of multiplication (X (0) × A (0)) is substituted into the variable (Y) of the predicted camera shake signal. Then, from step S26 to step S
28, the variable (Y) is multiplied (X (1) × A
(1)) result, multiplication (X (2) × A (2)) result,
The multiplication (X (3) × A (3)) is successively added and substituted. The variable (Y) obtained as a result is the predicted shake signal value.

Then, in step S29, it is determined whether or not the camera shake prediction signal means that the camera shake is small. As a reference value for making a judgment for that,
The predetermined value (Th) is compared with the absolute value of the predicted camera shake signal (Y). When the predicted camera shake signal (Y) is less than or equal to the predetermined value (Th), it is determined that there is no camera shake, and the process proceeds to step S30. On the other hand, when the predicted camera shake signal (Y) is larger than the predetermined value, the process returns to step S11 and the above-described operation is repeated.

In step S30, the exposure command signal of the exposure command output unit 13 is read. Then, step S31
At, it is determined whether the exposure command signal indicates the exposure operation. If not, the process returns to step S11. Then, it progresses to step S14. Next, a second embodiment of the present invention will be described.

In this embodiment, a photograph taken by detecting a camera shake with an angular velocity sensor, moving an actuator based on the detected value, and changing the position of the image pickup member with respect to the optical axis of the lens in conjunction with this actuator. It is applied to prevent deterioration due to camera shake. As shown in FIG. 3, three axes of x, y, and z are set for the camera.

In this embodiment, in order to eliminate the influence of rattling of the mechanical connecting portion of the image stabilizing unit 9 (FIG. 1C) that moves the optical axis, and to simplify the control of the actuator, the exposure is performed. This is an example of a camera-shake preventing device that moves the optical axis linearly at a constant velocity by approximating the camera-shake to a straight line.

Therefore, as shown in FIG. 1C, the camera shake signal average calculator 8 takes the time average of the signal of the predicted camera shake signal detector 1 based on the output of the exposure time setter 3.
It is stored in the camera shake signal storage unit 2 via. Then, based on this value, in order to prevent camera shake during exposure, the optical axis of the subject image that reaches the imaging system through the taking lens of the optical system is
Before the exposure is started, the driving direction and the driving speed of the camera shake correction unit 9 that positively move the camera shake correction direction are determined.

FIG. 6 is a block diagram showing the structure of the second embodiment. In the figure, the angular velocity sensor 16 is
In order to detect camera shake in the lateral direction of the screen of the camera, its sensitivity axis is aligned so as to detect the angular velocity due to rotation around the y axis. The angular velocity sensor 17 is installed so as to detect the angular velocity of rotation around the x-axis in order to detect camera shake in the vertical direction of the screen. Each of the angular velocity sensors 16 and 17 has an A / D converter (not shown) at the output stage, and its output is digitized. The shake signal of each axis is sampled every 2 msec.

The x-direction camera shake signal average calculator 18 and the y-direction camera shake signal average calculator 19 respectively take the average value of the angular velocity sensor signal for the time equivalent to the exposure time set by the exposure time setting unit 3. And Also, the maximum exposure time for performing this calculation is 50 msec, that is, 1 /
20 seconds.

Therefore, the camera shake signal average calculation section for each axis has a memory for storing 26 camera shake signals. The latest 26 camera shake signals are stored in each of the x and y directions, and the averaging calculation is performed on the data of the time corresponding to the exposure time for each of the x direction and the y direction. That is, a total of 52 data storage areas are required.

In this embodiment, with respect to camera shake in both the x-direction and the y-direction, the average camera shake signal from the latest camera shake signal averaging unit is included in each of six average camera shake signals at a time offset of 8 msec. Shall be predicted. That is, 0 msec past data and 8 ms for each axis
ec, 16 msec, 24 msec, 32 msec, 4
It is estimated using the past data of 0 msec.

In this case, the size of the storage area of the camera shake signal storage unit 2 including the storage area for the latest and current average camera shake signal is required to be 21 per axis. I need one. Also, past data exceeding 40 msec is updated by new data.

Further, in the camera shake prediction coefficient setting unit 4, the coefficient (A (0)) for the current latest camera shake signal and 8 msec, 16 msec, 24 msec and 32 m, respectively.
sec, 40 msec Coefficients A (1), A (2), A (3), A (4), A (5) for multiplying past data
Are set based on the output signals of the exposure time setting unit 3 and the camera shake signal delay information storage unit 6.

The hand-shake signal delay information storage unit 6 stores information about the estimated time such as the time required for the estimation other than the exposure time, the sensor delay, the actuator delay, and the calculation time.

In the multiplication operation section 5a, the latest camera shake signal at present is taken as Xx (0) and Xy in the x and y directions, respectively.
(0) and 8 msec, 16 msec, and 24 ms among the past camera shake signals stored in the camera shake signal storage unit 2.
ec, 32 msec, and 40 msec of the past data, Xx (1), Xy (1), Xx (2),
Xy (2), Xx (3), Xy (3), Xx (4), X
As y (4), Xx (5), and Xy (5), the predicted camera shake signal (Yx, Yy) in the x and y directions is calculated by the equation 33 and by the camera shake signal prediction calculation units 5b1 and 5b2 in the x direction and the y direction. It is obtained by the relational expression of the equation 34.

[0129]

[Expression 33]

[0130]

[Equation 34]

The x-direction camera shake correction unit 20 and the y-direction camera shake correction unit 21 are provided in the x-direction camera shake signal prediction calculation unit 5b.
Based on the predicted camera shake signal (Yx, Yy) obtained by the 1 and y-direction camera shake signal prediction calculation unit 5b2, according to the exposure start command of the exposure command output unit 13, the actuator speed control is performed with the predicted camera shake signal, This is to eliminate camera shake.

The exposure command output unit 13 detects the operation of the second release switch, prepares for exposure such as focus adjustment, and outputs an exposure start command when the preparation is completed. In accordance with this exposure start command, the x-direction and y-direction image stabilization units 20 and 21 and the exposure unit 14 that drives the shutter are activated.

FIG. 7 shows the camera shake correction unit 2 in the x and y directions.
It shows an example of the configuration of 0 and 21. In this embodiment, the optical axis center position of the image pickup means such as a film is set to x
It has a mechanism for translating in the y-direction. Therefore, the screen frame 22 is provided on the x-direction guide 23 that allows movement only in the x-direction, and is installed on the y-direction guide 24 that allows movement only in the y-direction.

The screen frame 22 has a piezoelectric actuator unit 25 for driving in the x and y directions.
26 is attached via slide members 27 and 28 made of rollers. Piezoelectric actuator unit 25,
Reference numeral 26 is composed of a laminated piezoelectric body, a displacement magnifying mechanism, and a drive circuit, which causes deformation at the instructed speed and drives the screen frame 22. The screen frame 22 is biased by a spring to each piezoelectric actuator unit.

Next, the above-described camera shake signal storage section 2, camera shake prediction coefficient setting section 4, multiplication calculation section 5a, camera shake signal prediction calculation sections 5b1 and 5b2, x-direction and y-direction camera shake signal average calculation sections 18 and 19 are provided. , FIG. 8 shows the flow of the program when the microcomputer is mainly configured.
9 and the flowchart of FIG. In FIGS. 8 and 9, this subroutine is executed after the operation of the first release switch of the camera.

Assuming that this calculation subroutine is called, first, in step S41, the exposure time (S1) is read from the exposure time setting unit 3. Then, in step S42, the shake signal delay information (S
2) is read. Then, in step S43, the predicted time (S) to the predicted future is calculated and set from the read exposure time (S1) and camera shake signal delay time (S2). When the unit is msec, the relational expression of the above Expression 32 is established. In the embodiment, the upper limit of the prediction time is 50 ms.
ec.

Next, in step S44, six types of prediction coefficients (A (0), A used for predicting the camera shake signal are used.
(1), A (2), A (3), A (4), A (5)) are set. Therefore, as shown in Table 2, the data is stored in a table and the value corresponding to the predicted time (S) is read. The coefficients stored in the table are values that have been optimized in advance so that the speed for appropriate linear image stabilization drive can be predicted from the signal from the camera shake signal detection unit used according to the prediction time. Is.

Next, in step S45, the head address (# VM0) of the RAM area for storing the camera shake signal is assigned to the data pointer 1 in order to average the camera shake signal from the sensor. This RAM area can store 52 pieces of data. Then, in steps S46 and S47, the addition data variable (S
x, Sy) is cleared. Furthermore, in step S48,
Variable (Nm) representing the number of data to be averaged
Then, the data is set according to the relational expression of Expression 35.

[0139]

[Equation 35] This is because the data is sampled once every 2 msec and there is data at the beginning and end of the time.

Next, in step S49, the start address (#) of the continuous RAM area for storing the camera shake signal is stored in the data pointer 2 used for writing the averaged camera shake signal in the RAM. Substitute V0). This is the start address of an area in which 26 pieces are stored in the x and y directions including the latest camera shake signal.

In step S50, the pre-data counter as a counter for detecting that the camera shake signal necessary for the prediction is stored in the camera shake signal storage unit 2 (the RAM area) is cleared. This counter is
First, the count-up is performed until 21 × 2 averaged data required for prediction are stored. That is, when the averaging work is included, (Nm + 21) angular velocity samples are counted.

Next, in step S51, the angular velocity sensors 16 and 17 are read as a camera shake signal. Then, in step S52, the addition value variable (S
x, Sy) is added by the relational expression of Expression 36.

[0143]

[Equation 36]

Here, in step S53, the value of the pre-data counter is checked to see whether or not the number of camera shake sensor data sufficient for performing the averaging operation is already stored. If sufficient data is not yet stored, the process proceeds to step S63 described later, and if necessary data is obtained, the process proceeds to step S54 to perform an arithmetic operation for averaging the sensor signals.

From step S54 to step S59,
In order to remove the data before the exposure time, it is first checked whether (data pointer 1-2 × Nm) is smaller than # Vm0. Here, when it becomes smaller, the address of the data to be read is corrected, and the oldest data added to the previous addition value, that is, (S1 + 2) msec past data is subtracted from the addition value. Then, the relational expression of Expression 37 is established.

[0146]

[Equation 37] Also, in steps S60 and S61,
The average value represented by the relational expression is calculated.

[0147]

[Equation 38]

Next, in step S62, the data pointer 2 is read by using a general indirect addressing method.
The averaged camera shake signal is stored in the RAM area indicated by. Then, in step S63, the value of the pre-data counter is checked to see if a sufficient number of data for prediction has already been stored in the camera shake signal storage unit 2. Here, if sufficient data is not yet stored, the process proceeds to step S64, and if the necessary data are available, the process proceeds to step S72 described later.

When the process proceeds to step S64, first, at step S64, the pre-data counter is incremented and incremented by 1. Next, in steps S65 and S66, the values of the data pointers 1 and 2 are incremented by two to indicate the area of the next data.

Then, in step S67, it is checked whether or not the value of the data pointer 1 exceeds the area for storing the camera shake signal of the averaging operation. In the case of the embodiment, as described above, a total of 52 data including the latest camera shake signal can be stored. Therefore, when the number of storage areas exceeds 51 from the start address, the process proceeds to step S68 and the value of the data pointer is reset to the start address (# VM0).

In step S69, it is checked whether the value of the data pointer 1 has exceeded the area for storing the camera shake signal. Here, as described above, since a total of 42 pieces of data including the latest camera shake signal are stored, when the storage area exceeds 41 pieces from the start address, the process proceeds to step S70 and the value of the data pointer is changed. Start address (#
Reset to V0).

Succeedingly, in a step S7, it is checked whether or not the processing of this routine is to be ended. The signal for the end depends on the switch operation of the photographer. If the termination condition is met, the processing of this subroutine is terminated. On the other hand, if the termination condition is not met, the above step S51 is performed.
Return to. The termination condition applies when an exposure start command is issued, when the processing is interrupted by an operation of the photographer, and the like.

If it is determined in step S63 that past data sufficient for prediction has been accumulated, steps S72 to S76 are the same as those described in the first embodiment. Using the method, and also remembering that the data for the two axes x and y are stored,
ec, 16 msec, 24 msec, 32 msec, 4
0msec past data (Xx (1), Xy (1), X
x (2), Xy (2), Xx (3), Xy (3), Xx
(4), Xy (4), Xx (5), Xy (5)) are read.

Using the current and past averaged camera shake signals thus obtained and the prediction coefficient set in step S44, steps S177 to S8 are performed.
At 8, the predicted camera shake signal (Yx, Yy) is calculated. Here, the operations of the multiplication calculation unit 5a and the x-direction and y-direction camera shake signal prediction calculation units 5b1 and 5b2 are developed in software in a complex manner.

Then, in step S89, the predicted camera shake signal Yx for driving is output to the x-direction camera shake correction section 20, and in step S90 the predicted camera shake signal Yy is output to the y-direction camera shake correction section 21, and then the process returns to step S65. The operation of is repeated.

By the way, when the focal length is changed like an interchangeable lens or a zoom lens and the relationship between the angular velocity and the blurring of the image changes, the focal length detecting means is set so that the coefficient of this prediction can be set according to the focal length. It is also possible to include it. This can be achieved by correcting the same prediction coefficient according to the focal length.

Even when the distance varies depending on the subject distance, the accuracy is improved by detecting the subject distance and setting the coefficient in consideration of the distance, or by correcting the coefficient according to the distance. be able to.

In the example described above, the camera shake signal detecting section 1
The example using the subject brightness detection unit and the angular velocity sensor is shown in the above. However, even in the case of camera shake detection using an image pickup device such as a CCD, of course, there are differences in the method and effect in principle. Absent.

[0159]

As described above, according to the present invention, the exposure time can be changed by switching the coefficient used for the prediction by the product-sum operation to change the prediction time according to the length of the exposure time to the image pickup system. It is possible to predict the magnitude and speed of the blurring at the central time point of exposure even when the values are different.
Therefore, when the exposure is started at the timing of reducing the blurring at the central point of the exposure, it is possible to accurately determine the start point of the exposure in accordance with the exposure time. Even when the shake correction mechanism is driven in one direction at a constant speed during exposure to reduce the shake of the image pickup system, the shake speed at the central point of exposure is set according to the exposure time before the start of exposure. It becomes possible to obtain a shake correction system by driving the correction mechanism in a direction to cancel the shake.

[Brief description of drawings]

FIG. 1 is a block diagram showing a basic configuration example of an image stabilization apparatus of an image pickup apparatus according to the present invention.

FIG. 2 is a block diagram showing a basic configuration of a camera shake preventing device for an image pickup apparatus according to the first embodiment of the present invention.

FIG. 3 is a diagram showing a relationship between an x-axis, a y-axis, and a z-axis with respect to a camera.

FIG. 4 is a diagram showing a detailed configuration of a camera shake signal detection unit 1 in FIG.

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

FIG. 6 is a block diagram showing a configuration of a second exemplary embodiment of the present invention.

FIG. 7 is a diagram showing a configuration example of x-direction and y-direction image stabilization units 20 and 21.

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

FIG. 9 is a flowchart illustrating the operation of the second embodiment.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Shake signal detection part, 2 ... Shake signal storage part, 3 ... Exposure time setting part, 4 ... Shake prediction coefficient setting part, 5 ... Calculation part, 5a ... Multiplication calculation part, 5b ... Shake signal prediction calculation part,
5b1 ... X-direction shake signal prediction calculation unit, 5b2 ... Y-direction shake signal prediction calculation unit, 6 ... Shake signal delay information storage unit, 7 ... Exposure permission control unit, 8 ... Shake signal average calculation unit,
Reference numeral 9 ... Shake correction unit, 10 ... Subject brightness detection unit, 11 ... Subject brightness differential calculation unit, 12 ... A / D converter, 13 ...
Exposure command output unit, 14 ... Exposure unit, 15 ... Exposure start signal generation unit, 16, 17 ... Angular velocity sensor, 18 ... X direction camera shake signal average calculation unit, 19 ... Y direction camera shake signal average calculation unit,
20 ... x-direction image stabilization unit, 21 ... y-direction image stabilization unit.

Claims (3)

[Claims] 1. An image stabilization apparatus in an image pickup apparatus for recording an image of a subject, and an image stabilization signal detecting unit for detecting an image stabilization vibration, and an image stabilization unit for storing at least one output from the image stabilization signal detecting unit. The signal storage means, the exposure time setting means for automatically or manually setting the length of the exposure time, and at least two or more kinds of prediction coefficients for predicting the camera shake signal at the time of photographing are output to the exposure time setting means. A calculation for calculating a predicted camera shake signal by multiplying and adding the camera shake prediction coefficient setting means set based on the above, the output of the camera shake signal detection means and / or the output of the camera shake signal storage means and the output from the camera shake prediction coefficient setting means. An image stabilization device for an image capturing apparatus, comprising: 2. A correction means for correcting camera shake based on the output of the calculation means, at least a delay time of detection of the camera shake signal detection means, and a delay time of operation of the camera shake correction means. Delay information storage means for storing one information, and the camera shake prediction count setting means sets the prediction coefficient based on the output signal of the exposure time setting means and the output signal of the delay information storage means. An image stabilizer for an image capturing device, which is characterized in that 3. The camera shake signal averaging means for averaging the camera shake signal based on the output of the exposure time setting means, and the output of this camera shake signal averaging means. At least one of a driving direction and a speed of the camera shake preventing means is determined before the exposure is started by storing the camera shake signal storage means in the camera shake preventing apparatus of an image pickup apparatus.