JPH04134316A - Moving speed detection device for body and image blur correction device for camera - Google Patents

Abstract

PURPOSE:To accurately detect the moving speed of the body without being affected by a mechanical shake by providing a movement detecting means and a vibration detecting means for detecting the vibration of the movement detecting means, and calculating the moving speed of the body. CONSTITUTION:A photoelectric converting means 3, the movement detecting means 5, the vibration detecting means 14, and a speed arithmetic means 5 are provided. The movement detecting means 5 detects the movement of an image formed on the surface of the photoelectric converting means 3 through an optical system with a video signal outputted by the photoelectric converting means 3, the vibration detecting means 14 detects vibration due to a hand shake, and further the moving speed of the body itself is calculated from the detected value of the movement detecting means 5 and the detected value of the vibrating means 14. Consequently, even when the influence of the hand shake is large, the moving speed of the subject is accurately detected to correct an image blur with high accuracy.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to an object moving speed detection device and a camera image blur correction device.

[Prior Art] Conventionally, it is known that a camera uses an image blur correction device to prevent blurred photographs or video images from being taken. This wobbling occurs when the position on the surface of the photographic film where the subject image is formed changes during exposure or photography (hereinafter, this change is referred to as "image wobbling").

There are two causes of image blur: vibration of the photographer's hand supporting the camera (hereinafter, this vibration will be referred to as "camera shake") and movement of the subject itself.

Conventionally, image stabilization devices that correct image blur caused by camera shake by the photographer use a vibration sensor such as an acceleration sensor to detect movement (vibration) of the camera due to camera shake, and adjust the lens barrel according to the detected vibration. It is known to move the entire image sensor or the image sensor (for example, Japanese Patent Publication No. 1-53957). Furthermore, as an image blur correction device that corrects both image blur caused by camera shake and image blur caused by movement of the subject itself, it is known that an object movement speed detection device is used. In this method, the image of the subject is formed not on the photographic film but on an image sensor equipped with an object moving speed detection device immediately before shooting, and the moving speed of the object image is determined by the video signal output from this image sensor. This is to prevent image blurring.

[Problems to be Solved by the Invention] However, such conventional image blur correction devices for cameras have had the following problems.

Although image stabilization devices using vibration sensors are effective in preventing image blur caused by the photographer's camera shake as described above, they cannot prevent image blur caused by movement of the subject during exposure. .

Therefore, if the image blur is caused mainly by the movement of the subject, there is no point in using this image blur correction device.

In addition, since an image stabilization device using an object moving speed detection device detects relative position changes between the camera and the subject, the detected image blur information can be divided into two types: one caused by camera shake and the other caused by the movement of the subject. It is a composite of the two. However, the moving speed detection device for a moving object requires time required for the integration operation of each photoelectric conversion element constituting the image sensor and calculation processing to determine the state of image blur from the video signal output from the image sensor. Since this method takes a long time, it is suitable for detecting the moving speed of objects that rarely make random movements in a short period of time, but it is difficult to accurately detect camera shake. For this reason, conventional object movement speed detection devices cannot accurately detect the object movement speed when the influence of camera shake is large, and therefore, image blur using the object movement speed detection device In the correction device,
Image blur could not be prevented with high precision.

Furthermore, since the image sensor cannot detect a video signal during film exposure, such an image blur correction device cannot detect image blur that changes during exposure.

The present invention provides an object moving speed detection device that can accurately detect the moving speed of an object even when the influence of camera shake is large, and a camera image blur correction device that can perform highly accurate image blur correction. The purpose is to provide

[Means for Solving the Problems] The object moving speed detection device of the present invention includes a photoelectric conversion means for converting an image of an object into a video signal, and an optical system for forming an image of the object on the surface of the photoelectric conversion means. a movement detection means for detecting movement of the image of the object using the video signal inputted from the photoelectric conversion means; and a vibration detection means for detecting vibration of the movement detection means using a mechanical vibration sensor. and a speed calculation for calculating a moving speed of the object using a detected value of vibration of the movement detecting means inputted from the vibration detecting means and a detected value of movement of the image of the object inputted from the movement detecting means. It is equipped with the means.

Further, the image blur correction device for a camera according to the present invention includes a photoelectric conversion means for converting a subject image into a video signal, an optical system for forming the subject image on the surface of the photoelectric conversion means, and an input from the photoelectric conversion means. a movement detection means for detecting movement of the subject image using the video signal obtained by the camera; a vibration detection means for detecting vibration of the movement detection means using a mechanical vibration sensor; a speed calculating means for calculating a moving speed of the subject using a detected value of vibration of the detecting means and a detected value of movement of the subject image inputted from the movement detecting means; and a correction means for correcting image blur based on the moving speed of the camera.

[Function] The object moving speed detection device of the present invention detects the movement of the image formed on the surface of the photoelectric conversion means by the optical system using the movement detection means based on the video signal output from the photoelectric conversion means, and also detects the movement of the image formed on the surface of the photoelectric conversion means by the movement detection means. The vibration of the movement detection means is detected by the vibration detection means, and the moving speed of the object itself is calculated based on the detection value of the movement detection means and the detection value of the vibration detection means.

The image stabilization device for a camera according to the present invention detects movement of a subject image formed on the surface of a photoelectric conversion means by an optical system using a video signal output from the photoelectric conversion means, and also detects movement due to camera shake. The vibration of the means is detected by the vibration detection means, and the moving speed of the subject itself is calculated from the detection value of the movement detection means and the detection value of the vibration detection means, and the image blur is corrected by the correction means using this calculated value. Do this.

[Embodiment 1] Hereinafter, as an embodiment of the present invention, an image stabilization device for a camera of the present invention configured using the object moving speed detection device of the present invention is installed in a single-lens reflex camera having a focus detection function. This will be explained using an example.

FIG. 1 is a block diagram schematically showing the configuration of a camera according to this embodiment. As shown in the figure, this camera includes a photographic lens 1, a focus detection optical system 2, an area image sensor 3, an interface circuit 4, a CPU (central processing unit) 5, and a ROM (
read-only memory) 6, a lens ROM 7, a lens drive circuit 8, a lens drive motor 9, a slit 10, and a light emitting diode lla that constitutes a photointerrupter 11.
and photoresistor 11b1 display device 12, vibration isolator 13
, a mechanical shake detection unit 15 having an anti-vibration sensor 14, a switch SWI for starting tracking of a moving object, a release switch SW2, and the like.

The area image sensor 3 photoelectrically converts an image formed by the focus detection optical system 2 into an electric signal (video signal). The interface circuit 4 also drives the area image sensor 3 and
The electrical signal (analog signal) from is converted into a digital signal and output to the CPU 5. The CPU 5 is in charge of controlling the operation of the entire camera, and includes, for example, calculating a subject distance based on the output of the area image sensor 3, display control such as focusing and defocusing on the display device 12, and image stabilization control. It is now possible to do this.

The ROM 6 stores the amount of deviation (interval between two images) at the time of focusing in a plurality of areas where focus detection is performed. Also, lens RO
M7 is provided in the lens barrel and stores various data necessary for focus detection, such as a conversion coefficient for determining the amount of defocus from the F number of the lens and the amount of image shift. Furthermore, the lens drive circuit 8 drives the motor 9 under the control of the CPU 5, thereby driving the photographing lens 1.
It moves the position of.

Normally, in AF control in which a subject distance is detected and the photographing lens 1 is driven based on the distance information, it is necessary to feed back the driving amount of the lens 1 to the CPU 5. In this case, it is common practice to substitute the actual movement amount of the lens 1 with the number of rotations of the driving motors. For this reason, here, the number of slits 10 is counted by a photointerrupter 11 to obtain the number.

That is, when the lens drive circuit 8 operates and the motor 9 is rotated, the slits 10 provided at equal intervals in the rotating member of the lens barrel are rotated. Then, this slit 10
passes between the light emitting diode 11a and the phototransistor llb, which are arranged opposite to each other, and the number of rotations thereof is counted by the CPU 5. Thereafter, when the count reaches a predetermined value, the motor 9 is controlled to stop rotating. The mechanical shake detection unit 15 includes a vibration sensor 14 such as an angular velocity sensor, and detects the amount of shake on the image plane from the vibration of the camera.

FIG. 2 is an optical layout diagram showing the detailed configuration of the focus detection optical system 2 shown in FIG. 1 above. This focus detection optical system 2
, a quick return mirror 2a movably provided in the direction of arrow A1, a prism 2b, a field lens 2c, an aperture diaphragm 2d1 and optical axes are provided by a prism 2d2.
The visual field image transmission optical system 2d consists of the following.

The quick return mirror 2a transfers the light flux (image information) from the photographing lens 1 to the prism 2b during AF operation.
, and in the direction of the film surface 2e during photographing.

Further, the visual field image transmission optical system 2d divides the guided light beam into two by pupil division, and displays the two on the area image sensor 3.
Images Ia and Ib are formed.

Here, the principle of focus detection of the camera according to this embodiment will be explained. In addition, here, two images Ia by pupil division
, Ib is used to determine the subject distance. That is, the light flux that has passed through the photographing lens 1 passes through the field lens 2c, and is then split by the visual field image transmission optical system 2d, and is formed on the area image sensor 3 as images Ia and Ib, respectively.

In this case, for example, when focusing (just in focus), the third
As shown in figure (a), regardless of the subject distance, the two images Ia,
Ib is imaged at a certain interval fa. Furthermore, in contrast to this time of just focus, when the front focus is in front of the subject, the distance fb between the two images 1a and Ib is greater than the distance fa during just focus, as shown in Figure 3(b). It gets narrower. Furthermore, when the subject is in focus after the subject, the two images I
The distance fc between a and Ib is wider than the distance fa when in just focus. Therefore, by calculating the interval between these two images Ia and Ib, the distance to the subject can be obtained. The above is the principle of focus detection using the phase difference method.

Next, a method for detecting movement of a subject image from a video signal will be described. Detection of such movement of the subject image is performed using only a part of the light receiving surface of the area image sensor 3 shown in FIG. 1 above. FIG. 4 is a conceptual diagram showing an example of a range of the light receiving surface of the area image sensor 3 used for detecting movement of a subject. In the figure, the area indicated by A is an area for forming a subject image 1a, and the area indicated by B is an area for forming a subject image 1b. Note that FIG. 4 shows a case where the subject image is formed near the center on the area image sensor 3. Here, as shown in Fig. 5, if the coordinate axis in the long side direction of the film surface is the y axis, the coordinate axis in the short side direction is the y axis, and the coordinate axes in the optical axis direction are two axes, the area image sensor corresponding to this The X-axis direction and the y-axis direction of No. 3 are as shown in FIG. Further, FIGS. 6(a) and 6(b) show the pixel configurations of area A and area B shown in FIG. 4. Here, FIG. 6(a) shows the pixel configuration of area A shown in FIG. 4, and FIG. 6(b) similarly shows the pixel configuration of area B. Hereinafter, as shown in FIGS. 6(a) and 6(b), the pixel output of area A will be expressed as anm, and the pixel output of area B will be expressed as bnm. Also, the sum of the pixel outputs of the first row to the area at time t is S A (xi, t) - Σ a i j
−<1) No.1. Note that rtJ indicates time, and rxiJ indicates the first row (X indicates the addition direction). Similarly, the sum of pixel outputs in the fifth column of area A at time t is S A (yj, t) - Σ aij (2), and further, the sum of pixel outputs in the first column of area B at time t is Let the sum of

FIGS. 7(a) and 7(b) are flowcharts for explaining the procedure for determining the moving speed of the subject in the X-axis direction, the y-axis direction, and the two-axis directions. Hereinafter, explanation will be given according to this flowchart.

First, the photographer decides on a composition so that the subject image is formed near the center of the screen (ST701).

When the composition is determined, the CPU 5 performs an integral operation of the area image sensor 3 and stores the start time t1 (ST
702). Subsequently, the above equations (1) to (3) are calculated and the calculation results are stored (ST703.5T704).

Thereafter, the area image sensor 3 performs an integral operation for time t2 (ST705), and then calculates the above equations (1) to (3) (ST706.
5T7C17).

Next, using the above calculation data, the moving speeds of the subject in the X-axis direction, the y-axis direction, and the X-axis direction are determined (ST708 to 5T712). Here, when the subject is moving in two axes, the subject distance changes, so the two images 1 on the area image sensor 3
Image spacing of a and Ib (see Figures 3(a) to 3(c))
changes, the image 1a (image of area A) shifts in the minus direction of the X-axis, and the image 1b (image of area B) shifts in the plus direction of the X-axis. On the other hand, the phase shift in the y-axis direction is not caused by the movement of the subject in the two-axis directions. On the other hand, for the movement of the subject in the X-axis direction and the y-axis direction, two images 1a, Ib
make the same amount of positional change in the same direction of the X-axis and y-axis, respectively. Therefore, the change in the y-axis direction of the two images 1a and Ib corresponds only to the movement of the subject in the y-axis direction, whereas the change in the X-axis direction corresponds to the change due to the movement of the subject in the It is a combination of changes due to movement of . Therefore, the moving speed of the subject in the y-axis direction can be easily determined by calculating the moving amount at time t2-t for either image 1a or image 1b, whereas The moving speed in the direction is the image Ia
It cannot be determined unless both the amount of movement of the image Ib and the amount of movement of the image Ib are considered. A method for determining the moving speed of a subject in each direction will be described below.

First, find the moving speed of the subject in the y-axis direction (S
T708). Among the pixels constituting area A, the range (reference image) used to detect the movement of the subject is divided from the 1st row to the 9th row (i.e., i- to -), as shown in FIG. 8(a).
41), the pixel output at time t1 and time t2
The correlation calculation output with the pixel output in is FAX (S) - Σ l 5A (X+, t+) 5A
It is given by (X+++s+, 12) l - (4). Here, the value of S that gives the minimum value of FAx(S) is 81
Then, Sl is the pixel output at time tl and time t2
represents the amount of temporal shift of the image that best matches the pixel output at .

That is, as shown in FIG. 9, this indicates that the image has moved by a distance S1 in the X-axis direction between time t1 and time t2. Therefore, if the imaging magnification of the focus detection optical system 2 is β, the speed vy of the subject in the y-axis direction is given by:

Next, find the moving speed of the subject in the X-axis direction (S
T709). Of each pixel constituting area A and area B, the range (reference image) used for detecting the movement of the subject is defined from the 'th row to the g'th row (
That is, i- is set to '~1'). Regarding area A,
The correlation calculation output between the pixel output at time t1 and the pixel output at time t2 is FBY(S) −: fl SB (y+, t+) −
3B(y++, s+, t2) is given by 1-(6). Here, if the value of S giving the minimum value of FAY(S) is 82, S2 represents the amount of image shift at which the pixel output at time t1 and the pixel output at time t2 most closely match. As shown in Fig. 10, this shift amount S2 is the sum of the image shift Δx1 caused by the subject moving in the X-axis direction and the image shift Δx2 caused by the subject moving in the two-axis direction. be. That is, the value of S2 is ΔX1
−Δx2. On the other hand, the correlation calculation output between the pixel output at time t1 and the pixel output at time t2 for area B is FBY(S)-Σ l5B(y+, t+) 5B(
It is given by yt++s+, t2) -(7).

Assuming that the value of S that gives the minimum value of FBY(S) is 83,
As shown in FIG. 10, this shift amount S3 is the sum of the image shift ΔX1 caused by the subject moving in the X-axis direction and the image shift ΔX2 caused by the subject moving in the two-axis direction. . That is, the value of S3 becomes ΔX1+Δx2. Therefore, if the imaging magnification of the focus detection optical system 2 is β as described above, the speed Vx of the subject in the X-axis direction is given by.

Next, find the moving speed of the subject in the Z-axis direction (
Sr110-3T712). In this example, first, the two-image interval f1 at time t1 and the two-image interval f2 at time t2 are determined (ST710ST711), and the determined fl and f
2 is used to find the moving speed Vz of the image in the Z-axis direction. Note that since a highly accurate calculation is required for the subject distance, the above-mentioned pixel addition output S A (yj,
t).

The calculation is performed by reducing the number of pixels to be added without using S B (yj, t). Here, only one row of pixels is used, and the row used is the second row (kh x g).

Pixel output a b+' of the second row of area A at time t1
= a be, abk' ~ah, - is a reference image, and this reference image and the pixel output of area B bhk・~bh,
・The amount of deviation between
ni' +52) '++ a (h+511(I'
+521 sb lll'sl++ 翫'+S21
~b Lh4SI) Perform the correlation calculation using l'·32+. Here, if the value of f that gives the minimum value of H+(Ω) is fl, then fl is 2 at time t1.
It represents the image spacing. Note that by performing an interpolation calculation on the two-image interval f1 determined by this correlation calculation, a more accurate two-image interval can be determined.

Next, find the two-image interval f2 at time t2 (Sr1
11). At time t2, since the image is moving on the area image sensor 3, the same correlation calculation for pixels bb1 to b□· as at time 1 cannot be performed.

When the image of area A is used as a reference image, the image is scanned in the X-axis direction and in the X-axis direction of this image. Here, if the value of f that gives the minimum value of H2(R) is f2, then f2 represents the interval between the two images at time t2.

Next, Sr110-calculated two-image interval f1 and 5T71
The moving speed Vz of the subject in the two-axis directions is determined based on the two-image interval f2 determined in step 1 (ST712). VZ is given by α as a coefficient.

Next, it is determined whether Vx, Vy, and Vz are within a range that can be controlled by blur prevention drive or moving object prediction drive, which will be described later. First, Vx, Vy, and Vz are compared with predetermined upper limit values C,, C2, and C3 (Sr113 to ST7
15), Vx, Vy.

If any of Vz exceeds the upper limit value, a release prohibition flag is set (ST 720).

Next, it is determined whether the accelerations ΔVX, Δ■y, and ΔVz of the subject in the X-axis direction, y-axis direction, and Z-axis direction are within the range that can be controlled by the blur prevention drive and moving object prediction drive described later. Let's do it. This flowchart is executed repeatedly, so Vx, Vy, and Vz are calculated repeatedly, so
Vx, Vy, Vz determined by two consecutive runs
By calculating the amount of change per unit time in Vx, Vy, and Vz from each calculated value, the accelerations ΔVx and ΔVy are determined.

ΔVz can be obtained. Compare these with the predetermined upper limit values C4, C5, and C6.
T718), if either exceeds the upper limit, a release prohibition flag is set (ST720).

Finally, the tracking range is changed so that the subject can be tracked the next time the flowchart is executed (Sr121).

Here, if the next integration start time is t3, the tracking range of area A is a position shifted from the y-axis direction only in the X-axis direction. In addition, the tracking range of area B is
The position is shifted by the number of pixels given by equation (12) in the y-axis direction, and shifted by the number of pixels given by equation (12) in the X-axis direction.

Note that in this example, the moving speed of the subject in the X-axis direction and the y-axis direction was determined using the output of the area sensor for phase difference AF, but it can be easily calculated using an ordinary non-eccentric imager. It is also possible to determine the moving speed of the subject.

Next, the vibration isolator 13 shown in FIG. 1 will be explained in detail.

11 and 12 schematically show the structure of such a vibration isolator 13, with FIG. 11 being a sectional view and FIG.
The figure is a perspective view. In both figures, 101 is the main body of the camera, and 102 is a photographic lens. The photographic lens 102 is
Photographing optical system 103, first acceleration sensor 119a for detecting vibration in the rotational direction of the y-axis shown in FIG.
.

A second acceleration sensor 119b is used to detect vibrations in the rotational direction of the X-axis shown in FIG. A correction optical element 108, a second correction optical element 109 configured to be rotatable about the axis 111a and for correcting vibrations in the xz plane, and a disk-type ultrasonic device for driving the first correction optical element 108. A sonic motor 112, a disk-type ultrasonic motor 113 for driving the second correction optical element 109, an encoder 114 for detecting the rotational state of the first correction optical element 108,
An encoder 115 for detecting the rotational state of the second correction optical element 109, an electric circuit section 120 for driving the disk-type ultrasonic motors 112 and 113, and exchange of signals between the camera body 101 and the photographing lens 102. It is constituted by a photographing lens side contact point 116 for performing this. The encoder 114 includes a drum 114a that rotates in accordance with the rotation of the first correction optical element 108, a magnetic sensor (such as a magnetic head, hereinafter referred to as an MR sensor) 114b for detecting the rotational state of the drum 114a,
An optical sensor (not shown, such as a photoreflector, hereinafter referred to as PR) is used to detect the end of rotation of the drum 114a.
(referred to as a sensor). Similarly, the encoder 115 includes a drum 115a that rotates in accordance with the rotation of the second correction optical element 109, an MR sensor 11.5b for detecting the rotational state of the drum 115a,
It is composed of a PR sensor (not shown) for detecting the end of rotation of the drum 115a. On the other hand, the main body 101 of the camera includes a quick return mirror 104, a main body optical system 106 consisting of a focus detection optical system 2 and a finder optical system, an eyepiece 107, and an electric circuit section for performing signal processing, calculation of camera shake amount, etc. 118, and a camera body side contact 117 for exchanging signals between the camera body 101 and the photographing lens 102.

Further, the camera body 101 is equipped with an X-axis angular velocity detection section 119a and an X-axis angular velocity detection section 119b in order to detect the amount of mechanical shake of the camera. Here, the X-axis direction and the y-axis direction of the camera are determined as shown in FIG. The X-axis angular velocity detection unit 119a is for detecting the angular velocity of camera shake in the rotational direction about the y-axis. This rotational blur around the y-axis causes image blur on the film 105 in the x-axis direction. Further, the X-axis angular velocity detection unit 119a is for detecting the angular velocity of the camera shake in the rotational direction around the X-axis. This causes image blur. The shake information detected by the X-axis angular velocity detection section 119a and the X-axis angular velocity detection section 1]9b is transmitted to the electric circuit section 118 of the camera body 10].
The data is transferred to the CPU 5 provided within the computer.

Next, the electric circuit system of the vibration isolator 13 will be explained using FIG. 14. In FIG. 14, components designated by the same reference numerals as in FIGS. 11 and 12 are the same as in FIGS. 11 and 12, respectively. Further, 7 is a lens ROM 137 that stores information regarding the photographic lens.
138 is a drive control unit for driving the disk-type ultrasonic motor 112; 138 is a drive control unit for driving the disk-type ultrasonic motor 113; and 139 is a storage unit for storing various programs.

The subject velocities Vx and vy based on the video signals obtained by the above equations (5) and (8) and the X-axis angular velocity detection section 1
The blur information detected by the X-axis angular velocity detection section 19a and the X-axis angular velocity detection section 119b is taken into the CPU 5. CPU5 is
Based on the captured angular acceleration information and various constants and programs stored in the lens memory 136 and storage unit 139, the angular velocity of the correction optical elements 108 and 109 necessary to cancel the movement of the image is calculated. The obtained angular velocity is sent to drive control units 137 and 138. Drive control units 137 and 138 drive disk-type ultrasonic motors 112 and 113 based on the acquired angular velocity to correct image blur.

Next, the encoders 114 and 115 will be explained. Figures 15(a) to (C) are diagrams showing the configuration of the encoder 1]4, with Figure 15(a) being a top view and Figure 15(b) being a top view.
is a perspective view, and FIG. 15(c) is a side view.

A magnetic pattern is magnetized on the outer peripheral portion 114a' of the drum 114a, which rotates in accordance with the rotation of the first correction optical element 108. This magnetic pattern is the MR sensor 114b.
The amount and direction of rotation of the disk-type ultrasonic motor 112 are detected. Furthermore, the rotation speed is also detected from this value. In addition, the outer peripheral portion 11 of the drum 114a
Patterns with different reflectances of infrared light are attached to 4a, and are read by PR sensors 114c and 114d. These are used as limit switches to detect the rotational limits of the first corrective optical element 108 and prevent it from colliding with mechanical ends. Figure 16(a)
And FIG. 16(b) shows PR sensors 114c and 114d.
FIG. When the drum 114a rotates to the left, the pattern of the drum outer periphery 114a'' changes from a bright area (the area shown in white in the figure) to a dark area (the area shown in black in the figure), and the PR sensor 114C rotates to the left. A signal indicating the limit is outputted, and when the PR sensor 114C rotates clockwise, the dark area changes to a bright area, and the PR sensor 114C outputs a signal indicating the limit of clockwise rotation.

The rotation signal and direction signal output by the MR sensor 114b are
Disk type ultrasonic motor per unit time 112.113
is converted as a rotation amount signal, and the drive control unit 137 (first
4) is used as a speed control feedback signal for the disc type ultrasonic motors 112 and 113 to control the speed of the disc type ultrasonic motors 112 and 113, or as a displacement signal from a certain reference weight position for position control. Ru.

Furthermore, the sequence control section within the CPU 5 also monitors the rotation signal and direction signal of the disk-type ultrasonic motors 112, 113 output by the MR sensor 114b, and uses the signals to correct the angular velocity signal for driving. When the sequence control unit in the CPU 5 monitors the rotation signal, the interval between the magnetic patterns on the outer circumferential portion 114a' of the drum 114a is determined so that the signal interval will not be monitored, or the latch section is It is desirable to prevent the rotation signal from being overlooked even if the operation of the monitor is delayed due to the interference.

For example, the spacing of the magnetic patterns on the outer circumference 114a- of the drum 114a is such that even when the disk-type ultrasonic motors 112, 11B rotate at the fastest speed, rotation signals are output at a rate of 1 signal or less every 1/1.024 seconds. , and hold that signal in a latch circuit. The sequence control section monitors the rotation signal every 1/1024 seconds,
If there is a signal, it is understood that a rotation signal was output within the past 1/1024 seconds, and the latch circuit is cleared. The sequence control section can grasp where the disk type ultrasonic motors 112 and 113 are located by integrating the rotation signal and the direction signal using an up/down counter or the like. The reference position for integration is created by the PR sensors 114c and 114d.

Here, the right rotation limit of the drive range of the disk-type ultrasonic motors 112 and 113 is taken as the reference position. Therefore, when the disk-type ultrasonic motors 112, 113 reach their right rotation limit, the drive is stopped and a counter or register indicating the position of the disk-type ultrasonic motors 112, 113 is cleared to "0". On the other hand, when the disc-type ultrasonic motor 112° 113 reaches its left rotation limit,
At the same time as stopping the drive, the counters or registers indicating the positions of the disk type ultrasonic motors 112 and 113 are corrected to predetermined values. By doing so, the positions of the disk-type ultrasonic motors 112 and 113 can be more accurately grasped.

Regarding the image blur caused by the mechanical blur of the camera, if a rotation of θX occurs in the X-axis direction (13th
(see figure) will be explained as an example.

FIG. 17 is a schematic cross-sectional view showing the relationship between the photographing optical system 103 and the film 105, as viewed from the positive side of the X-axis. Note that the X axis serving as the center of rotation is located at the center of the film. The amount of image blur Δyc that occurs when a rotation of θ occurs in the X-axis direction is given by Δyc−−f−(1+β)2tanθX, where β is the imaging magnification and f[mm] is the focal length. Here, assuming that there is infinite light, that is, the photographing magnification β is zero, then Δyc−fatanθx−(15). Since the imaging magnification is small in general photography, the imaging magnification β may be considered to be approximately zero in this way. Also,
If the amount of movement of the image is sufficiently small compared to the subject distance, equation (15) can be approximated by the following equation (16). In addition, if necessary, Δyc
−−f ・(1+β)2 ・θx =(17
) may be approximated as

Here, considering the time differentiation on both sides of approximate equation (16), d(Δyc)/dt −−f・(dθx/dt) ・(18
) holds true. Here, dθx/dt is the angular velocity, and d(Δyc)/dt is the moving speed Vcy of the image when the angular velocity of θ is generated.

Furthermore, even when a rotation of θY occurs in the y-axis direction, the moving speed of the image can be determined in the same manner.

FIG. 18 is a schematic cross-sectional view showing the relationship between the photographing optical system 103 and the film 105, as viewed from the positive side of the y-axis. The amount of image blurring Δxc that occurs when rotation of θY occurs in the y-axis direction is Δxc−f・(1+β
)2tanθ7, therefore dθY/dt
The image moving speed Vcx when an angular velocity of Vcx occurs is Vc
x=d(Δxc)/dt-f・(dθY/dt)...(1
9) It is expressed as

Next, the mechanical shake detection section 15 shown in FIG. 1 will be explained. FIG. 19 is a block diagram schematically showing the configuration of the mechanical shake detection section 15. As shown in the figure, the mechanism shake detection unit 15 includes an X-axis angular velocity detection unit 210 that detects angular velocity around the X-axis, an A focal length information output unit 212 that outputs information regarding focal length, and a subject distance detection unit 213 that detects and outputs information about subject distance.
and mechanical blur image movement calculating sections 2 and 4 that calculate the moving speed of the image based on the output information of each of these sections.

Regarding the operation sequence of the mechanical shake detection unit 15, the 20th
This will be explained using figures. First, the focal length f, the lens position,
Read out of focus 5T2001ST2002, 5T
2003), then the subject distance g is calculated as explained using FIG. 3 (ST2004). Subsequently, the imaging magnification β is calculated (ST2005). The imaging magnification β can be calculated by a known method using the focal length f and the subject distance Ω. In addition, calculations can be simplified by using approximate expressions or the like according to requirements for accuracy. It should be noted that if the photographing magnification is determined by approximate formula calculation without using the subject distance, the subject distance detecting section 213 shown in FIG. 19 is not necessary. Also,
In the case of a fixed focus, the focal length information output unit 212 is also not required by performing calculations using fixed values.

After that, Step 5T2006 to Step 5T2011
Calculate the mechanical deflection. , X-axis angular velocity detection section 11
9b to detect the angular velocity dθX/dt around the X axis 5T2
(ST2
007) and outputs the calculation result (ST2008). Next, the y-axis angular velocity detection unit 119a detects the angular velocity dθ around the y-axis.
To detect Y/dt 5T2009) This detected value dθ
Using Y/dt, the image movement speed Vcx is determined by the above equation (19) (ST2010), and finally, the calculation result is output (ST2011), and the sequence ends.

In this example, the image movement information obtained from the video signal obtained by the above equations (5), (8) and (11) and the above equation (1) are used.
8) and the image movement information obtained from the mechanical shake detection unit 15 obtained in steps (19) are used to detect information regarding the movement of the subject.

The image movement information obtained from the video signal is information about the relative movement between the subject and the camera (area image sensor 3 of the camera), so the image movement information obtained from the image signal includes the image movement component due to the movement of the subject and the image due to mechanical blur of the camera. It is a composite of the movement component.

Therefore, by finding the difference between the image movement speed obtained from the video signal and the image movement speed obtained from the mechanical shake detection section 15, it is possible to calculate the image movement speed due to the movement of the subject. That is, if the image movement speed in the X-axis direction due to the movement of the subject is Vox, and the movement speed in the y-axis direction is Voy, then the image movement speed in the X-axis direction and the y-axis direction due to the video signal is VX.

Since the image movement speeds in the X-axis direction and y-axis direction due to mechanical blur of the Vy1 camera are Vcx and Vcy, Vox-V
x-Vex --(20)Vo y=V
y-Vc y-(21) holds true.

FIG. 21 shows the image movement speed Vox, V due to the movement of the subject.
The procedure for finding oy is shown below. First, the above equations (5), (8), and (11) grin, Vx, and Vy are calculated using a video signal (ST2101).

Next, the y-axis angular velocity detection unit 119a and the X-axis angular velocity detection unit 119b detect the angular velocity dθ, /dt and X
Detect the angular velocity dθx/dt around the axis (ST2102
) Next, detect the focal length f of the photographic lens 102. 5T2103), dθY/dt, dθx/dt, f
The image movement speed Vcx due to the mechanical shake of the camera is calculated using
, Vcy are calculated (ST2104). Finally, using equations (20) and (21) above, Vox.

Find Voy (ST2105).

In this way, prior to photographing, the image movement speeds Vox and Voy due to the movement of the subject and the image movement speeds Vcx and Vcy due to mechanical shake are detected, and the image movement speeds Vcx and Vcy are detected using the image stabilization device 13 using Vcx and Vcy. When correcting camera shake, by taking into account the image movement speeds Vox and Voy caused by the movement of the subject, it is possible to correct the mechanical shake of the camera as well as the image blur caused by the movement of the subject. It becomes possible.

FIG. 22(a) shows the image movement speed Vox due to the movement of the subject according to the procedure shown in FIG.

5 is a timing chart showing the execution timing of each process when calculating Voy. In the figure, t.

is the distance measurement start time (the first integration start time of the area image sensor 3), tl is the first readout timing of the video signal, and tl is the second readout timing of the video signal. Further, Δt1 is the integration time of the first integration of the area image sensor 3, and Δt2 is the integration time of the second integration. The calculation of the image movement speeds Vx and Vy determined from the video signal is started at the same time as the second reading of the video signal.

When the calculation of Vx and Vy is completed, the image movement speed Vcx and Vcy due to the mechanical shake of the camera are detected (angular velocity dθx
Measurement of /dt, dθ, /dt and calculation of Vcx and Vcy are started (start timing t3). Vcx, V
When the detection of cy is completed, Vox and Voy are finally obtained using the above-mentioned equations (20) and (21) (start timing 14).

Here, the video signal obtained by the first reading of the video signal is at time t. It can be considered that it indicates the average position of the subject from time t1 to time t1 (Δ1.). Similarly, the video signal obtained by the second readout can be considered to be the average of the subject positions during Δt2. Therefore, the image movement speeds Vx and Vy obtained from these video signals can be considered to be the average of the image movement speeds in the X-axis direction and the y-axis direction from to to tl. By taking this into consideration when detecting the image movement speed due to the mechanical shake of the camera, the image movement speed Vox, Voy due to the movement of the subject can be
It is possible to improve the accuracy when determining . FIG. 22(b) shows an example of a timing chart when calculating the image movement speeds Vox and VoY due to the movement of the subject by taking such consideration into account. In FIG. 22(b), the image movement speed Vex due to the mechanical shake of the camera between to and tl is
, Vcy are detected multiple times, and then at timing t
In step 3, the average value of the obtained image movement velocities Vex and Vcy is calculated. Note that in this case, the detection of Vcx and the detection of Vcy do not necessarily need to be performed at the same time, and it is sufficient that the average value of each can be determined. For example, both can be detected alternately, thereby reducing the time required for one measurement.

Next, a description will be given of image stabilization control for preventing image blur during photographic exposure.

Object movement generally does not occur at such high frequencies. Particularly, when taking panning shots, the movement of the image of the subject occurs at approximately constant speed. Therefore, no problem will arise in terms of accuracy even if image stabilization control during exposure is basically performed using the moving speed of the subject image detected immediately before the start of exposure. on the other hand,
In this embodiment, mechanical blur of the camera due to vibrations of the human body, etc. is detected, image movement (image blur) due to this blur is calculated, and anti-shake control is performed so that this image blur can also be corrected. However, this image blur detection can be performed even during exposure. That is, in this embodiment, the image movement speed Vox, V
Image stabilization control is performed in real time so as to correct during exposure the blur that is the sum of oy and image movement speeds Vcx and Vcy due to mechanical blur of the camera that are sequentially detected during exposure.

FIG. 23 shows a schematic flow of vibration isolation control.

First, the mechanical shake is detected as shown in FIG. 20 above, and the image movement speed V due to the mechanical shake of the camera is
cx and Vcy are determined (ST2301). Next, the image movement speeds Vox and Voy due to the movement of the subject detected as shown in FIG. 21 described above are read out (ST2302
), continued, VCX+VOX (7) calculation, Vc y
+Voy calculation (ST2303, 5T2304
), Vcx+Vox and Vcy+Voy are image blurs that should be prevented in real time during exposure. Then, using these data, perform shake correction (ST2
305). As described above, in this embodiment, values immediately before exposure are used as expected values for Vox and Voy, which cannot be detected during exposure, and detected values during exposure are used for Vcx and Vcy, which can be sequentially detected even during exposure. By using , the accuracy of correction is improved. Note that this series of operations is repeated during exposure.

Next, the operation of the single-lens reflex camera according to this embodiment will be explained according to the flowcharts shown in FIGS. 24(a) and 24(b). Figure 24 (a
) and FIG. 24(b) are flowcharts showing the operation sequence of the CPU 5 in this embodiment.

When the power is turned on, the C
Reset PU5, initial settings, check each electric circuit,
Perform resets of various mechanical parts as necessary (ST2
401). Next, execute the photometry subroutine (ST2
402). This photometry subroutine is a subroutine that measures the brightness of a subject and calculates an aperture value and shutter speed that will provide appropriate exposure. Subsequently, a key manual subroutine is executed (ST2403).

This key manual subroutine is a subroutine that reads the states of various switches on the camera and performs mode settings, etc. according to the read switch states.

After that, the AF mode is set to single AF mode or continuous AF mode. First, single AF
Please check the on/off switch 5T2404)
, if it is on, set the SAF flag to "1" and perform AF.
Set the mote to single AF mote (ST2405
). On the other hand, if the single AF switch is off, SAF
The flag is cleared to "0" and the AF mode is set to continuous AF mode (ST2406). Here,
Single AF mode is a mode that prohibits automatic focus adjustment by autofocus (AF) after focusing once, and continuous AF mode prohibits automatic focus adjustment by AF even after focusing once. This is the mode in which it does not.

Next, a blue routine to be displayed is executed (ST2407).

In this display blue routine, the camera's operating mode and exposure value are displayed. After that, check the status of switch 1 (5T2408), and if it is off, step 5T2
Execute steps 402 and subsequent steps again. On the other hand 1. If switch 1 is on, then check the state of the first release switch (ST2409).If the first release switch is off, step 5T24
Execute steps 02 and later again. On the other hand, if the first release switch is on, then a distance measurement subroutine is executed (ST2410). This distance measurement subroutine measures the focus state of the subject and the amount of out-of-focus.

Next, execute the AF drive subroutine (ST2411
). In this AF drive subroutine, a judgment is made as to whether or not the subject is in focus using the focus state and the amount of focus deviation of the subject measured in the distance measurement subroutine, and if it is in focus, it is determined that the subject is in focus. A flag indicating that is set is set, and a display to that effect is displayed. On the other hand, when the camera is not in focus, different processing is performed depending on whether distance measurement is not possible. When distance measurement is not possible, a display and action to that effect are performed. In addition, when distance measurement is possible and the focus is simply blurred, the drive amount of the focus detection optical system 2 is calculated from the amount of focus shift, and the focus detection optical system 2 is driven based on the result. bring to a state of focus.

After completing the AF drive subroutine, check the state of the first release switch again (S
T2412). Here, if the first release switch is off, steps 5T2402 and subsequent steps are executed again. On the other hand, if the first release switch is on,
Check the state of the SAF flag (ST2413).

When the SAF flag is "1", that is, when the mode is AF mode or single AF mode, next, it is determined whether or not it is in focus (ST2412).If it is not in focus, step 5T2410 and subsequent steps are performed. Execute again. On the other hand, when the camera is in focus, the VOCAL subroutine (see FIG. 21), which is a subroutine for determining the image movement speeds Vox, Voy, and Voz due to the movement of the subject described above, is executed (ST2413). After that, check the condition of the first release switch (5T2416).
If it is off, step 5T2402 and subsequent steps are executed again. On the other hand, if the first release switch is on, check whether the release is prohibited or not.
T2417), if release is prohibited, step 5T
Execute steps 2415 and later again. On the other hand, if the release is not prohibited, then check the state of the release switch (5T2418), and if it is off, step 5T24).
Execute steps 15 and later again. On the other hand, if the release switch is on, steps 5T2422 and subsequent steps are executed.

On the other hand, in step 5T2413, the SAF flag is set to "
0'', that is, when the mode is AF mode or continuous AF mode, next, the VOCAL subroutine (see FIG. 21) is executed (ST2419). Next, check whether the release is prohibited or not.
T2420), if release is prohibited, step 5T
Execute steps 2410 and later again. On the other hand, if the release is not prohibited, then check the state of the release switch (5T2421); if it is off, step 5T2
Execute steps 410 and later again. Furthermore, if the release switch is on, steps 5T2422 and subsequent steps are executed.

Step 5T2422 and subsequent steps are an exposure sequence.

First, a moving object prediction drive subroutine is executed (ST24
22). The moving object predictive drive subroutine is a subroutine that predicts the amount of movement of the object during the release time lag from the moving speed Vz of the object image in the optical axis direction, and drives the focus detection optical system 2 so that the object is in focus during film exposure. be.

Next, start driving the vibration isolator 13 (ST2424)
. After that, exposure is started, and the mirror is raised, the aperture is driven, the shutter is controlled, etc. (ST2424)
. Next, the vibration isolation control subroutine is executed (ST24).
25). As described above (see FIG. 23), this anti-shake control subroutine uses the image movement speed Vo due to the movement of the subject determined in the vOCAL subroutine (see FIG.
Using x, Voy, Voz and image movement speeds Vcx and Vcy due to mechanical camera shake, first, image blur due to mechanical shake of the camera is corrected, and then image blur due to movement of the subject is also corrected. It is something. This subroutine is repeated until it is determined in step 5T2426 that the exposure has ended, thereby preventing image blurring during exposure. When the exposure is finished, first execute the anti-vibration reset subroutine to return the anti-vibration device to its initial state5.
T2427), then executes the exposure mechanism reset subroutine to perform operations necessary for the next shooting such as lowering the mirror, charging the shutter, resetting the aperture to open the aperture, and advancing the film (ST2428), and then returns to step 5T2402.
Return to

Figure 25 shows the CPU 5 when the camera is in continuous shooting mode.
2 is a flowchart showing the basic operation of FIG.

First, the first integration of the area image sensor 3 is performed, and addition is performed in the X-axis direction and the y-axis direction for the area where moving object tracking is performed (ST2501). Next, perform the second integration of the area image sensor 3, and similarly,
Perform addition in the X-axis direction and y-axis direction (ST2502
). Subsequently, a correlation calculation is performed in the same manner as explained in FIGS. 7(a) and 7(b) above (ST25).
03) Image movement speed Vx from the video signal.

Vy and Vz are determined (ST2504), and then image movement speeds Vcx and Vcy due to camera mechanical shake are determined in the same manner as described in FIG. 20 (ST2505).

Furthermore, according to equations (20) and (21), the image movement speed Vox due to the movement of the subject.

Voy is determined (ST2506). Thereafter, the release inch is checked (ST2507). If it is off, step 5T2501 and subsequent steps are executed again.

On the other hand, if the release button is on, moving object predictive driving and anti-shake control are performed (ST2508.5T2509), and then exposure control is performed (ST2510) in the same manner as described in FIG. Then step 5T2
By returning to step 501 and repeating the same operation, it is possible to perform photography using the continuous shooting mode.

Figure 26 shows the CPU 5 when the camera is in continuous shooting mode.
12 is a flowchart showing another example of the basic operation of FIG. The flowchart shown in FIG. 26 differs from the flowchart shown in FIG. 25 in that the process returns to step 5T2502 after performing exposure control. In this case, the image movement speed V
The calculation of xVy and Vz (ST2504) may be performed using the pixel output from the second integration (ST2502) in the previous execution. In this way, the number of integral operations between releases can be reduced by one, thereby enabling high-speed continuous shooting.

Figure 27 shows the CPU5 when the camera is in continuous shooting mode.
12 is a flowchart showing a third example of the basic operation of FIG. The flowchart shown in FIG. 27 differs from the flowcharts shown in FIGS. 25 and 26 in that steps 5T2501 to 5T2506 are not executed from the second time onwards.

In this case, the first detected value is used as the image movement speed. In this way, even higher speed continuous shooting becomes possible.

[Effects of the Invention] As explained in detail above, according to the object moving speed detection device of the present invention, the object movement speed detection device of the present invention can detect
It becomes possible to accurately detect the moving speed of an object. Further, according to the image stabilization device for a camera of the present invention, even when image blur occurs due to both image blur due to camera shake and image blur due to movement of the subject, high accuracy can be achieved. This makes it possible to perform image blur correction, which is particularly effective when taking panning shots of moving objects.

[Brief explanation of drawings]

FIG. 1 is a block configuration diagram schematically showing an embodiment of the present invention, FIG. 2 is an optical layout diagram showing the detailed configuration of the focus detection optical system in FIG. 1, and FIGS. Figure 3 (C) is a diagram for explaining the principle of focus detection of a camera, and Figure 4 shows an example of the area of the light-receiving surface of the area image sensor shown in Figure 1 that is used to detect the movement of a subject. A conceptual diagram, FIG. 5 is a conceptual diagram showing the coordinates of the film plane, FIGS. 6(a) and 6(b) are conceptual diagrams showing the pixel configuration of the A area and B area shown in FIG. Figures 7(a) and 7(b) are flowcharts for explaining the procedure for determining the moving speed of a subject, and Figures 8(a) and 8(b)
, FIG. 9 and FIG. 10 are conceptual diagrams for explaining the procedure for determining the image movement speed from the video signal, FIG. 11 is a sectional view schematically showing the structure of the vibration isolator, and FIG. 12 is the vibration isolator. 13 is a conceptual diagram showing the X-axis direction and y-axis direction of the camera, FIG. 14 is a block diagram showing the electric circuit system of the vibration isolator, and FIG. 15(a) ~(C) are diagrams showing the configuration of the encoder, where FIG. 15(a) is a top view, FIG. 15(b) is a perspective view, FIG. 15(C) is a side view, and FIG. 16(a) 16(b) is a schematic perspective view for explaining the operation of the PR sensor, FIGS. 17 and 18 are schematic sectional views showing the relationship between the photographing optical system and the film, and FIG. 19 is the mechanism. A block diagram schematically showing the configuration of the shake detection section, FIG. 20 is a flowchart showing the operation sequence of the mechanical shake detection section, FIG. 21 is a flowchart showing the procedure for determining the image movement speed due to the movement of the subject, and FIG.・A) and FIG. 22(b) are timing charts when determining the image movement speed due to the movement of the subject.
Figure 3 is a flowchart showing the procedure for anti-shake control, Figure 24 is a flowchart showing the CPU operation sequence shown in Figure 1, and Figures 25 to 27 are the basics of the CPU when the camera is in continuous shooting mode. 2 is a flowchart showing a typical operation. 1...Photographing lens, 2...Focus detection optical system, 3...
- Rear image sensor, 4... Interface circuit, 5... CPU, 6... ROM, 7... Lens ROM. 8... Lens drive circuit, 9... Lens drive motor, 10... Slit, 11... Photo interrupter, 11a... Light emitting diode, 11b... Photo resistor, 12... Display device , 13... vibration isolator, 1
4... Anti-vibration sensor, 15... Mechanical shake detection section.

Claims (2)

[Claims] (1) a photoelectric conversion means for converting an image of an object into a video signal; an optical system for forming an image of the object on the surface of the photoelectric conversion means; a movement detection means for detecting movement of an image of an object; a vibration detection means for detecting vibration of the movement detection means using a mechanical vibration sensor; and a detected value of vibration of the movement detection means inputted from the vibration detection means. and a speed calculation means for calculating the movement speed of the object using the detection value of the movement of the image of the object inputted from the movement detection means. (2) a photoelectric conversion means for converting a subject image into a video signal; an optical system for forming the subject image on the surface of the photoelectric conversion means; and an image of the subject using the video signal inputted from the photoelectric conversion means. movement detection means for detecting the movement of the movement detection means; vibration detection means for detecting the vibration of the movement detection means using a mechanical vibration sensor; and a detection value of the vibration of the movement detection means inputted from the vibration detection means and the movement. a speed calculation means for calculating a moving speed of the object using a detected value of movement of the object image input from the detection means; and image blur correction based on the movement speed of the object input from the speed calculation means. What is claimed is: 1. An image blur correction device for a camera, comprising: a correction means for performing the following steps.