JPH01174076A - Photographing device - Google Patents

Abstract

PURPOSE:To correct the movement of the screen due to hand blur and to obtain a normal pattern by providing,e.g., >=2 acceleration sensors to a photographing device case, applying arithmetic processing to outputs of plural acceleration sensors and giving the result to a memory circuit or an address control circuit controlling a photoelectric conversion means. CONSTITUTION:A difference of outputs of the acceleration sensors 9, 10 in pairs represents a horizontal turning of the photographing device 1 and its absolute value depicts the horizontal movement. Then the output of the acceleration sensors 9, 10 is subject to arithmetic operation by an arithmetic circuit 2 to obtain a horizontal moving vector signal on the screen by the movement. Similarly, the vertical movement vector signal is obtained and given to the address control circuit 3. On the other hand, a photographing element 8 applies photoelectric conversion to a range larger than the desired video range and applies analog-digital conversion and stores the result in a memory circuit 4. The address control circuit 3 uses the horizontal movement vector signal and the vertical movement vector signal to control the readout start address of the memory section to control the output of the memory circuit 4 so that there is no movement of the screen. As a result, the movement of the pattern due to hand blur is corrected.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a photographing device such as a video camera, and more particularly to a photographing device that prevents screen shake due to camera shake or the like.

[Conventional technology]

In recent years, with the spread of home video tape recorders (hereinafter referred to as VTRs), opportunities to operate photographing devices such as video cameras have increased.

When taking pictures with this type of photographic device, it is unlikely to be a problem if you use a fixed device such as a tripod.
When operating the camera by hand, "shake" may occur,
The image vibrates vertically and horizontally, resulting in an extremely unsightly image.

Conventionally, as a photographing device that takes measures to prevent camera shake, there is a photographing device that uses a directional gyro that rotates at high speed and a vertical gyro that are connected via a support and a movable part, as shown in Japanese Patent Laid-Open No. 53-64175. As a method for stabilizing the device, there is a method of mechanically controlling the photographing device connected to the support body through a movable part using a rotation 1n signal using an angular velocity sensor, as shown in Japanese Patent Application Laid-Open No. 177867/1983. . Furthermore, as disclosed in Japanese Patent Application Laid-Open No. 198879/1983, there is a method of controlling the reading position of a frame memory storing A/D converted video signals based on motion information between frames of the video signal +a.

[Problem that the invention seeks to solve]

Among the conventional techniques, the one disclosed in Japanese Patent Laid-Open No. 53-64175 is equipped with a gyro that rotates at high speed, resulting in an increase in the size of the photographing device, an increase in weight, and an increase in power consumption. In addition, the device disclosed in Japanese Patent Application Laid-Open No. 177867/1983 uses an angular velocity sensor to correct camera shake that does not involve rotation, that is, screen shake caused by a parallel component. It may not be possible.

Furthermore, since both of the above devices have mechanically movable parts, there is a possibility that the single most visible device becomes larger in size. In addition, in the video signal disclosed in Japanese Patent Application Laid-open No. 61-198879, the frame is divided into a plurality of blocks at L2, and the correlation between blocks is calculated between frames, and the position and direction where the correlation is the largest is determined. Although motion vectors are detected, malfunctions may occur if the subject has a uniform shape and color, such as a wall, a continuous repeating pattern, or multiple objects moving randomly. There is sex.

An object of the present invention is to provide a photographing device that eliminates image blur caused by camera shake.

[Means for solving problems]

The above purpose is to install a means for detecting acceleration, such as acceleration sensors, in two or more locations on the photographing device, a calculation circuit for calculating a motion vector based on the acceleration, a photoelectric conversion means, for example, a memory circuit for storing video signals from an image sensor; This is achieved by providing an address control circuit that controls the address of the memory section based on the output of this arithmetic circuit, and a timing control circuit that provides a reference signal to each of them.

[Function] According to the present invention, the acceleration sensor detects only the -axis direction, and detects the movement of the optical axis in the normal direction of the horizontal plane including the optical axis of the photographic lens, that is, in the vertical direction of the photographing device. One pair is provided with a distance in the direction, and a further pair is provided with a distance in the optical axis direction for detecting vibration in the normal direction of the vertical plane including the optical axis, that is, in the horizontal direction of the photographing apparatus. Taking horizontal movement as an example, the output from the acceleration sensor is a signal proportional to the horizontal acceleration, and the difference between the outputs of the pair of acceleration sensors indicates the rotation of the imaging device in the horizontal direction. The absolute value indicates the amount of horizontal movement.

Therefore, the output of the acceleration sensor is calculated by a calculation circuit to obtain a horizontal motion vector signal on the screen due to the oscillation. Similarly, a vertical motion vector signal is also determined and input to the address control circuit.

On the other hand, the photoelectric conversion means photoelectrically converts an area larger than the desired image, converts it from analog to digital, and stores it in a memory circuit. The address control circuit controls the read start address of the memory section using the horizontal motion vector signal and the vertical motion vector signal, and controls the output of the memory circuit so that the screen does not fluctuate. That is, a part of the memory area of the memory circuit is cut out and used as an output signal. Furthermore, the output signal is digital-to-analog converted into a video output, which is input to video equipment such as a VTR. In the above manner, screen fluctuations caused by camera shake are corrected.

〔Example〕

Hereinafter, one embodiment of the present invention will be described with reference to the drawings.

FIG. 1 is a block diagram of the entire imaging device system illustrating an embodiment of the present invention, in which an image sensor is used as a photoelectric conversion means, an acceleration sensor is used as an acceleration detection means, and a digital memory is used as a memory circuit.

In the figure, a photographing device 1 includes a photographing lens.

The photographing information circuit 11 converts information on the angle of view of the photographing lens and the distance to the object into an electrical signal, and inputs the electric signal to the arithmetic circuit 2.

FIG. 2 is a perspective view of the imaging device according to FIG. 21.2 for
This configuration uses two acceleration sensors. At this time, as long as the acceleration sensor can detect in two axial directions, one pair may be sufficient.

In Figure 1, it is X.

A first acceleration sensor 9 and a second acceleration sensor 10 are shown as acceleration sensors capable of detecting two axes of the y-axis.

Returning to FIG. 1, the arithmetic circuit 2 includes a first acceleration sensor 9,
The acceleration signal from the second acceleration sensor 10, the timing signal from the timing control circuit 7, and the distance information from the photographing information circuit 11 are input, and from the acceleration signal, the information on the screen when the photographing device 1 is shaken due to camera shake is determined. A motion vector is calculated and the calculation result is output to the address control circuit 3.

On the other hand, in the video signal system, the image sensor 8 photoelectrically converts the optical image from the photographing lens L, synchronizes with the first synchronization signal of the timing control circuit 7, and outputs it to the analog-digital conversion circuit 5. The analog-digital conversion circuit 5 converts the video signal from the image sensor 8 into a digital signal, and outputs the digital signal to the memory circuit 4 with the first synchronization signal from the timing control circuit 7.

Here, the photographing range desired by the photographer is the second one in Fig. 3.
The field 41 is synchronized with the image sensor 8.
As shown in the field 40 of the synchronous system, photoelectric conversion is performed over a range more than one time. Therefore, the memory circuit 4 has a large photographing range as a whole, that is, a field 40 of the first synchronization system.
is stored, and a part of it, that is, a range desired by the photographer, is outputted to a field 41 of the second synchronization system.
is output to the digital-to-analog conversion circuit 6. At this time, the address control circuit 3 controls the data range of the second synchronous field 41 on the memory circuit 4 using the motion vector signal from the arithmetic circuit 2. For example, in FIG. 3, after outputting the second synchronization field 41 at time T1, at time T2, if the motion vector on the screen due to camera shake between time T1 and time T2 is 43, the calculation circuit A signal is input from 2 to the address control circuit 3. The address control circuit 3 controls the memory circuit 4 to output the range of the second synchronous field 42 at the time T2. In this way, by moving the field of the second synchronization system on the memory using the motion vector, the fluctuation of the screen is corrected.

Next, the digital-to-analog conversion circuit 6 converts the second synchronous field signal from the memory circuit 4 into an analog signal and outputs it to a video device, such as a VTR (not shown).

In the above manner, screen fluctuations are corrected.

In addition, with this method, when performing operations such as panning and tilting, there is almost no output from the acceleration sensor because the movement is at a constant velocity, so when panning, vertical shake and lateral speed changes are corrected. , when tilting, it compensates for horizontal fluctuations and vertical speed changes. For example, if the second synchronization system field 41 in FIG. 3 comes to the end point of the first synchronization system field 40 due to large acceleration or camera shake when panning is started, the vibration in the direction that came to the end point will be An address control circuit so as to move the second synchronization system field 41 to the center of the first synchronization system field 40 when no correction is performed and the motion vector signal from the arithmetic circuit 2 is smaller than a reference value. may be operated. Next, the acceleration sensors 9 and 10 and the image sensor 8
Although the positional relationship between the two acceleration sensors is arbitrary, the algorithm is simpler when one of the acceleration sensors is on the imaging surface of the imaging element, and it is suitable for correcting vibration.

Note that although this embodiment has described correction in units of fields, the same applies to corrections in units of frames.

Next, a specific example of the arithmetic circuit 2 in FIG.
This will be explained using FIG.

Figure 4 shows the first part in Figure 1. second acceleration sensor 9,
10 is a block diagram showing the relationship between the calculation circuit 10 and the arithmetic circuit 2. FIG.

In this embodiment, it is assumed that the first acceleration sensor 9 is located on a plane that includes the imaging surface of the image sensor 8, and only one direction in the horizontal plane (vertical plane) will be described.

FIG. 5 is an explanatory diagram of the operation of the correction means.

In FIG. 5, at a certain time T, the photographing device 1
is being photographed, and at the next time point T2, the photographing device 1 is
Suppose that when the user moves to the position due to camera shake, an acceleration of 31 is applied to the first acceleration sensor 9 and an acceleration of b is applied to the second acceleration sensor 10. Here, the oscillation component due to camera shake is divided into two types: a translational component and a rotational component. First, how to obtain the parallel component will be explained.

In Fig. 5, l is the distance between the acceleration sensors 9 and 10, k is the distance between the center of rotation and the imaging plane, ψ is the rotation angle, θ□ is the angle of view in the horizontal (vertical) direction, and d is the light to be corrected. The amount of axis movement, do is the amount of optical axis movement to be corrected by the parallel movement component, d'' is the amount of optical axis movement to be corrected by the rotational component, and L is the distance between the subject and the image sensor.Here, originally the acceleration is also Because it changes,
Double integration is performed and the first. Second acceleration sensor 9,1
This will be explained as a uniformly accelerated motion to find the moving amounts m and n of 0.

The movement ff1m,n is determined as follows. However, t is the unit time for performing the correction.

The amount of optical axis movement d" to be corrected by the parallel movement component is equal to the movement amount m. Therefore, the apparent upper rotation angle ψ2 of the subject A due to the parallel movement at time T2 is as follows.

Next, find the rotational component. The distance between the center of rotation and the imaging plane is as follows.

k-□・1 (4) -b Therefore, the rotation angle ψ is as follows.

Therefore, ψ1 + ψ2 is. Therefore, the correction ratio 1c to the horizontal field angle θ8 is found as (ψ, +ψ2)/θ8.

The above algorithm is executed in the block diagram shown in FIG.

Said 1st. The outputs of the second acceleration sensors 9 and 10 are input to the arithmetic circuit 2. Inside the arithmetic circuit 2, an addition circuit 31 calculates the extraction speed for the parallel movement of the imaging plane for the parallel movement component, and a coefficient circuit 33 calculates the shooting information signal, that is, the distance to the subject, the inter-sensor distance i, and the timing control circuit. Calculation of equation (3) is performed with the unit time from 7 to obtain ψ2. Further, the rotational component is detected by the differential circuit 30. The difference between the second acceleration sensors 9 and 10 is taken, and the coefficient circuit 32 calculates ψ1 using the photographing information and the unit time t to obtain ψ1.
1 +ψ2)/θ□ is calculated, and then the correction amount C for the first synchronous frame is sent to the address control circuit 3.
Output to.

Another embodiment for detecting the movement amount C to be corrected will be described with reference to FIGS. 6 and 7.

FIG. 6 is an explanatory diagram of the operation of the correction means, in which the movement id of the optical axis due to the blurring of the image pickup device is calculated by separating it into a parallel component and a rotational component, and each component is added to obtain an in-focus object. This is a diagram showing a means for detecting the amount of movement d of the optical axis over a distance of ''.At this time, a case will be described in which one of the acceleration sensors 9 is on a plane that includes the image sensor 8.

If the output of the acceleration sensor 9 is a, the parallel movement component d'' is an integral form of this output.

d.
It can be detected by the difference (a-b) between the outputs of 0.

In FIG. 6, the difference (a-
Since the triangles 51 including b) are similar, the relationship d": "-(a-b)t"=L: 1 holds true. Therefore, the amount of movement of the optical axis due to rotation d" is d"=2 .(・-b)t” (9).

The optical axis movement amount d to be corrected is the sum of the above two.

d=d'+d" FIG. 7 is a block diagram showing an example of a circuit for calculating the amount of movement d of the optical axis. 9 on the plane containing the image sensor 8, the adder 31 can be omitted.In this configuration, the optical axis movement ld can be detected by setting the coefficients of the coefficient circuits 32 and 33 to be the same.

Since the optical axis movement amount d is the length to be corrected as a distance to the subject, it is converted into a movement hectare to be corrected in the first synchronization system.

The correction amount C for the first synchronous frame to be delivered to the address control circuit 3 is as follows: C=d/D (11). Here, D is the length that appears in the frame of the first synchronization system at a position separated by the distance. This length D is as follows, assuming that the angle of view of the lens is θ.

From equations (10) 2 and (12) above, the correction 1c delivered to the address;b' control circuit 3 is the value obtained by dividing the optical axis movement amount d by the length D of the first synchronization system frame. Become.

FIG. 8 is a more detailed block diagram of the memory system of the photographing device, in which 100 is an image sensor, 101a to 101
c is a signal line, 102a to 102C are analog-to-digital converters (hereinafter referred to as A/D converters), and 103a to 1
03C is a signal line, 104a to 104C are field memory sections, 105a to 105c are digital-analog converters (hereinafter referred to as D/A converters), 106a to 106C
1 is a digital signal line, 107a to 107c are analog signal lines, 108 is a chroma signal generation circuit, 109 is a video signal generation circuit, 110 is an output terminal, and 111 is a timing control terminal.

First, the operation shown in FIG. 8 will be explained. Image sensor section 10
0 consists of a picture tube, MOS type, or CCD type solid-state image sensor, and its peripheral circuits, and the luminance signal (Y
A signal) is output from the signal line 101a, an RY color difference signal is output from the signal line 101b, and a B-Y color difference signal is output from the signal line 101c. The chrominance signal and the two color difference signals are converted into digital signals by A/D converters 102a to 102c, respectively, and stored in field memories 104a to 104, respectively.
written to C. The number of quantization bits of these A/D converters is 8 bits for 102a for Y signal and 10 bits for color difference signal.
Approximately 6 bits are suitable for 2b and 102c. Image sensor 100 and A/D converters 102a to 102c
All operations of the field memories 104a to 104C are as follows.
The timing control circuit 111 operates in synchronization with the first video synchronization system. On the other hand, after reading from the field memories 104a to 104C, operations are performed in synchronization with the second video synchronization system of the timing control circuit. The digital signals read from the field memories 104a to 104c each have 10
Each is converted into an analog signal by the D/A converters 5a to 105C, and the Y signal is sent to the signal line 107a, and the signal line 107
A RY signal is output to b, and a B-Y signal is output to signal line 107c. The chroma signal generation circuit 108 generates a chroma signal from the R-Y signal, the B-Y signal, and the color subcarrier from the timing control circuit, and the video signal generation circuit 109 adds the Y signal and the chroma signal. , a synchronization signal is added, and output from the output terminal 110 as a composite video signal.

The first synchronization system of the above-mentioned imaging section and the second synchronization system after the field memory reading section have the same field period.
Regarding the number of scanning lines and the video area, the first synchronization system includes the second synchronization system.

FIG. 9 is an explanatory diagram of the relationship between the first synchronous system and the second synchronous system, where 120 is the first synchronous system screen, 121 is the second synchronous system screen, and 122 is the first synchronous system screen. The arrow 123 represents the scanning state of the second synchronous system. Here, it is assumed that the second synchronous system screen 121 is twice as large as the first synchronous system screen 120 (double in area ratio). The field synchronization is 18 between the first synchronization system and the second synchronization system. Since the ratio is equal in seconds, the number of scanning lines is 2:1, and the area displayed by the same scanning line is also 2:1, the pixel repetition frequency is 4:1. That is, the clocks of A/D converters 102a to 102C and D
The frequency ratio of the clocks of the /A converters 105a to 105C is 4:1. Further, the screen 121 needs to be moved to an arbitrary position within the screen 120 according to the output signal of the arithmetic circuit.

The structure of the field memory that performs synchronization conversion and can move the readout screen position to an arbitrary position will be described in detail below.

FIG. 10 is a more detailed block diagram of the field memory 104a in FIG.
34 is a control signal for field memories 130 and 131, 135 is a control signal for switch 133, 136 is a control signal for switch 132, 137
is an input terminal, and 138 is an output terminal. Writing to and reading from the field memory is performed at 18 along with screens 120 and 121 in FIG. Since writing and reading are performed in seconds, when writing and reading to the same field memory at the same time (simultaneous in terms of l-field period), writing and reading can be performed using a dual-port memory, or finely time-divided. If writing and reading are performed alternately), the address of the writing system (that is, the screen 120) will overtake the address of the reading system on the way, so the read screen 121 will have two
The screen for one field is mixed up, which is bad.

FIG. 1θ shows two field memories 130 and 131.
In order to solve the above problem, one field period is set exclusively for writing, and the other field is set only for reading, and writing and reading are performed alternately in each field period.

A more detailed block diagram of the above field memory 130 is shown in FIG. 11, and timing charts of the main parts of FIG. 11 are shown in FIGS. 12 and 13.

In FIG. 11, 140 is a memory mat, 141 is a row decoder, 142 is a column decoder and an I10 switch,
143 is a data input terminal, 144 is a data output terminal, 1
45 is a row address counter, 146 is a column address counter, 147 is a row address set register, 148 is a column address set register, 149 is a row address counter 1
45 and column address counter 146 clock (CL
K) input terminal, 150 is the horizontal synchronization signal (hereinafter referred to as H5y
nc) input terminal, 151 is a vertical synchronization signal (hereinafter referred to as
152 is a row address set register 147 and a column address set register 1.
48 is a clock input terminal, 153 is a row address input terminal, and 154 is a column address input terminal.

In FIG. 11, the addresses on the memory mat have a one-to-one correspondence with the positions on the screen, and the l scanning line corresponds to one row of the memory. , address (0,0) to address (M
, N). Similarly, when reading, address (X, Y) and address (X+m, Y+n)
The data in the rectangular area surrounded by will be read out.

FIG. 12 is a time chart when reading HI
VIIe serves as a carry signal (ie, enable signal) for row address counter 145, and serves as a set signal for column address counter 146. Therefore, when Hs□wa is manually operated (becomes Lo-level), the row address counter 145 is incremented by 1, and the column address counter 146 is set to the value of the column address set register 148. V5ync acts as a set signal only for row address counter 145. Therefore, when V5sync is input (becomes Loiv level), the value of the row address counter 145 is set to the value of the row address set register 147.

Prior to a read operation, the row address set register 14
7, row address X, column address set register 148
The column address Y is set in advance by the address input 154, 153 and the clock input 152. In other words, the memory address (X, Y) for correcting hand movement is determined and set based on the acceleration information obtained from the acceleration sensor.The method for determining the (X, Y) coordinates will be described later. .

In FIG. 12, the read operation is started at time t,
H5ync sets the column address to X, and yync sets the row address to Y. Thereafter, the column address is incremented by 1 each time a clock is input.

Next, at time t2, H8yA again. and increment the row address by 1. H5ync+ V 5y in Figure 12
The nc and clock are input in synchronization with the second video signal synchronization system described above, and for example, in the case of the NTSC system, the clock frequency is 4f 5c-x14.3M H.

V,, nc -------------60H
A frequency such as z is suitable.

FIG. 13 is a time chart during writing. During writing, since writing is performed from addresses (0, 0) to addresses (M, N) in FIG. The operation is similar to that in FIG. 12, except that the set register 148 is cleared in advance. The clock frequency and the H5Vne frequency differ depending on the overscan ratio of the first video synchronization system and the second video synchronization system, but if the vertical and horizontal directions are doubled, then the clock frequency -------4f scX 4 = 57
．． 3M H2V -y, 1cm---------
----------60H.

H39, -・・−・−・・・525X 60 =
31.5 KH.

becomes.

Of course, number one. The second synchronous system can also be operated at frequencies other than those mentioned above, and it is also possible to stop the clock during the blanking period to save memory capacity.

Next, from the acceleration sensor output, (X
, Y) address, that is, a method for determining the read address.

The memory read address setting is 18. It is sufficient to perform this every second, and it is necessary to integrate the sensor output for '/6o seconds. Furthermore, since it is necessary to convert the sensor output of analog information into digital information, an A/D converter is required.

In this embodiment, analog integration is performed and 18. We have described an example of converting the movement amount per second and then performing A/D conversion, but in addition to this, it is also possible to directly A/D convert the sensor output,
It is also possible to obtain the (X, Y) address by digital calculation. An advantage of digital calculation is that when the amount of correction is large, the screen can be easily clipped (control is performed so that the screen 121 does not go outside the screen 120 in FIG. 9). At this time, since there is a time of '/6° seconds, it is also possible to calculate the coordinates using a microprocessor.

Regarding the amount of correction, in addition to the method of correcting changes in position,
It goes without saying that methods of compensating for speed changes are possible as well.

Above, we have explained how to correct screen fluctuation by controlling the address of the memory circuit using the detected screen motion vector.Next, we will explain how to control the address of the image sensor, which is a photoelectric conversion means, using the detected screen motion vector. A method of controlling and correcting screen fluctuation will be described using an embodiment shown in FIG. 14.

FIG. 14 is a system block diagram of a photographing apparatus showing another embodiment of the present invention, in which the same components as in FIG. 1 are given the same reference numerals.

In this embodiment as well, an image sensor is used as the photoelectric conversion means, an acceleration sensor is used as the acceleration detection means, and a digital memory is used as the memory circuit. Acceleration signals from the first acceleration sensor 9 and second acceleration sensor lO, timing signals from the timing control circuit 7, and photographing information circuit 1
1, and from the acceleration signal, calculates the hector of movement on the screen when the imaging device 1 is shaken due to camera shake, and outputs it to the address control circuit 3. The address control circuit 3 controls the read address of the image sensor. As a result, a normal image can be obtained from the output video signal, with the screen shake corrected due to camera shake. Therefore, in this embodiment, an A/D conversion circuit, a memory circuit, and a D/A conversion circuit in the video system are not necessarily required, and the system can be simplified.

In the present invention, it is advantageous if the acceleration sensor is of a type constructed on a chip, since manual correction can be performed with one chip.

FIG. 15 is a system block diagram of a photographing apparatus showing still another embodiment of the present invention, in which the same components as in FIG. 1 are denoted by the same reference numerals. In this embodiment, the points that are different from the first embodiment will be mainly explained.

1st. The second acceleration sensors 9 and 10 output acceleration signals to the arithmetic circuit 2. The calculation circuit 2 calculates the movement direction of the screen due to camera shake and outputs a movement direction signal to the address control circuit 3 and the movement detection circuit 150. A video signal from an image sensor 8, which is a photoelectric conversion means operating in the first synchronous system, is converted into a digital signal by an A/D conversion circuit 5, and is input to a motion detection circuit 150 and a memory circuit 4. The motion detection circuit 150 compares the video signal of one field with the video signal of one field one cycle before it, and since the direction of movement is known, it easily detects the amount of movement of the screen and sends it to the address control circuit. Output to 3. The address control circuit 3 controls the read address by the second synchronization signal system of the memory circuit based on the motion direction signal from the arithmetic circuit 2 and the motion amount signal from the motion detection circuit 150, as in the first embodiment. Corrects screen shake. Thereafter, the video signal is converted into an analog signal by the I)/A conversion circuit and output.

In this embodiment, the motion amount of the motion detection circuit can be quickly and accurately determined by determining the direction of screen oscillation by the acceleration detecting means and the motion amount by the video signal.

Furthermore, if only the first acceleration sensor 9 is used as the acceleration sensor, even if the same acceleration is detected by the acceleration sensor, the direction of the screen oscillation may be completely opposite, but since the direction of the oscillation can be known, it is different from the above example. It has a similar effect.

FIG. 16 is a system block diagram of a photographing apparatus showing still another embodiment of the present invention, in which the same components as in FIG. 1 are given the same reference numerals. In this example, the first
．． The second acceleration sensor 9° 10 outputs an acceleration signal to the arithmetic circuit 2. The calculation circuit 2 calculates the movement direction of the screen due to camera shake and outputs a movement direction signal to the address control circuit 3 and the movement detection circuit 160. Since the motion detection circuit 160 knows the direction of motion, it easily compares the video signal of one field with the video signal of one field one cycle before, detects the amount of motion, and outputs it to the address control circuit 3. Similar to the embodiment shown in FIG. 14, the address control circuit 3 controls the readout address of the image sensor to correct screen fluctuations. In this manner, also in this embodiment, by determining the shaking direction of the screen by the acceleration detecting means and the motion amount by the video signal, it is possible to quickly and accurately determine the motion amount of the motion detection circuit. Also in this embodiment, only the first acceleration sensor 9 can be used as the acceleration sensor. At this time, there are four possible movement directions: the X-axis and y-axis directions (horizontal and vertical directions), so the movement detection circuit 160 needs to check in four directions.

In both of the embodiments shown in FIGS. 15 and 16, the motion vector of the shaking of the screen can be roughly determined using the signal from the acceleration sensor, and the motion vector can be determined with high resolution and accuracy using the motion detection circuit.

At this time, there is no problem even if the resolution of the acceleration sensor is slightly lowered, so the sensor interval l can be reduced, the imaging device can be made smaller, and there is a possibility that it can be enclosed in the same IC package. .

The operations performed on a field-by-field basis in the explanation of the above embodiments may also be performed on a frame-by-frame basis.

〔Effect of the invention〕

As explained above, according to the present invention, the outputs of a plurality of acceleration detection means are processed and inputted to an address control circuit that controls a memory circuit or a photoelectric conversion means, thereby correcting screen fluctuations caused by camera shake, etc. A normal screen can be obtained, and a highly functional t-most shadow device can be provided.

[Brief explanation of the drawing]

Figure 1 is a system block diagram of one embodiment of the present invention, Figure 2 is a system block diagram of an embodiment of the present invention.
The figure is a perspective view of the imaging device showing the position of the acceleration sensor.
4 is a block diagram of the arithmetic circuit; FIGS. 5 and 6 are explanatory diagrams of the correction operation of the arithmetic circuit in FIG. 4; FIG. 7 is a block diagram of the arithmetic unit; FIG. 8 is a block diagram of the memory section, FIG. 9 is an explanatory diagram of the first and second synchronization systems, FIG. 10 is a block diagram of other memory sections, and FIG.
The figure is a block diagram of the address control section of the memory, FIG. 12 is a time chart of address changes during reading, FIG. 13 is a time chart of address changes during writing, FIG. 14 is a system block diagram of another embodiment, and FIG. The figure is a system block diagram of still another embodiment, and FIG. 16 is a system block diagram of still another embodiment. 1・-・・・Photographing device, 2−・−・・・Arithmetic circuit, 3
----1 Address control circuit, 4...Memory circuit, 7...-1111 Timing control circuit, 8--
---1 image sensor, 9.10--11 first and second acceleration sensors, 30--1 differential circuit, 32°3
3-11--coefficient circuit, 34--1 addition coefficient circuit, 104--1--field memory section, 14
0---1-...Memory array, 150.160-m-
・−・Motion detection circuit. Figure 1 Loss 4 Signal Figure 2 Figure 3 Figure 4 Tie-n number Figure 6 Figure 7 Figure 8 Figure 9 Figure 10 Figure 11 Figure 12 Figure 14 Gorge 4 Anger Figure 15 N! :No.6 Figure 16

Claims (1)

[Scope of Claims] 1. In a photographing device having a photographing lens and a photoelectric conversion means for converting an optical image of the photographing lens into an electrical signal, the photographing device includes means for detecting two or more accelerations, and a means for detecting two or more accelerations; What is claimed is: 1. A photographing device comprising: arithmetic processing means for inputting a signal from a means for detecting a motion; and an image stabilization means, and configured to prevent a screen from shaking due to camera shake. 2. In the photographing device according to claim 1,
A photographing device characterized in that the camera shake correction means comprises a memory circuit and an address control circuit that controls an address of the memory circuit. 3. In the photographing device according to claim 1,
A photographing device characterized in that the camera shake correction means includes an address control circuit that controls a read address of the photoelectric conversion means. 4. In the photographing device according to claim 1,
An imaging device characterized in that an acceleration sensor is used as means for detecting the acceleration. 5. In the photographing device according to claim 2,
A photographing device characterized in that the memory circuit is a frame memory that stores one frame of television. 6. In the photographing device according to claim 2,
A photographing device characterized in that the memory circuit is a field memory that stores one frame of television. 7. In the imaging device according to claim 3,
A photographing device characterized in that the photoelectric conversion means is a solid-state image sensor. 8. In the photographing device according to claim 3,
A photographing device characterized in that the photoelectric conversion means is an oxide film (MOS) type solid-state image sensor. 9. In the photographing device according to claim 1,
The photographing device, wherein the photoelectric conversion means is configured such that its imaging surface and the axis of the acceleration detection means in the detection direction coincide with each other. 10. In the photographing apparatus according to claim 1, the camera shake correction means comprises a motion detection circuit for detecting screen movement from a video signal, a memory circuit, and an address control circuit for controlling an address of the memory circuit. A photographing device comprising: 11. In the photographing device according to claim 1, the camera shake correction means includes a motion detection circuit that detects screen movement from a video signal and an address control circuit that controls a read address of the photoelectric conversion means. A photographing device characterized by: 12. The photographing device according to claim 1, wherein the arithmetic processing means has two systems of arithmetic processing units, one for a translational component and one for a rotational component.