JP2002016834A - Apparatus and method for correction of shake of imaging device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To correct the motion of an image in an image signal due to the shake of an imaging device. SOLUTION: The output of an angular velocity sensor 13 is amplified by an AGC amplifier 14 at an arbitrary amplification factor, and it is converted into a motion vector-corresponding signal S2 by a correction part 103. The signal S2 and a motion vector signal S1 from a motion-vector sensing circuit 7 are compared in a comparison and decision part 106. On the basis of its comparison result, the reliability of a motion vector is determined. When the degree of the reliability is in a tolerance, the processing operation of a memory control circuit 8 is executed, and the processing operation of a gain control unit 104 is executed. When the degree of the reliability is not in the tolerance, the processing operation of the part 104 is stopped, and the correction processing operation of the motion of the image in the video image signal by the circuit 8 is stopped.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to a shake correction apparatus and method for an image pickup apparatus for correcting the movement of an image in a video signal due to a shake of a video camera, for example.

[0002]

2. Description of the Related Art As is well known, for example, a video camera used for outdoor monitoring may be shaken by an external factor such as wind depending on a place where the video camera is installed. For example, in road monitoring, when a video camera is attached to a pole, the pole is shaken by the wind or the vibration of an automobile, and the video camera attached to the pole is also shaken. In this case, the image of the video camera becomes very unsightly because the angle of view fluctuates with the swing of the pole.

In order to solve such a problem, in a conventional video camera, a shake correction circuit is added to reduce the movement of an image in a final video signal due to the shake, thereby making the shake less noticeable. This type of shake correction circuit detects a motion vector between temporally continuous screens obtained by an image pickup device and, based on the detection result, corrects a motion of an image in a video signal due to a shake of a video camera. The first method and the second method of correcting the movement of an image in a video signal from the detection result by the angular velocity sensor using an angular velocity sensor that physically detects the amount of shake of the video camera are adopted.

[0004] Japanese Patent Application Laid-Open No. Hei 6-339063 and Japanese Patent Publication No. Hei 5-23 disclose the shake correction technique based on each of the above methods.
No. 545 describes the details, and the description thereof will be omitted here.

[0005]

By the way, in the first method, the detection accuracy is high, there is no variation due to individual differences and aging, and the memory cutout can be performed by the actual moving amount (pixel pitch) itself. Although it is possible, it has the following problems. (1) For a moving subject, the motion of the subject is detected as a vector amount instead of the motion of the video camera. (2) Vectors cannot be detected well in subjects with low contrast (including dark subjects). (3) In the high sensitivity mode, correction becomes difficult because the exposure sample time of an image is reduced.

[0006] In the second method, since the shaking of the video camera itself is detected, it is not affected by the state of the subject (video signal) unlike the motion vector. However, it has the following problems. (4) Depending on the mounting state of the video camera, if the movement does not involve rotation, the detection signal becomes small. (5) The sensor output is vulnerable to noise because of an analog small signal. Therefore, when an image is enlarged by zooming the lens, the amount of motion of the image with respect to the shake of the video camera increases, and the sensor output signal per unit pixel (the amount of shake converted into the image) decreases, resulting in a poor SN. The detection accuracy deteriorates. (6) The variation (sensitivity, offset) of the sensor output is large. Therefore, an object of the present invention is to provide an automatic imaging apparatus without being affected by variations due to individual differences in sensor outputs and a difference between an actual shaking of an imaging apparatus and a conversion rate of a rotational motion due to installation conditions of the imaging apparatus. To improve the detection accuracy of
An object of the present invention is to provide a shake correction apparatus and method for an imaging device that can reliably correct the movement of an image due to the shake of the imaging device.

[0007]

According to a first aspect of the present invention, there is provided a shake correcting apparatus for an image pickup apparatus, comprising: a motion vector detecting means for detecting a motion vector from a video signal output from an image sensor; A motion sensor that corrects the motion of the image in the video signal due to the image device shaking based on the physical sensor that physically detects the amount of vibration of the image capturing device; Amplifying means for amplifying the output of the sensor at an arbitrary amplification factor; signal converting means for converting an output signal of the amplifying means into a signal corresponding to a motion vector detected by a motion vector detecting means; and an output signal of the signal converting means Amplification control means for controlling the amplification factor of the amplification means so as to substantially match the output signal of the motion vector detection means, and an output signal of the signal conversion means Comparing the output signal of the motion vector detection means with the output signal of the motion vector detection means to determine the reliability of the motion vector. Execution / stop control means for controlling execution / stop of the process with respect to the means.

In this configuration, when the motion of the image in the video signal due to the motion of the image pickup device is corrected using the motion vector, the actual amount of fluctuation of the image pickup device is detected using the physical sensor. If the amount of fluctuation detected by the physical sensor is small relative to the magnitude of the motion vector, it is determined that the reliability of the motion vector is low, and the motion correction processing is stopped. At this time, in the comparison between the two, a problem due to an individual difference of the physical sensor output and a difference between the actual shaking of the imaging device and the conversion ratio of the rotational motion due to the installation condition of the imaging device become problems. Therefore, the above problem is solved by amplifying the output of the physical sensor, converting it to a signal equivalent to a motion vector, comparing the signal with the motion vector, and controlling the amplification factor with respect to the output of the physical sensor so that the two substantially match. To improve the accuracy of detecting the amount of shaking by the physical sensor. However, if the amplification factor for the sensor output is controlled in the case where the reliability of the motion vector is low, the accuracy will conversely decrease. Therefore, based on the result of comparison between the sensor output and the motion vector, the amplification factor for the sensor output is controlled when the reliability of the motion vector is obtained to some extent, and when the reliability is low, the amplification factor is controlled. I try to stop.

According to another aspect of the present invention, there is provided a shake correction device for an image pickup apparatus, wherein a physical sensor for physically detecting a shake amount of the image pickup apparatus, and an image sensor based on the shake amount of the image pickup apparatus detected by the physical sensor. A motion compensation unit for compensating the motion of an image in a video signal output from a camera, a motion vector detection unit for detecting a motion vector from a video signal output from the image sensor, and a physical sensor Amplifying means for amplifying an output of the amplifying means at an arbitrary amplification factor, a signal converting means for converting an output signal of the amplifying means into a signal corresponding to a motion vector detected by a motion vector detecting means, and providing the signal to a motion correcting means Amplification control means for controlling an amplification factor of the amplification means so that an output signal of the conversion means substantially matches an output signal of the motion vector detection means; The reliability of the motion vector is determined by comparing the output signal of the means with the output signal of the motion vector detection means, and the execution / stop of the process is controlled by the amplification control means according to the degree of the reliability. And execution / stop control means.

With this configuration, when correcting the movement of the image in the video signal due to the shaking of the imaging device using the physical sensor, the variation due to the individual difference of the output of the physical sensor and the imaging condition based on the installation condition of the imaging device are used. Focusing on the problem of the difference between the actual swing of the device and the conversion rate of the rotational movement,
Using the motion vector obtained from the video signal, the output of the physical sensor is amplified, converted into a signal equivalent to the motion vector, and compared with the motion vector. By controlling, the accuracy of detecting the amount of swing by the physical sensor is improved. However,
If the amplification factor for the sensor output is controlled in the case where the reliability of the motion vector is low, the accuracy will conversely decrease. Therefore, based on the result of comparison between the sensor output and the motion vector, the amplification factor for the sensor output is controlled when the reliability of the motion vector is obtained to some extent, and when the reliability is low, the amplification factor is controlled. I try to stop.

In the above configuration, the execution / stop control means determines the reliability of the motion vector based on at least one of the contrast, brightness, histogram, and sensitivity mode of the video signal obtained from the image sensor. When the degree of reliability does not fall within the allowable range, stop control is forcibly performed.

That is, if the contrast of the video signal is insufficient, or if the video signal is dark or the high sensitivity mode is set, a sufficient motion vector cannot be obtained from the video signal. Needless to say, compared with the physical sensor output, the reliability of the motion vector output is low. Therefore, in the above configuration, the reliability of the motion vector is determined from at least one of the contrast, brightness, histogram, and sensitivity mode of the video signal, and when the degree of reliability does not fall within the allowable range, stop control is forcibly performed. Like that.

In the above configuration, the amplification control means includes a storage means for storing a control result when an output signal of the signal conversion means and a detection signal of the motion vector detection means substantially coincide with each other during a control operation, and When the stop control is received by the stop control means, an amplification factor corresponding to the control result stored in the storage means is set in the amplification means.

In this configuration, the reliability of the motion vector is low, and when the amplification control unit receives the stop control of the process, the motion detected from the signal corresponding to the motion vector and the video signal stored during the control operation from the storage unit. The control result when the vector substantially matches the control result is read out, and an amplification factor corresponding to the control result is set in the amplification means. This makes it possible to maintain the accuracy of detecting the amount of fluctuation by the physical sensor, and correct the movement of the image in the video signal due to the fluctuation of the imaging device without being affected by the state of the subject. In addition, every time the reliability of the motion vector is obtained, the control result stored in the storage unit is updated.
It can respond to aging and can have a learning effect.

[0015] In the above arrangement, the signal conversion means is provided when a zoom lens is provided in the imaging device.
An output signal of the amplifying unit is converted into a signal corresponding to a motion vector detected by the motion vector detecting unit in accordance with a zoom magnification.

By doing so, the output of the physical sensor can be converted into a motion vector corresponding to the zoom magnification of the lens, thereby preventing the SN from deteriorating and the detection accuracy from deteriorating.

[0017]

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

(First Embodiment) FIG. 1 is a block diagram showing the configuration of an embodiment of a shake correction device for a video camera according to the present invention.

In FIG. 1, an image picked up via a zoom lens 1 is converted into an electric signal by an image pickup device 2 and subjected to a sampling process in a sample / hold circuit (S / H circuit) 3. The output signal of the sample-and-hold circuit 3 is converted into a digital signal by an A / D (analog / digital) converter 4, subjected to processing such as white balance, gamma correction, and details by a signal processing circuit 5. Is stored in

The output signal of the A / D converter 4 is also input to a motion vector detection circuit 7 to obtain a motion vector signal S1. The motion vector signal S1 indicates how many pixels the pixel has moved per unit time (for example, one field) in the X and Y directions. This motion vector signal S1 is input to the memory control circuit 8 and the microcomputer 10. The memory control circuit 8 determines a cut-out area from a writing area of the memory 6 and controls reading of the memory 6 according to the input vector amount (movement amount of the video camera) so as to cancel the vector amount. The output signal of the memory 6 is enlarged by an electronic zoom circuit 9 and converted into an NTSC video signal by an encoder 11 so as to match an angle of view of a television monitor, and is converted into a D / A (digital / analog) converter 1.
2 obtains an analog final video output signal.

The microcomputer 10 includes an A / D conversion unit 102, a correction unit 103, a gain control unit 104, a nonvolatile memory 105, and a comparison determination unit 106.

The A / D converter 102 outputs the output of the angular velocity sensor 13 for physically detecting the amount of shake of the video camera to A / D.
A signal amplified at an arbitrary gain by a GC (auto gain controller) 14 is input. The angular velocity sensor 13 has a dual configuration for the horizontal and vertical directions, and the AGC amplifier 14 has a dual configuration for amplifying the detection signal in each direction.

The amplified signal of the AGC amplifier 14 is A / A
The signal is converted into a digital signal by the D conversion unit 102 and
3, after being converted into a signal S2 equivalent to a motion vector detected by the motion vector detection circuit 7,
04 and the comparison determination unit 106. The correction unit 103 performs a correction process by changing a correction amount depending on the zoom position of the zoom lens 1 so that a signal corresponding to a motion vector can be obtained at any zoom position. As a result, it is possible to prevent the SN from deteriorating and the detection accuracy from deteriorating.

The gain control unit 104 converts the comparison result signal S4 to A so that the output signal S2 of the correction unit 103 substantially matches the motion vector signal S1 from the motion vector detection circuit 7.
The signal is output to the GC amplifier 14 to control the amplification factor. Further, the gain control unit 104 stores the amplification setting value when the motion vector signal S1 matches the output signal S2 of the correction unit 103 in the nonvolatile memory 105 during the control operation.

The comparing and judging section 106 calculates the motion vector signal S
1 and the output signal S2 of the correction unit 103, and the correlation is determined to determine the reliability of the motion vector signal S1. Further, the reliability of the motion vector is determined from the contrast information output from the signal processing circuit 5 and the signal S7 indicating the sensitivity mode, and the processing is executed on the memory control circuit 8 according to the degree of the reliability. Control the stop and execute the processing to the gain control unit 104
Control the stop.

When the gain control unit 104 receives the control start notification S5 from the comparison determination unit 106, the gain control unit 104 executes S1 and S2.
The gain of the AGC amplifier 14 is controlled so that
The set value at the time of coincidence is stored in the nonvolatile memory 105. Thereafter, the AGC amplifier 104 becomes the nonvolatile memory 105
It operates with the value stored in.

Next, the principle of shake correction in the above configuration will be described with reference to FIG. In the memory 6, the memory cutout area (the range to be actually read) is moved by the memory control circuit 8 in a direction opposite to the moving direction of the video camera. Since the readout signal is smaller than the actual imaging range, the readout signal is enlarged by the electronic zoom by the electronic zoom circuit 9 to match the angle of view of the television monitor.

Next, the correlation judgment method in the comparison judgment section 106 will be described with reference to FIG. The comparison / determination unit 106 determines the difference between the motion vector signal S1 and the output signal S2 of the correction unit 103, and if the value is within the threshold range, determines that the motion vector has detected the motion of the video camera. That is, since the output of the angular velocity sensor 13 responds only to the movement of the video camera, the angular velocity sensor 1
If the correlation with the detection result of No. 3 is strong, it can be determined that the shaking of the video camera has been detected. The reason why the threshold value is given to the determination is that the output of the angular velocity sensor 13 varies due to individual differences, and the difference between the actual shaking of the video camera and the conversion rate of the angular velocity due to the mounting state of the video camera.
This is because the relationship between the corrected signal and the actual motion vector varies (it does not completely match).

Next, the operation of the comparing and judging section 106 will be described with reference to FIG.
This will be described with reference to the flowchart of FIG. First, the comparison determination unit 106 obtains a difference value α between the motion vector signal S1 and the output signal S2 of the correction unit 103 (step a1), and determines whether the difference value α is smaller than the threshold value TH (step a1). Step a2). Here, if the difference value α is larger than the threshold value TH (No), the comparison determination unit 106 stops the image motion correction process of the memory control circuit 8 and
The gain control process of the gain control unit 104 is stopped (step a3).

On the other hand, if the difference value α is smaller than the threshold value, the comparing and judging section 106 executes the following judging process from the signal S6 from the signal processing circuit 5.

(1) Whether the mode is the standard sensitivity mode (step a4). (2) Whether the contrast of the video signal is sufficient (step a5). Here, when the above conditions (1) and (2) are satisfied, the comparison determination unit 106 gives the control instruction S3 of the image motion correction to the memory control circuit 8 and the motion vector to the gain control unit 104. An amplification factor control instruction S5 based on a comparison result between the signal S1 and the output signal S2 of the correction unit 103 is given (step a6). If any one of the above conditions (1) and (2) is not satisfied, the comparison / determination unit 10
Step 6 executes the process of step a3.

As a result, the influence of the variation of the angular velocity sensor 13 (individual difference, aging) and the detection sensitivity of the angular velocity sensor 13 due to the condition of the installation location of the video camera are absorbed, and the accuracy of the comparison and judgment section 106 is improved (threshold). Value can be set small), and erroneous determination can be prevented. Since this control always operates with the video camera installed, the latest data is always stored in the non-volatile memory 105. Therefore, even if the installation location of the video camera is changed, the control is automatically performed. It is possible to respond to the current installation environment.

As described above, in the first embodiment, when correcting the motion of the image in the video signal due to the shake of the video camera using the motion vector, the actual angular velocity sensor 13 is used. The shake amount of the video camera is detected, the output of the angular velocity sensor 13 is amplified by the AGC amplifier 14, converted into a signal S 2 corresponding to a motion vector by the correction unit 103, and then the signal S 2 and the motion vector signal are output by the gain control unit 104. By comparing S1 with the gain of the AGC amplifier 14 and controlling the gain of the AGC amplifier 14 so that the two substantially coincide with each other, variations in the output of the angular velocity sensor 13 due to individual differences, and actual video camera installation conditions due to the installation conditions of the video camera are considered. The angular velocity sensor 13 eliminates the difference between the conversion rate of the shaking and the rotational motion.
, The accuracy of detecting the amount of shaking can be improved.

Further, the correction unit 1 is
The reliability of the motion vector signal S1 is determined by comparing the output signal S2 of FIG. 03 with the motion vector signal S1, and when the reliability of the motion vector is obtained to some extent based on the determination result, the memory control circuit 8 To the gain control unit 104, and
The gain is controlled so as to control the gain of the GC amplifier 14, and when the reliability is low, the processing of the memory control circuit 8 is stopped and the processing of the gain control unit 104 is stopped. Then, when the gain control unit 104 receives the stop control of the process from the comparison determination unit 106, when the signal S2 corresponding to the motion vector stored from the nonvolatile memory 105 during the control operation substantially matches the motion vector signal S1. Read out the amplification setting value of AGC
The amplifier 14 is set.

For this reason, when the shake amount of the video camera detected by the angular velocity sensor 13 is small relative to the magnitude of the motion vector, it can be determined that the reliability of the motion vector is low, and the memory control circuit 8 and the Gain control unit 104
Is stopped, the motion of the image in the video signal is corrected when the reliability of the motion vector is low, or the amplification rate of the AGC amplifier 14 is controlled to decrease the accuracy of detecting the amount of vibration by the angular velocity sensor 13. Can be prevented. At the time of this stop, the nonvolatile memory 105
By setting the amplification setting value stored in the AGC amplifier 14 to the AGC amplifier 14, it is possible to maintain the accuracy of detecting the amount of shaking by the angular velocity sensor 13, and the video signal caused by the shaking of the video camera without being affected by the state of the subject. The movement of the middle image can be corrected. Further, each time the reliability of the motion vector is obtained by the comparison / determination unit 106, the amplification setting value stored in the nonvolatile memory 105 is updated by the gain control unit 104. It can also have a learning effect.

Further, the comparing and judging section 106 includes the image pickup device 2
, The reliability of the motion vector is determined based on the contrast and sensitivity mode of the video signal obtained from the video signal, and when the degree of reliability does not fall within the allowable range, stop control is forcibly performed. That is, when the contrast of the video signal is insufficient or when the mode is set to the high-sensitivity mode, a sufficient motion vector cannot be obtained from the video signal. Needless to say, the reliability of the motion vector signal S1 is low.
Therefore, the reliability of the motion vector is determined from at least one of the contrast of the video signal and the sensitivity mode, and when the degree of the reliability does not fall within the allowable range, the stop control is forcibly performed.

(Second Embodiment) FIG. 5 is a block diagram showing a configuration of a second embodiment of a shake correction device for a video camera according to the present invention. In FIG. 5, the same parts as those in FIG. 1 are denoted by the same reference numerals, and different parts will be described here.

The difference from FIG. 1 is that the control processing of the memory control circuit 8 is performed based on the detection result of the angular velocity sensor 13. Since the angular velocity sensor 13 detects the movement of the video camera itself, the comparison determination unit 106
Does not control the memory control circuit 8. That is, the output signal S2 of the correction unit 103 is provided to the memory control circuit 8.
Is input, and memory cutout control is performed based on the signal S2.

Next, the comparison / decision unit 106 in the above configuration
Will be described with reference to the flowchart of FIG. Steps b1 to b4 are the same as steps a1, a2, and a4a5 in FIG.

The comparison / determination unit 106 calculates the motion vector signal S
1 and the output signal S2 of the correction unit 103 are smaller than the threshold value TH, and if the contrast of the video signal is sufficient in the standard sensitivity mode, the motion vector signal S1 and the correction unit An instruction S5 for amplification control based on the result of comparison with the 103 output signal S2 is given (step b).
5) If the difference value α is larger than the threshold value and the contrast of the video signal is insufficient in the high sensitivity mode, the gain control process of the gain control unit 104 is stopped (step b6).

In the second embodiment, the angular velocity sensor 1
In the case where the motion of the image in the video signal caused by the shaking of the video camera is corrected by using the video camera 3, the fluctuation due to the individual difference of the output of the angular velocity sensor 13 and the actual shaking and rotation of the video camera due to the installation condition of the video camera are used. Paying attention to the fact that the difference from the conversion rate of the movement becomes a problem, the output of the angular velocity sensor 13 is amplified by the AGC amplifier 14, converted into the signal S 2 corresponding to the motion vector by the correction section 103, and By comparing the vector signal S1 and controlling the amplification factor of the AGC amplifier 14 so that the two substantially match each other, the accuracy of detecting the amount of swing by the angular velocity sensor 13 is improved. However, if the reliability of the motion vector is low,
If the gain of the AGC amplifier 14 is controlled, the accuracy will be reduced. Therefore, the comparison determination unit 106 compares the signal S2 corresponding to the motion vector with the motion vector signal S1 from the motion vector detection circuit 7, and based on the comparison result, when the reliability of the motion vector is obtained to some extent, the gain control unit 104 Is executed to control the gain of the AGC amplifier 14, and when the reliability is low, the process of the gain control unit 104 is stopped. At the time of this stop, the gain control unit 104 reads from the nonvolatile memory 105 an amplification setting value when the signal S2 corresponding to the motion vector and the motion vector signal S1 stored at the time of the control operation substantially coincide with each other, and reads the amplification setting value. The AGC amplifier 14 is set.

For this reason, the motion vector is used without being affected by the variation due to the individual difference of the output of the angular velocity sensor 13 or the difference between the actual swing of the video camera and the conversion rate of the rotational motion due to the installation condition of the video camera. In addition to improving the accuracy of detecting the amount of swing of the angular velocity sensor 13, the motion of the image due to the swing of the video camera can be reliably corrected.

(Other Embodiments) The present invention is not limited to the above embodiments. As shown in FIG. 7, for example, the processing of the correction unit 103 can eliminate the need for controlling the correction of the angular velocity sensor signal based on the zoom ratio information of the zoom lens 1. In FIG. 7, the same parts as those in FIGS. 1 and 5 are denoted by the same reference numerals, and description thereof will be omitted.

That is, the shake correcting device of the video camera shown in FIG.
By controlling so that the signal S2 corresponding to the motion vector of the angular velocity sensor 13 has the same level, the difference in detection sensitivity due to the zoom ratio of the angular velocity sensor 13 is absorbed.

In other words, the correction unit 103 does not need to perform correction control based on the zoom ratio, but simply converts the detection signal of the angular velocity sensor 13 into a signal equivalent to a motion vector. Will not be.

For example, a box-type surveillance video camera sold generally cannot capture information on the zoom ratio of a lens into the video camera. Further, even in the case of a single focus lens, since the focal length varies depending on the lens, the angle of view varies depending on the lens. Therefore, conventionally, it has been difficult for a general surveillance video camera with an external lens to control the shake correction based on the detection signal of the angular velocity sensor 13. However, according to the present invention, the angular velocity sensor 1
3 can be performed without using information on the zoom ratio.

In each of the above embodiments, the reliability of the motion vector is determined based on the sensitivity mode and the contrast of the video signal. However, the reliability of the motion vector is determined based on the brightness of the video signal and the histogram. You may make it. In this case, when the subject is dark and a sufficient motion vector cannot be obtained from the video signal, the signal S2 corresponding to the motion vector from the nonvolatile memory 105 matches the motion vector S1 detected from the video signal. By reading the amplification setting value at the time of the setting and setting it in the AGC amplifier 14, it is possible to correct the movement of the image in the video signal due to the shaking of the video camera without being affected by the state of the subject. .

In each of the above embodiments, each unit in the microcomputer 10 has a hardware configuration. However, the present invention can be similarly implemented with a software configuration.

In addition, the type and configuration of the video camera, the type of the angular velocity sensor, and the like can be variously modified without departing from the gist of the present invention.

[0050]

As described in detail above, according to the present invention,
Even when the installation location of the imaging device changes, the detection sensitivity of the angular velocity sensor can be automatically corrected, which allows the sensor output to change over time, variations due to individual differences, and the actual imaging device installation conditions due to the installation conditions of the imaging device. Provided is a shake correction device and method for an imaging device that can reliably correct the motion of an image in a video signal due to the swing of the imaging device without being affected by the difference between the conversion rate of the shake and the rotational motion. it can.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of a first embodiment of a shake correction device for a video camera according to the present invention.

FIG. 2 is a view for explaining the principle of shake correction in the first embodiment.

FIG. 3 is a view for explaining a correlation determination method in the comparison determination unit shown in FIG. 1;

FIG. 4 is a flowchart for explaining the operation of the comparison / determination unit shown in FIG. 1;

FIG. 5 is a block diagram showing a configuration of a shake correction device for a video camera according to a second embodiment of the present invention.

FIG. 6 is a flowchart for explaining the operation of the comparison / determination unit shown in FIG. 5;

FIG. 7 is a block diagram showing a configuration of a video camera shake correction device according to another embodiment of the present invention.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Zoom lens, 2 ... Image sensor, 3 ... Sample hold circuit (S / H circuit), 4 ... A / D converter, 5 ... Signal processing circuit, 6 ... Memory, 7 ... Motion vector detection circuit, 8 ... 9: electronic zoom circuit, 10: microcomputer, 11: encoder, 12: D / A converter, 13: angular velocity sensor, 14: AGC (auto gain controller) amplifier, 102: A / D conversion Unit 103: correction unit 104: gain control unit 105: nonvolatile memory 106: comparison and determination unit

Claims (7)

[Claims] 1. A motion vector detecting means for detecting a motion vector from a video signal output from an image pickup device, and a motion of an image in the video signal due to a shaking of an image pickup device based on a detection result by the motion vector detecting means. A motion sensor for physically correcting the amount of shaking of the imaging device; an amplifying device for amplifying an output of the physical sensor at an arbitrary amplification factor; Signal converting means for converting the output signal of the amplifying means into a signal corresponding to a motion vector detected by the motion vector detecting means; and amplifying the output signal of the signal converting means so as to substantially coincide with the output signal of the motion vector detecting means. Amplification control means for controlling the amplification factor of the means, and comparing the output signal of the signal conversion means with the output signal of the motion vector detection means, In accordance with the degree of the reliability, the execution / stop of the process is controlled by the motion correction means, and the execution / stop of the process is controlled by the amplification control means. A shake correction device for an imaging device, comprising: a stop control unit. 2. A physical sensor for physically detecting a shake amount of an imaging device, and a motion of an image in a video signal output from an image sensor based on the shake amount of the image pickup device detected by the physical sensor. A motion vector detecting means for detecting a motion vector from a video signal output from the image sensor; and an amplifying means for amplifying an output of the physical sensor at an arbitrary amplification factor. A signal converting means for converting an output signal of the amplifying means into a signal corresponding to a motion vector detected by the motion vector detecting means and providing the signal to the motion correcting means; and an output signal of the signal converting means, Amplification control means for controlling an amplification factor of the amplification means so as to substantially match an output signal; and an output signal of the signal conversion means and an output of the motion vector detection means. An execution / stop control unit that determines the reliability of the motion vector by comparing the signal with a signal, and controls the execution / stop of the process with respect to the amplification control unit according to the degree of the reliability. A shake correction device for an imaging device, characterized in that: 3. The execution / stop control unit includes a control unit that controls at least a contrast, a brightness, and a brightness of a video signal obtained from an image sensor.
The reliability of the motion vector is determined based on one of a histogram and a sensitivity mode, and when the degree of reliability does not fall within an allowable range, stop control is forcibly performed. A shake correction device for an imaging device according to any one of the preceding claims. 4. The amplification control means includes storage means for storing a control result when an output signal of the signal conversion means and a detection signal of the motion vector detection means substantially coincide with each other during a control operation, and wherein the execution / stop control is performed. 3. The apparatus according to claim 1, wherein an amplification factor corresponding to a control result stored in the storage unit is set in the amplification unit when stop control is performed by the unit. 5. The signal conversion means, when a zoom lens is provided in the image pickup device, outputs the output signal of the amplification means in accordance with a zoom magnification corresponding to a motion vector detected by the motion vector detection means. 3. The apparatus according to claim 1, wherein the image is converted into a signal. 6. A shake correction method for an image pickup apparatus, comprising: detecting a motion vector from a video signal output from an image pickup device; and correcting a motion of an image in the video signal due to a shake of the image pickup apparatus based on the detection result. Physically detecting the amount of shaking of the imaging device by a physical sensor, amplifying the output of the physical sensor at an arbitrary amplification factor, converting the amplified signal into a signal equivalent to a motion vector detected from a video signal, Controlling the amplification factor when amplifying the physical sensor output so that the signal corresponding to the motion vector substantially matches the motion vector, comparing the signal corresponding to the motion vector with the motion vector, Judgment of the reliability, control of execution / stop of the motion compensation process according to the degree of the reliability, and control of execution / stop of the amplification control process. Shake correction method of that imaging device. 7. A shake correction method for an image pickup device for correcting a motion of an image in a video signal output from an image pickup device based on a shake amount of the image pickup device detected by a physical sensor. A motion vector is detected from the video signal, and the output of the physical sensor is amplified at an arbitrary amplification factor. The amplified signal is converted into a signal corresponding to a motion vector detected from the video signal. By correcting the motion of the image in the video signal, controlling the amplification factor when amplifying the physical sensor output so that the signal corresponding to the motion vector substantially matches the motion vector detected from the video signal, The reliability of the motion vector is determined by comparing the signal corresponding to the motion vector with the motion vector detected from the video signal, and the degree of reliability is determined. Shake correction method for an imaging device and controls execution and stop of the amplification control processing Flip.