JP2002094877A - Electronic camera equipment and hand shake correcting method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To enable photographing easily an image in which horizontallity is ensured without user's attention about a horizontal position in the case of photographing by handheld or tripod application or table setting. SOLUTION: A storage device 22 for storing a video signal subjected to hand shake correction, and a coordinate transformation means 21 for performing rotational coordinate transformation in which the image center of the video signal connected with the storage device 22 is made an origin are installed. Concerning an video signal output read from the storage device 22, horizontal positions on an image are maintained in accordance with the inclination of an imaging element 2 by the coordinate transformation means 21.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an electronic camera device such as a so-called video camera device or electronic still camera, and to a camera shake correction method for correcting camera shake in such an electronic camera device.

[0002]

2. Description of the Related Art Conventionally, a so-called "digital electronic camera" which is an electronic camera device or a so-called "camera integrated VT"
As the “R device”, a device having a camera shake correction function for correcting shaking (camera shake) of a photographing screen generated during hand-held photographing has been proposed.

In this camera shake correction function, the shake component of the electronic camera device is detected separately in X direction (horizontal direction of the screen) and Y direction (vertical direction of the screen), and these are corrected independently. . Then, on the imaging screen, X
Correction of direction (screen horizontal direction) and Y direction (screen vertical direction)
Is corrected in a vector manner to realize the correction.

[0004]

However, in the above-described electronic camera apparatus, the problem with the hand-held photographing screen is that, in addition to the above-described camera shake in the X and Y directions, the camera shakes in the tilt direction of the screen. Or, the screen remains tilted. That is, when a captured image is reproduced on a monitor or the like, it is very anxious if the horizontal and vertical in the captured screen are inclined with respect to the horizontal and vertical screen frames of the monitor. On a screen monitor, there is a problem that this inclination becomes even more noticeable.

[0005] In addition, the image quality at the time of photographing, which occurs in a digital camera suitable for photographing still images, can be suppressed by increasing the shutter speed even without a camera shake correction function. However, the photographer still needs to pay attention to the tilt of the screen.

As a mechanism for detecting the inclination of a camera device, Japanese Patent Application Laid-Open No. HEI 6-178190 discloses a mechanism utilizing gravity movement of a conductive liquid in a sealed cylinder. However, such a tilt detection mechanism using a conductive liquid has problems in responsiveness and stability against temperature change, and is difficult to miniaturize.
There is also a problem that it cannot be manufactured at low cost.

Accordingly, the present invention has been proposed in view of the above-mentioned circumstances, and by adding a tilt correction function to a camera shake correction function, the present invention can be applied to hand-held shooting.
Even when shooting on a tripod or on a table, the user can easily shoot horizontal images without worrying about the horizontal position, and has excellent responsiveness and stability against temperature changes. , Electronic camera device that can be miniaturized,
Another object of the present invention is to provide a camera shake correction method applicable to such an electronic camera device.

[0008]

SUMMARY OF THE INVENTION In order to solve the above-mentioned problems, an electronic camera apparatus and a camera shake correction method according to the present invention provide an electronic camera apparatus that obtains a video signal by capturing an image of a subject through an imaging lens with an image sensor. An electronic camera device having a camera shake correction function for correcting image shake in the vertical and vertical directions, comprising a memory for storing a video signal corrected for camera shake, and a rotation centered on the screen center of the video signal connected to the memory. Coordinate conversion means for performing coordinate conversion.

[0009] In the video signal output read from the memory, the horizontal position on the screen is maintained by the coordinate conversion means according to the inclination of the image sensor.

[0010] The camera shake correction method according to the present invention comprises:
In an electronic camera apparatus for obtaining a video signal by capturing an image of an object by an image sensor through an imaging lens, a camera shake correction method for correcting horizontal and vertical image shakes, wherein the image signal corrected for the camera shake is connected to a coordinate conversion unit. The rotation of the image signal is performed by the coordinate conversion means using the center of the screen as the origin, and the video signal output is read from the memory so as to maintain the horizontal position on the screen according to the inclination of the image sensor. It is characterized by the following.

[0011]

Embodiments of the present invention will be described below with reference to the drawings.

At present, camera shake correction in electronic cameras is
Several methods have been proposed, for example, by placing a prism whose shape is variable between a lens and a subject, and based on shake information detected by angular velocity sensors in the X and Y directions,
How to absorb camera shake by shifting the optical axis of the lens,
Alternatively, there is a method in which a position at which an image of a CCD (individual imaging device) is cut out is similarly changed based on shake information detected by an angular velocity sensor in the X and Y directions to absorb hand shake. As a method of detecting the shaking in the X direction and the Y direction, a method of detecting the direction of the shaking motion vector from image information without using an angular velocity sensor has also been proposed.

However, even if the camera shake in the X and Y directions of the screen is absorbed by such a camera shake correction function,
Still, the shake in the tilt direction of the screen remained. An object of the present invention is to provide an electronic camera device having the above-described camera shake correction function that suppresses almost all of the shaking of the electronic camera device that is unsatisfactory for the user, and provides an easily viewable captured image.

The camera shake correction method according to the present invention can be added as an additional function to an electronic camera device using any of the above-described camera shake correction methods. In the following description, an embodiment will be described in which a camera shake correction method for cutting out an image of a CCD (individual imaging device) is used, and an angular velocity sensor is used as a mechanism for detecting a shift in the X direction and the Y direction.

In the electronic camera device to which the present invention is applied, first, as shown in FIG. 1, a subject image is formed on a CCD (individual imaging device) 2 through a zoom lens 1. The CCD 2 is driven by a CCD drive signal from the CCD drive circuit 4 and performs photoelectric conversion. In the case of the camera shake correction method in this electronic camera device, an H (horizontal) direction,
In the V (vertical) direction, a CCD having a pixel area larger than the effective pixel area actually projected on the monitor is adopted, and the V direction is effective in the CCD 2 so that the camera shake is absorbed in accordance with the amount of camera shake that occurs. Pixels are cut out,
In the H direction, effective pixels are cut out by the line memory 5,
The end result is a screen with an aspect ratio of 3: 4 without camera shake.

The signal output from the CCD 2 is
In the camera system signal processing 3, sampling processing, AG
C processing is performed, and after digital conversion by the ADC,
Gamma processing, encoding processing, and the like are performed, and input to the line memory 5. In the line memory 5, the above-described pixel cutout in the H direction is performed, and a predetermined video signal without camera shake is output to the recording system signal processing 6.

In the recording system signal processing 6, for example, in the case of digital video, a video signal is subjected to a blocking process, a shuffling process, a DCT process, and a framing process after requantization, and an audio signal is subjected to an interleave process and a framing process. VTR that has been multiplexed with the previous video signal and has a recording medium such as tape.
It is sent to a recording device and recorded. Illustration and description of the audio signal processing and the VTR recording device are omitted.

Next, camera shake detection is performed by the V direction angular velocity sensor 8 and the H direction angular velocity sensor 9. This angular velocity sensor generates Coriolis force when a rotational angular velocity is applied to a piezoelectric element such as a ceramic, and outputs a voltage proportional to the angular velocity. The angular velocity sensors are arranged one by one in the V direction and one in the H direction. I have. From the output signals of the angular velocity sensors 8 and 9, only signals in a necessary band are extracted by band limits 10 and 11, amplified to a predetermined signal level by AMPs 12 and 13, converted from analog to digital by an ADC 14, and 7 is input.

The microcomputer 7 receives zoom position information from a wide angle to a telephoto position from the zoom lens 1 as a zoom position signal. The microcomputer 7 generates an H-direction camera shake correction signal to be output to the line memory 5 from the zoom position signal.
A V-direction camera shake correction signal to be output to the CCD drive circuit 4 is generated. The zoom position signal is used because the lens angle of view changes according to the zoom position of the zoom lens 1 and the correction amount on the screen differs depending on the lens angle of view even if the same amount of camera shake occurs. This is because the camera shake correction amount needs to be changed accordingly.

As described above, the camera shake correction mechanism is constituted, and the camera shake correction is realized by performing image cutout processing so as to always cancel the camera shake amount detected by the angular velocity sensors in the H direction and the V direction. are doing.

As shown in FIG. 2, in the electronic camera device according to the present invention to which the camera shake correction method according to the present invention is applied, as described above, the output signal from the CCD 2 is processed by the camera system signal processing 3 and It passes through the line memory 5 and is input to the coordinate conversion means 21. This coordinate conversion means 21
A coordinate conversion process is performed for each pixel constituting the screen in the rotation direction with respect to the center (origin) of the screen in which the camera shake in the H direction and the V direction has been corrected, and screen tilt information based on an input tilt direction correction signal; The connected frame memory 22 operates to always cancel the generated inclination.

In this electronic camera device, the coordinate transformation in the rotational direction is always performed after the camera shake correction described above with reference to FIG. This is because if the rotational coordinate conversion is performed before the camera shake correction, the camera shake direction is also rotated from the H and V directions at the same time as the rotational movement, so that the camera shake cannot be corrected. The coordinate conversion means 21 will be further described later. The video signal, the inclination of which has been corrected by the coordinate conversion means 21, is subjected to predetermined processing in the recording system signal processing 6 as in FIG.

The detection of the amount of camera shake in the H and V directions is shown in FIG.
The description is the same as that described above. For tilt direction detection,
First, the inclination amount is detected by a G-direction piezoelectric sensor 23 composed of a quartz oscillator, a ceramic oscillator, or the like. That is, the inclination is detected from the change in the vibration frequency due to the gravity sensitivity characteristic of the piezoelectric vibrator. The G-direction piezoelectric sensor 23 will be further described later.

The output signal from the G direction piezoelectric sensor 23 is
In the FV (frequency-voltage) conversion processing 24, the frequency change is converted into a voltage change, the level is optimized by the AMP 25, the digital is converted by the ADC 26, and input to the microcomputer 7. Microcomputer 7
1, the H-direction correction signal and the V-direction correction signal are output to the line memory 5 and the CCD drive circuit 4, respectively, and the tilt direction correction signal is output to the coordinate conversion means 21 in the same manner as in FIG. , For correcting the tilt direction.

Next, as shown in FIG.
A specific operation example 1 will be described. In FIG.
The point P (X, Y) corresponds to an arbitrary pixel point on the screen, and a luminance signal and a chroma signal exist at the point P in the video signal. Assuming that the point P (X, Y) is located at a distance r from the arbitrarily set origin O and at an angle Φ with respect to the X axis by the coordinate conversion means 21, the point P (X, Y)
Y) is a point P ′ (X ′,
Consider the conversion when moving to Y ′).

First, from X = rcosΦ and Y = rsinΦ, X
'And Y' are represented as follows. X ′ = rcos (Φ + θ) = rcosΦcosθ−rsinΦsinθ
= Xcosθ−Ysinθ Y ′ = rsin (Φ + θ) = rsinΦcosθ−rcosΦsinθ
= Y cos θ−X sin θ Therefore, when the conversion in this case is represented by a determinant, the following expression (1) is obtained.

[0027]

(Equation 1)

By performing the coordinate conversion according to the equation (1) for all the pixels on the screen, the rotation coordinate conversion is performed. At this time, extra pixels are written in the frame memory 22 by the amount corresponding to the rotational movement from the screen to be actually cut out, for tilt correction. Then, the angle θ is obtained based on the tilt direction correction signal input from the microcomputer 7, and while varying so as to absorb the generated tilt,
By reading from the frame memory 22, it is possible to always generate a horizontal screen without tilt.

Here, in the present invention, the origin O in FIG. 3 is, for example, the center of the screen after the camera shake correction in the H direction and the V direction, and coordinate conversion is performed based on the origin O.

Next, FIG. 4 shows a specific example of the G-direction piezoelectric sensor 23. As shown in FIG. 4, the G-direction piezoelectric sensor 23 includes a base 23a, a lead 23b, a supporter 23c, a conductive adhesive 23d, a piezoelectric vibrator 23e, and a piezoelectric vibrator 23e.
f is a lead electrode part, 23g is an excitation electrode. This sensor is hermetically sealed in a metal package not shown in FIG. First, in (a) of FIG. 4, a disk-shaped piezoelectric vibrator 23e is fixed to a supporter 23c at both ends by a conductive adhesive 23d. Also, the excitation electrodes 23 deposited on the front and back of the piezoelectric vibrator 23e
g is electrically connected to the conductive adhesive 23d through the lead electrode portion 23f.
Is electrically connected to a lead 23b which hermetically passes through the base 23a. Therefore, this piezoelectric sensor oscillates stably at the natural frequency of the piezoelectric vibrator in the oscillation circuit.

Here, the inclination of the piezoelectric vibrator 23e in the direction of the arrow shown in FIG. 4A, which is the front and back direction of the piezoelectric vibrator 23e, is deformed by its own weight, causing frequency fluctuation.
This has sensitivity in the direction of gravity, and the amount of frequency variation is greatest when tilted to the side.

FIG. 4B shows a state in which the strip-shaped piezoelectric vibrator 23e stands on the base 23a, and only one side is supported by the conductive adhesive 23d. An excitation electrode 23g is vapor-deposited, and is electrically connected to a lead 23b hermetically penetrating the base 23a through a conductive adhesive 23d.

Here, as in (a) of FIG. 4, the oscillation frequency of the piezoelectric vibrator is changed with respect to the inclination in the direction of the arrow shown in (b) of FIG. Since the piezoelectric vibrator 23e has a one-sided support structure, it is deformed by its own weight, and has a gravity sensitivity characteristic. In this case, the frequency fluctuation becomes largest when the piezoelectric vibrator 23e is tilted right beside.

Further, in order to further increase the gravity sensitivity, an electrode may be further deposited on the upper end of the piezoelectric vibrator to adjust the weight balance.

FIG. 5 shows an example of the output characteristics of the G-direction piezoelectric sensor 23 described above. The vertical axis in FIG. 5 is the ratio between the oscillation frequency F inherent to the piezoelectric vibrator and the frequency variation ΔF due to the gravity sensitivity characteristic, that is, the frequency deviation, and the horizontal axis is the piezoelectric direction in which the gravity direction (downward) is 180 °. The inclination angle of the sensor is shown. Here, looking at the output characteristics, 0 ° and 180 °
The characteristic is such that the deviation is minimum at °, becomes maximum in the + direction at an inclination of 90 °, and becomes maximum in the ゛-direction at an inclination of 270 °. Therefore, by storing the relationship between the frequency deviation and the inclination angle in a microcomputer or the like as a memory map, and reading out the inclination angle with respect to the generated frequency deviation amount from the memory map, the inclination angle can be calculated.

Therefore, the inclination of the electronic camera device can be determined in real time by mounting the G direction piezoelectric sensor 23 shown in FIG. 4 horizontally in the same manner as the horizontal direction of the image pickup device of the electronic camera device.

As described above, the camera shake correction method according to the present invention performs tilt correction on signals from an image pickup device such as a CCD by pure electronic processing. It is characterized in that tilt correction is performed after the direction camera shake correction processing. If the inclination correction is performed before the camera shake correction processing, the camera shake detection directions of the H-direction and V-direction sensors do not match the image shake direction due to the tilt correction, and the camera shake correction processing is complicated.

Further, in the present invention, the inclination correction is performed after the camera shake correction processing, so that the inclination correction can be performed in combination with any of the above-described conventional camera shake correction functions. Stop the function,
It is also possible to operate only the inclination correction function independently.

Next, the coordinate conversion means 21 and the frame memory 22 will be described with reference to FIG. First, in the input video signal, it is necessary to widen the pixel area by the correction amount with respect to the effective pixels after the inclination correction. For example, "NT
"VGA" compatible with the SC standard television signal
If the video output after the correction is required in the “standard” (640 horizontal × 480 vertical), for example, “SVGA” (80 horizontal
O × Vertical 600) or “XGA” (Horizontal 1024 × Vertical 7)
68) A video signal captured by a CCD having a considerable number of pixels is input. If the number of pixels of an image sensor such as a CCD cannot be increased, a so-called “electronic zoom unit” is used in advance.
In this case, a signal may be input after the image size is appropriately enlarged and interpolated.

Next, the aforementioned input video signal is supplied to the SW 34
Are sequentially written to the field memory A or the field memory B in the frame memory 22 composed of two field memories. At the same time, the coordinate origin setting 30 sets the origin for rotation coordinate conversion. This coordinate origin setting 3O is performed, for example, as shown in FIG. 7A, by fixing the lower left corner of the cut-out image frame as the origin, as shown in the coordinate conversion example shown in FIG. (B), the origin may be set at the center of the screen. More specifically, the horizontal and vertical dot clocks and the number of lines are counted and set by the counter means. Further, in the signal in which the coordinate origin is set, the distance from the origin for each pixel is counted by the counter means and converted into coordinates in a coordinate conversion operation 31. In addition, according to the tilt information from the tilt direction correction signal, the rotational coordinate transformation shown in (Equation 1) is calculated for each pixel, and the memory control 32 determines the address to be read from the memory.

Reading from the memory is sequentially performed from the frame memory 22 in accordance with the address determined by the memory control 32 described above. The memory read out at this time is read out from the other B side if the field memory into which the current field signal has been previously written is one A side. Therefore, in the next field, B
SWs 34 and 35 are operated so that the side A on which the current field is written is read out. This makes it possible to always cut out an arbitrary image frame from all the pixels taken into the memory. At this time, the signal of the previous field is read out and used for the current tilt direction correction signal. However, in the case of the video signal, no problem occurs because the inter-field image has a high correlation.

Next, the signal read from the field memory B is input to the interpolation processing 33 via the SW 35, and the signal is interpolated in accordance with the actual pixel coordinates from the theoretical value obtained by the coordinate conversion operation 31 described above. Is performed. This is because, for example, in FIG. 3, the coordinate P ′ (X ′, Y ′) after the conversion is not always on the actual pixel coordinates with respect to the pixel coordinate P (X, Y) before the conversion. Here, an interpolation value on the actual pixel coordinates is calculated and output from the read peripheral pixel signal by, for example, a linear interpolation method.

As described above, as shown in FIG. 7, by extracting an image from a memory capture image frame with a cut-out image frame indicated by a dotted line, the converted image is corrected for inclination as shown by a solid line. The output is made.

In the present invention, as described above, tilt correction can be performed in combination with the conventional camera shake correction method. However, as in the active area method and the vector method described above, in conventional image stabilization, an image frame is cut out from an image sensor such as a CCD in accordance with the amount of image stabilization, and further in the tilt correction, the image frame is cut out. Leads to a decrease in Therefore, in the present invention, the microcomputer 7 determines based on the zoom position information from the zoom lens and the amount of camera shake from the H direction and the V direction, restricts the amount of tilt correction, and prevents the deterioration of resolution as much as possible. I have. This is because, at the wide end of the zoom lens, the amount of camera shake may be small on the screen, so that the amount of tilt correction is increased accordingly, and conversely, the amount of camera shake correction is increased toward the telephoto end. The amount of tilt correction is reduced, and the total amount of pixels used is suppressed. In addition, the amount of tilt correction may be varied in real time in accordance with the amount of occurrence of camera shake correction. Furthermore, when so-called “electronic zoom” is used, a margin is provided for the pixels to be used, so that the margin pixels can be used for inclination correction.

The flow of the above processing is shown in the flowchart of FIG. First, in FIG. 8, the left side shows the flow of processing of the microcomputer 7 described above, and the right side shows the flow of processing of the coordinate conversion means 21 and the frame memory 22.

First, in the microcomputer, it is determined in step 50 whether or not the electronic zoom is used. If so, the margin pixels of the CCD can be used for tilt correction. In step 52, a tilt direction correction signal is generated based on information from the direction sensor. Next, if the electronic zoom is not used, it is determined in step 51 whether or not the camera shake amount is small. For example, if the camera shake amount is small such as when a tripod is used, the amount is reduced accordingly. Since pixels can be used for tilt correction, a tilt direction correction signal is generated in step 52 based on information from the current G direction sensor. Next, when the camera shake amount is large, the current zoom position and the camera shake amount are evaluated in step 53, and the correction amount is limited in the tilt correction amount in step 54, for example, so that the correction is not performed up to the complete horizontal position. Pixels are used for camera shake correction. Then, in step 52, a tilt direction correction signal subjected to correction restriction is generated.

Next, the coordinate conversion means 21 sequentially writes the input signal in the field memory A or B in the frame memory 22 in step 40, and further determines in step 41 the origin of the coordinate conversion in the field image. Is set, and in step 42, the above-described step 52
A coordinate conversion operation is performed for each coordinate from the tilt direction correction signal generated in step (1). The operation result calculated in step 42 is used for generating an address to be read from the field memory in step 43, and further in step 44, a field memory B different from the side to which the data is written in step 40 is used.
Alternatively, the data is sequentially read from A in accordance with the address generated in step 43. Then, in step 45, the read signal is subjected to pixel interpolation processing to correct the theoretical coordinate pixel value and the actual coordinate pixel value, and is output as a tilt correction signal.

As described above, in the present invention, inclination correction is performed. However, in the actual use state of an electronic camera or the like, a necessary inclination also occurs. It is necessary to define a correction range of about 20 ° so that no further tilt correction is performed.

In the present invention, the coordinate origin can be arbitrarily set in the coordinate origin setting 30 shown in FIG. Therefore, for example, by always setting the origin so as to cut out using pixels on the opposite side to the direction in which camera shake occurs, the pixel use efficiency of camera shake correction and tilt correction can be further optimized. As an example, in the case of the inclination correction shown in FIG. 7A, since the origin is set at the lower left corner of the screen, no memory (pixel) is used on the right and lower sides of the screen.
Therefore, the use of pixels does not expand during camera shake in the right and lower directions, and the efficiency is high. This is the same in each of the four corners of the screen, and the pixel use efficiency can be improved in all camera shake directions.

[0050]

As described above, the camera shake correction function of the conventional electronic camera apparatus alone cannot perform the shake correction in the tilt direction of the screen, and the photographer must always take a picture while being conscious of the horizontal. In the electronic camera device and the camera shake correction method according to the present invention, since the inclination is easily corrected, it is only necessary to match the photographing image frame, and photographing can be facilitated. In addition to hand-held shooting, for example, in shooting fixed to a tripod, horizontal shooting can be easily performed without being aware of the inclination of the tripod or the inclination of the ground.

As an electronic camera device, not only a video camera device but also a so-called “digital camera” suitable for photographing still images can similarly use a CCD to perform tilt correction, and can easily set a horizontal camera. Captured still images can be taken.

In comparison with the conventional tilt detecting sensor, a piezoelectric vibrator which has a quick response, a wide operating temperature range, and is inexpensive and small is used. Further, the present invention can be implemented by a circuit configuration independent of a conventional camera shake correction method, and therefore can be applied to any electronic camera device of a camera shake method. All of the coordinate conversion means can be constituted by digital circuits. Therefore, due to recent advances in computer graphics (CG) technology, it is relatively easy to realize a “DSP” or “LSI” at high speed.
Due to recent miniaturization and higher density of semiconductors, increase in circuit scale can be realized with almost no problem.

That is, according to the present invention, by adding a tilt correction function to a camera shake correction function, the user can adjust the horizontal position even when shooting with a handheld device or when shooting with a tripod or placed on a table or the like. An electronic camera device that can easily take a horizontal image without worrying, has excellent responsiveness and stability against temperature change, and can be downsized, and can be applied to such an electronic camera device. It is possible to provide a camera shake correction method.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of an electronic camera device to which a camera shake correction method according to the present invention can be applied.

FIG. 2 is a block diagram illustrating a configuration of an electronic camera device according to the present invention in which the above-described camera shake correction method is performed.

FIG. 3 is a graph showing a state of coordinate conversion in the electronic camera device and the camera shake correction method.

FIG. 4 is a perspective view showing a configuration of a G-direction piezoelectric sensor constituting the electronic camera device.

FIG. 5 is a graph showing characteristics of the G-direction piezoelectric sensor.

FIG. 6 is a block diagram illustrating a configuration of a coordinate transformation and a frame memory included in the electronic camera device.

FIG. 7 is a front view showing an image whose inclination has been corrected by coordinate conversion.

FIG. 8 is a flowchart showing a procedure of a tilt correction process in the electronic camera device.

[Explanation of symbols]

1 zoom lens, 2 CCD, 5 line memory, 7
Microcomputer, 8, 9 angle sensor, 21
Coordinate conversion means, 22 frame memory, 23G direction piezoelectric sensor

Claims (4)

[Claims] 1. An electronic camera device for obtaining a video signal by capturing an image of a subject by an image sensor through an imaging lens, wherein the electronic camera device has a camera shake correction function of correcting image shake in a horizontal direction and a vertical direction. A memory that stores the corrected video signal; and a coordinate conversion unit that is connected to the memory and that performs a rotational coordinate conversion with respect to the screen center of the video signal as an origin. The video signal output read from the memory An electronic camera device, wherein a horizontal position on a screen is maintained by a coordinate conversion unit according to the inclination of the image sensor. 2. The electronic camera device according to claim 1, wherein the camera shake correction function operates based on a tilt from a direction of gravity detected based on a frequency change due to a gravity sensitivity characteristic of the piezoelectric vibrator. 3. An electronic camera apparatus for obtaining a video signal by capturing an image of a subject by an image sensor through an imaging lens, the method comprising: correcting a camera shake in a horizontal direction and a vertical direction; The image data is stored in a memory connected to the conversion means, and the coordinate conversion means performs rotational coordinate conversion with the screen center of the video signal as an origin. A camera shake correction method characterized by reading a video signal output so as to maintain a horizontal position. 4. The camera shake according to claim 3, wherein the camera shake in the electronic camera device is detected based on a tilt from the direction of gravity detected based on a frequency change due to gravity sensitivity characteristics of the piezoelectric vibrator. Correction method.