JPH0918779A - Vibration buffer device - Google Patents

Abstract

PURPOSE: To relax vibration in a short time by providing a signal processing means conducting all of multiplication processing through arithmetic shift operation to the device so as to conduct all the multiplication processing in the signal processing means by means of a digital arithmetic shift operation. CONSTITUTION: A gyro 11 is used for a detection means to detect an angular velocity of a behavior of the device. An angular velocity signal being an output of the gyro 11 is given to an operational amplifier circuit 12, in which the signal is amplified by a prescribed gain (set by a resistor R11 and a variable resistor VR 12). The amplified angular velocity signal is converted into a digital output V1 by an A/D converter circuit 13 immediately and then processed digitally. Then the device is provided with a signal processing means conducting multiplication processing by arithmetic shift operation, all of arithmetic processing is conducted by digital arithmetic shift operation in the signal processing means and the multiplication processing in the signal processing means is eliminated. Thus, an output of the gyro 11 uncertain in terms of DC is changed around a prescribed potential and the hand-shake buffer device is realized with a very small circuit scale and it is suitable for circuit integration.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an improvement of a shake reducing device applied to a photographing device such as a movie camera.

[0002]

2. Description of the Related Art An example of the circuit configuration of a conventional device of this type is shown in FIG. 4 and described below.

This example is an application to an image stabilization system of a movie camera, and has the structure shown in Japanese Patent Application No. 6-309560.

A gyro 11 is used as a detecting means to detect the angular velocity of the behavior of the device. The angular velocity signal output from the gyro 11 is amplified by the operational amplifier circuit 12 with a predetermined gain (set by R11 and R12).
Then, the amplified angular velocity signal is immediately converted into a digital value output (V1) by the A / D conversion circuit 13, and thereafter digitally processed.

This output (V1) is used for three processes.

The first destination of the output (V1) is the multiplication means 257 (GAIN). The role of the multiplication means 257 is an adjustment means for correcting the sensitivity variation of the gyro 11 and is multiplied by the value obtained by converting the adjustment value of the external volume (VR20) by the A / D conversion circuit 18. The angular velocity signal having a constant sensitivity characteristic by the adjusting means is integrated by the integrating means 251 and converted into an angle signal. The integrating means has a characteristic of attenuating with a predetermined time constant, so that it is a low-pass filter having a fairly low cut-off frequency.

On the other hand, the output (V2) of the integrating means 251 is converted into an analog signal by the D / A conversion circuit 16,
It is transmitted to the correction means 17 by the optical system, and the camera shake is optically corrected. On the other hand, it is transmitted to the control means 252, where it is converted into a gain setting value (KK). Control means 252
Then, the absolute value of the output (V2) is taken and the gain setting value from the minimum to the maximum is selected by a predetermined function.

The gain setting value KK is variable gain amplifier circuit 2
It is input to 53, whereby the amplification factor changes from 0 times to 1 time with respect to the output (V1).

The second destination of the output (V1) is the above-mentioned variable gain amplifying means 253, the output of which is multiplied by a constant gain (K5) by the multiplying means 258 and added to the high pass filter 255. Obtain the output (V4).

The third destination of the output (V1) is a predetermined gain amplifier circuit 254, where K_FIL times (V
5) is added to the above output (V4) and output (V
6) is input to the integrating means 256 and integrated. The output (V7) of the integrating means 256 is the D / A conversion circuit 14
Is again converted into an analog signal by the operational amplifier circuit 1
2 is added to the positive input.

The processing from the A / D conversion circuit 13 to the D / A conversion circuit 16, that is, the processing in the integrated circuit IC will be described below.

A first-order low-pass filter is formed by the amplifier circuit constituted by the operational amplifier circuit 12, R11 and R12, the predetermined gain amplifier circuit 254, and the integrating means 256. As a result, the output of the gyro 11, which is indeterminate in terms of direct current, changes around a predetermined potential and is applied to the A / D conversion circuit 13 as a signal that has undergone the necessary amplification.

In addition, at the time when the system needs to be initialized, such as when the power is turned on, the predetermined gain amplifying circuit 254 for only the necessary time is required.
It is possible to quickly set the system to a stable state by setting the predetermined gain of 1 to a large value and operating the system.

When the device greatly changes the angle, such as during panning, the output (V2) of the integrating means 251 tends to deviate largely from the center. Then, the control means 25
2 detects this, and changes the gain of the variable gain amplifying means 253 to the larger one.

A / D conversion circuit 13 to D / A conversion circuit 1
The processing parts up to 6 may be configured by a hardware logic circuit or may be realized by a computer.

An example of a program when implemented by a computer is shown below.

The description language is C (Compiler; Compil)
er), and only the digital processing portion is extracted and described using the variables corresponding to the respective values shown in FIG.

The constants are examples when the sampling frequency is about 500 Hz, and the gyro sensitivity is about 1 mV / deg.
/ Sec, the amplifier gain is about 45 times, the optical correction amount of the correction system is about 1 deg / V, and the unit of each variable is Vo.
It is lt.

[0019] When the power is turned on, initialization can be realized by setting the constant K_FIL to a value having a larger absolute value and repeating the process a predetermined number of times, and turning the drive ON / OFF can be realized by switching the constant GAIN to zero.

The above operation will be described with reference to the flow chart shown in FIG.

The system shown here starts its operation when the power is turned on (step 501).

First, the initial setting of the entire system including hardware. This is performed in a loop of processing from step 502 to step 506. In step 502, constants are set so that each variable easily converges.
In 03, the operation from the input of the A / D conversion circuit 13 to the output of the D / A conversion circuit 16 is performed.
The calculation from the input of the / D conversion circuit 13 to the output of the D / A conversion circuit 14 is performed, and in the subsequent step 505, the delay element is updated.

The constant K_FIL is set to an absolute value larger than a normal value, so that the feedback loop including the operational amplifier circuit 12 quickly converges to a stable state. Here, the loop count is set to 100, and the stable state is reached in about 0.2 seconds.

When the initial setting is completed, each constant in the normal operating state is set in step 507, and the normal 500 Hz sampling infinite loop is entered.

First, in step 508, ON / OFF of the anti-vibration switch is checked, and when it is ON, GAIN is set to an appropriate value, and when it is OFF, it is set to zero.

Step 509 is a calculation process from the input of the A / D conversion circuit 13 to the output of the D / A conversion circuit 16.

In the next step 510, the process of obtaining the absolute value of the integrated output value is performed, then in step 511 the process of the polygonal line function, and in the processes of the following steps 510 and 511, the conversion shown in FIG. 6 is performed. The value KK obtained here indicates the strength of the panning control.

At the next step 512, the A / D conversion circuit 1
The arithmetic processing from the input of 3 to the output of the D / A conversion circuit 14 is performed, and the value KK obtained above is involved.

In the following step 513, the delay element is updated.

The above is the processing of one sampling,
The process waits until the next sampling process starts.

In the conventional example shown above, several multiplication processes are performed, and the assets (multiplication circuit in the case of hardware processing, multiplication time in the case of software processing) spent on this processing are It accounted for a considerable amount.

[0032]

SUMMARY OF THE INVENTION As described above,
In the prior art, many assets were spent on the multiplication process.

(Object of the Invention) It is an object of the present invention to provide a shake mitigation device capable of reducing the circuit scale and reducing the shake in a short time.

[0034]

In order to achieve the above object, the present invention according to claim 1 is provided with signal processing means for performing all multiplication processing by arithmetic shift operation, and all the signal processing means are provided. The multiplication process is performed by digital arithmetic shift calculation so that the multiplication process is eliminated from the signal processing means.

In order to achieve the above-mentioned object, the present invention according to claim 2 further comprises a gain adjusting means for adjusting a predetermined gain of the amplifying means, and eliminates the multiplication processing from the signal processing means, thereby providing the gain adjusting means. The amplification means is used.

[0036]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, the present invention will be described in detail with reference to the illustrated embodiments.

In this embodiment, the technique disclosed in Japanese Patent Application No. 6-309560, that is, the configuration shown in FIG.
Replace the multiplication process with 1) arithmetic shift operation.

2) Replace with an analog circuit or the like, and drive it to the outside. The two types of changes are added to finally realize a device that does not require any multiplication means.

More specifically, in the configuration of FIG. 4, the objects to be replaced are constant multiplication K_FIL constant multiplication K3 constant multiplication K_INT constant multiplication K5 variable multiplication KK variable multiplication GAIN. It is a constant value that is determined, and is realized by replacing it with a simple arithmetic shift process or a combination of arithmetic shift process and subtraction, and regarding it as multiplication of close values.

Is realized by switching some arithmetic shift processes having different constants according to the output value of the control means 152 described later.

Since the gain adjusting means requires a certain degree of accuracy, it cannot be said that a close value is used as a substitute.

Therefore, this part is realized by substituting an appropriate scaling by a constant arithmetic shift operation and providing an analog adjusting means outside the digital processing part in order to further satisfy the accuracy. Further detailed numerical values will be described in the following examples.

(First Embodiment) FIG. 1 is a block diagram showing a circuit configuration according to a first embodiment of the present invention. Here, it is assumed that an image stabilization apparatus for a movie camera is used.

The gyro 11 is used as the detecting means to detect the angular velocity of the behavior of the device. The angular velocity signal output from the gyro 11 is amplified by the operational amplifier circuit 12 with a predetermined gain (set by the resistor R11 and the volume VR12). Then, the amplified angular velocity signal is immediately converted into a digital value output (V1) by the A / D conversion circuit 13, and thereafter digitally processed.

This output (V1) is used for three processes.

The first destination of the output (V1) is the integrating means 151, and the angular velocity signal is integrated here and converted into an angle signal (V2). The integrating means has a characteristic of attenuating with a predetermined time constant, so that it is a low-pass filter having a fairly low cut-off frequency.

In the conventional example, this low-pass filter is realized by performing a process of multiplying the output of the input delay value by a predetermined constant (K_INT: 0.9999) and adding the result to the input. + Input delay value − Input delay value is 12-bit right arithmetically shifted to obtain the same function.

To be precise, the value corresponding to K_INT is "0.999.
It is "0.99976" instead of "9", but it is not a difference that can be recognized by humans.

On the other hand, the output (V2) of the integrating means 151 is converted into an analog signal by the D / A conversion circuit 16,
It is transmitted to the correction means 17 by the optical system, and the shake is optically corrected. On the other hand, it is transmitted to the control means 152, where it is converted into a gain setting value (KK). This control means 15
At 2, the absolute value of the output (V2) is taken and the gain setting value from the minimum to the maximum is selected by a predetermined function.

The gain setting value KK is gain switching means 153.
Is controlled to switch the gain with respect to the output (V1) in a predetermined step.

The second destination of the output (V1) is the gain switching means 153 described above, the output of which is scaled to a constant gain (K5) and added to the high-pass filter 155 to output the output. (V4) is obtained.

K5 is a constant multiplication of "0.002" in the conventional example, but it is a 9-bit right arithmetic shift in this embodiment, and is precisely a constant multiplication of "0.00195". In the filter 155, the constant multiplication of "K3 = 0.995" was performed in the conventional example, but in the present embodiment, the right arithmetic shift of bits is subtracted from 1 to be a constant multiplication of "0.996" accurately.

The third destination of the output (V1) is the predetermined gain means 154, which is multiplied by K_FIL (V5) in the scaling process of the predetermined gain, and added with the above output (V4) to output (6). And integrated by the integrating means 156.

Although the scaling of the predetermined gain means 154 is a constant multiplication of "0.0003" in the conventional example, it is a right arithmetic shift of 12 bits in the present embodiment, and precisely "0.00024".
4 ”is a constant multiplication.

The output (V7) of the integrating means 156 is D /
The signal is again converted into an analog signal by the A conversion circuit 14,
It is applied to the positive input of the operational amplifier circuit 12.

As described above, the replacement from any multiplication is a slight difference compared to the conventional example, and is not a difference that can be recognized by humans.

A / D conversion circuit 13 to D / A conversion circuit 1
The processing up to 6 will be described below, that is, the processing in the integrated circuit IC.

An amplifier circuit composed of the operational amplifier circuit 12, R11 and VR12, a predetermined gain means 154, and
The structure of the integrating means 156 forms a first-order high-pass filter. As a result, the output of the gyro 11 which is uncertain in terms of DC changes around a predetermined potential, and
It is added to the A / D conversion circuit 13 as a signal subjected to the necessary amplification.

The cutoff frequency of this high-pass filter actually depends on the closed loop gain of the amplifier circuit, and its transfer characteristic H (s) is expressed by the following equation.

H (s) = − G / (1+ (1 + G) / Ts) where G is a gain and T is a time constant determined by the predetermined gain means and the integration means. That is, the cutoff frequency changes in proportion to “1 + G”.

However, the quality of the vibration gyro for detecting a shake in recent years has been considerably improved, and a variation in sensitivity is suppressed to about ± 5%. Therefore, by using such a gyro, the adjustment range may be ± 5%, and the change in the cutoff frequency can be suppressed within ± 5%.

It can be said that the camera shake reduction system of the movie camera according to the present embodiment has no practical problem.

According to this system, the image stabilizing device can be realized with an extremely small circuit scale as described above, and it can be said that it is suitable for integration.

(Second Embodiment) A second embodiment of the present invention shows an example of a program for realizing the same effect as that of the first embodiment by a computer.

Since various constants are set to be almost equal to those of the conventional example, refer to the description of the conventional example.

A program (excerpt) written in the language C is shown below, and its operation will be described with reference to the flowchart of FIG.

Here, SH (val, shift) is defined as an arithmetic right shift function, which is a value obtained by shifting the input variable val to the right by shift times.

[0068] The following is a description of the flowchart of FIG.

Starting from the start of step 301, the initial setting is for the entire system. This is performed in the loop from step 303 to step 306.

At step 303, the operation from the input of the A / D conversion circuit 13 to the output of the conversion circuit 16 is performed, and at the next step 304, the operation from the input of the A / D conversion circuit 13 to the output of the D / A conversion circuit 14. However, in the following step 305, the delay element is updated.

The shift amount in step 303 is set to a value smaller than the normal value, and the feedback loop including the amplifier 12 is quickly converged to a stable state. Here, the loop count is set to 100, and the stable state is reached in about 0.2 seconds.

When the initialization is completed, a normal 500 Hz sampling infinite loop is entered.

First, in step 308, integration processing is performed on the output of the D / A conversion circuit 16, and then in step 30.
ON / OFF of the anti-vibration switch is checked at 9,
When it is ON, the input value from the A / D conversion circuit 13 is appropriately scaled and added to the output of the integration processing.

In step 310, the absolute value of the integrated output value is obtained, in step 311, the polygonal line function is processed, and in the subsequent steps 310 and 311, the variables shown in FIG. 3 are executed. The strength of the panning control is determined by the evaluation of the value (KK) obtained here. Output (V
When the absolute value of 2) is 0.3 or less, zero is assigned to the output (V3), and when it is 0.3 or more, the shift amount for the output (V1) is selected stepwise and assigned to the output (V3). is there.

At step 312, the arithmetic processing from the input of the A / D conversion circuit 13 to the output of the D / A conversion circuit 14 is performed.
The output (V3) obtained above is involved.

At the next step 313, the delay element is updated.

The above is the processing of one sampling,
The process waits until the next sampling process starts.

According to the present system, it is possible to realize an apparatus capable of alleviating camera shake in an extremely short time as described above.

According to each of the above-mentioned embodiments, all the multiplication processing is replaced with the arithmetic shift operation, or the variable gain amplifying means such as an external analog circuit is replaced so that the multiplication processing is eliminated from the digital processing section. , The circuit scale became very small, and it became possible to construct a shake reduction system at low cost. This is a very advantageous condition for circuit integration.

(Modification) Two embodiments of a hardware configuration and a computer software configuration have been described above, but the present invention is not limited to the configuration described above, and various modifications are possible. It can be applied to systems with various configurations.

Further, although the present invention has been described in the case of being applied to a movie camera, the present invention is not limited to this, and is an apparatus that is operated by a human being while holding it with a hand or the like and does something with respect to a certain object. However, it is applicable when it is desirable to mitigate the effect of minute vibration (camera shake, etc.) that occurs with the operation.

That is, both a motion based on the intention of the operator, such as pointing the device toward a certain object, and a motion that occurs against the intention act on the device. It suppresses the movement of and transmits only the former movement.

Specifically, it can be applied to a photographing device such as a still camera other than the above-mentioned movie camera, an observing device such as a scope, an illuminating device such as a laser stick or a flashlight used for a presentation, and the like. It is possible to stabilize the image, the observation image, and the irradiation position and illumination position.

By changing the setting of the frequency and the threshold value, not only the vibration caused by the human operation but also the jerky feeling of the mechanical operation can be alleviated. For example, the present invention can be applied to a device that smoothes the movements of the limbs of a robot and a device that dampens vibrations of a vehicle while it is running.

The detecting means may be any of a velocity sensor, an angular velocity sensor and an angular acceleration sensor.

Further, the correcting means is not limited to the variable apex angle prism shown in FIG. 1, but a method of shifting or rotating the lens in the direction perpendicular to the optical axis, a method of changing the reflection angle of the mirror, and May be a method of electronically changing the cutout area of the imaging screen.

[0087]

As described above, according to the present invention,
A signal processing means for performing all multiplication processing by arithmetic shift calculation is provided, all multiplication processing in the signal processing means is performed by digital arithmetic shift calculation, and the multiplication processing is eliminated from the inside of the signal processing means.

Further, according to the present invention, the gain adjusting means for adjusting the predetermined gain of the amplifying means is provided, the multiplication processing is excluded from the signal processing means, and the gain adjusting means and the amplifying means are used.

Therefore, it becomes possible to provide a shake mitigation device capable of reducing the circuit scale and mitigating the shake in a short time.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of a shake mitigating device according to a first embodiment of the present invention.

FIG. 2 is a flowchart showing a shake mitigation operation of an apparatus according to a second exemplary embodiment of the present invention.

FIG. 3 is a diagram showing a panning function in the first embodiment of the present invention.

FIG. 4 is a block diagram showing a configuration of a conventional shake reduction device.

FIG. 5 is a flow chart showing a shake reducing operation of the conventional device.

6 is a diagram showing a panning function in the conventional apparatus of FIG.

[Explanation of symbols]

 11 Gyro as detection means 12 Operational amplification circuit as amplification means 13, 18 A / D conversion circuit 14, 16 D / A conversion circuit 17 Optical system as correction means 151 Integration means 152 Control means for panning detection 153 Gain switching Means 154 Predetermined gain means 156 Integrating means IC Integrated circuit including signal processing means VR12 Volume that is gain adjusting means

Claims (2)

[Claims] 1. A detection means for detecting a shake state of the device,
In a shake reduction apparatus including a signal processing unit that processes an output of the detection unit and a correction unit that operates according to an output of the signal processing unit, the signal processing unit performs all multiplication processes by a digital arithmetic shift operation. A shake mitigation device comprising means for performing 2. A detection means for detecting a shake state of the device,
Amplification means for amplifying the output of the detection means with a predetermined gain;
A / D conversion means for converting the output of the amplification means into a digital signal, signal processing means for changing the first value and the second value according to the output of the A / D conversion means, and the signal processing means Of the amplifying means according to the analog output signal of the D / A converting means. The correcting means operates according to the first value of the D / A converting means and the D / A converting means for converting the second value into an analog signal. A shake mitigation device configured to control a point, wherein the shake mitigation device is provided with gain adjusting means for adjusting a predetermined gain of the amplifying means.