JP3508225B2 - Image timing coincidence detection method and apparatus - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image timing coincidence detecting method and apparatus for detecting coincidence of timings of a plurality of images in the time direction.

[0002]

2. Description of the Related Art Generally, when a scene of a moving image is captured by a plurality of image capturing apparatuses having different input systems, the same scene can only be captured at the same time.
For this reason, when a moving image is recorded, in order to extract the same scene, it is necessary to record the real time at the same time at the time of recording and then perform the cutout with reference to the time code.

[0003]

By the way, when the same scene is imaged by using a plurality of image pickup devices, the plurality of image pickup devices cannot be physically arranged on one optical axis. It's not exactly the same scene. In particular, when an image of a three-dimensional object is to be captured, it is necessary to devise a method such as spectral separation on the optical axis.

Even if the optical axis is spectrally separated, it is necessary to record real time in a plurality of image pickup devices having different input systems. If such a time code cannot be recorded, a plurality of picked-up images are taken. The timing in the time direction cannot be adjusted.

It is an object of the present invention to provide an image timing coincidence detecting method and device capable of detecting coincidence of timings of images picked up by a plurality of image pick-up devices, without needing to record a time code. And

[0006]

An image timing coincidence detecting method according to the present invention comprises a first image output from a first image sensor having a first pixel number and a second image having a second pixel number. Image timing coincidence detection method for detecting coincidence in time direction with a second image output from the image pickup device, the average value, standard deviation, and time difference between unit images of the first image and the second image. The problem is solved by detecting the coincidence of the timings in the time direction between the images by using at least one piece of information of the movement amount.

In this case, the above information is the average value, standard deviation, and unit image of the first image and the second image in the unit image.
It is at least one of the time difference and the time series change value of the motion amount.

Further, the average value, standard deviation, time difference,
The information on the amount of movement is evaluated using a correlation coefficient.

Further, in the image timing coincidence detection method according to the present invention, the first image output by the first image sensor having the first number of pixels and the second image sensor having the second number of pixels are output. An image timing coincidence detecting method for detecting coincidence in time direction with a second image to be output, comprising: an average value, a standard deviation, a time difference, and a motion amount of a unit image of the first image and the second image. The above problem is solved by performing evaluation using a correlation coefficient for the time-series change value, and comprehensively judging the evaluation value to detect the coincidence of the timing in the time direction between the images.

Further, in the image timing coincidence detecting apparatus according to the present invention, the first image output by the first image sensor having the first number of pixels and the second image sensor having the second number of pixels are provided. In an image timing coincidence detection apparatus for detecting coincidence in time direction with a second image to be output, an average value, a standard deviation, a time difference, and a motion amount of a unit image of the first image and the second image are detected. The above problem is solved by detecting the coincidence of the timing in the time direction between the images using at least one information in the above.

In this case, the above information is the average value, standard deviation, and unit deviation of the first image and the second image in the unit image.
It is at least one of the time difference and the time series change value of the motion amount.

The average value, standard deviation, time difference,
The information on the amount of movement is evaluated using a correlation coefficient.

Further, in the image timing coincidence detecting apparatus according to the present invention, the first image outputted by the first image pickup device having the first number of pixels and the second image pickup device having the second number of pixels are provided. In an image timing coincidence detection apparatus for detecting coincidence in time direction with a second image to be output, an average value, a standard deviation, a time difference, and a motion amount of a unit image of the first image and the second image are detected. The above problem is solved by performing evaluation using a correlation coefficient for the time-series change value, and comprehensively judging the evaluation value to detect the coincidence of the timing in the time direction between the images.

In this case, the first and second image pickup devices image the same subject. Also, for the subject,
A marker indicating the reference pixel position is attached.

[0015]

In the image timing coincidence detecting method according to the present invention, the first image output by the first image sensor having the first number of pixels and the second image sensor having the second number of pixels are output. Average value, standard deviation, in unit image of the second image,
Since at least one of the time difference and the amount of movement is used, it is possible to detect the coincidence of the timings of the images in the time direction without the need to record the time code.

In the image timing coincidence detection method according to the present invention, the first image output by the first image sensor having the first number of pixels and the second image sensor having the second number of pixels are output. The average value, the standard deviation, the time difference, and the time series change value of the motion amount in the unit image of the second image are evaluated using the correlation coefficient, and the evaluation value is comprehensively determined. It is possible to detect the coincidence of the timings of the images in the time direction even though the recording is unnecessary.

In the image timing coincidence detecting apparatus according to the present invention, the first image output by the first image sensor having the first number of pixels and the second image sensor having the second number of pixels output. Since at least one of the average value, the standard deviation, the time difference, and the amount of motion in the unit image of the second image is used, it is possible to detect the coincidence of the timings of the images in the time direction while not recording the time code. it can.

In the image timing coincidence detecting apparatus according to the present invention, the first image output by the first image sensor having the first number of pixels and the second image sensor having the second number of pixels output. The average value, the standard deviation, the time difference, and the time series change value of the motion amount in the unit image of the second image are evaluated using the correlation coefficient, and the evaluation value is comprehensively determined. It is possible to detect the coincidence of the timings of the images in the time direction even though the recording is unnecessary.

[0019]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Preferred embodiments to which the image timing coincidence detection method and apparatus according to the present invention can be applied will be described below. This embodiment is an image timing coincidence detection device for detecting coincidence of timings in time direction of two images obtained by capturing two images of the same subject by different image capturing devices. Here, at least one of the two image capturing apparatuses having different input systems is an image capturing apparatus that cannot record a time code at the same time as image capturing, and the image timing coincidence detecting apparatus according to the present embodiment is capable of performing real-time operation. It is impossible to detect timing coincidence by reference.

Therefore, the image timing coincidence detection apparatus of the present embodiment reads out an image for a certain fixed time, calculates a characteristic amount of the image in units of frames or fields, and time-series data of the fixed amount for the characteristic amount. Is held, matching is performed using the time-series data, and a timing coincidence point is detected to output a timing phase difference. As the feature amount, the average luminance m of the supplied image signal, the standard deviation σ, the interframe difference Δ, and the motion vector v are used.

That is, as shown in FIG. 1, the image timing coincidence detecting apparatus of the present embodiment detects the first image signal supplied from the input terminal 1a and the second image signal supplied from the input terminal 1b. An average brightness value calculation circuit 2 for calculating an average brightness value m of a signal, a time series value storage unit 3 for holding time series data of the average brightness value m for a fixed time, a signal of the first image and the first image. 2, a standard deviation calculating circuit 4 for calculating the standard deviation σ of the signal of the image, a time series value storage unit 5 for holding the time series data of the standard deviation σ for a certain time, the signal of the first image and the above Interframe difference Δ of the signal of the second image
Of the inter-frame difference calculation circuit 6, a time-series value storage unit 7 that holds time-series data of the inter-frame difference Δ for a certain period of time, a signal of the first image and a signal of the second image. A motion vector calculation circuit 8 for calculating a motion vector v, a time-series value storage unit 9 that holds time-series data of the motion vector v for a certain time, the time-series value storage unit 3,
By matching the various time series data of the signal of the first image and the various time series data of the signal of the second image, which are supplied from 5, 7, and 9 in the time direction, and detecting the timing coincidence point. And a time direction matching circuit 10 that outputs a timing phase difference.

The first terminal supplied from the input terminal 1a
FIG. 2 shows a method for obtaining the signal of the image and the signal of the second image supplied from the input terminal 1b.

The image pickup device 11 on one side picks up an image of the subject 16 transmitted through the half mirror 15, and the other image pickup device 12 picks up an image of the subject 16 reflected on the half mirror 15. That is, since it is not possible to physically arrange the two image pickup devices 11 and 12 on one optical axis, the same image is obtained by splitting with the half mirror 15.

Here, the one image pickup device 11 is provided with a first image pickup device having a first pixel number, for example, and outputs a first image. Further, the image pickup device 12 includes a second image pickup device having a second number of pixels, for example, and outputs a second image. Here, specifically, the first image is one of the standard television systems, for example, NTSC.
The low resolution image (hereinafter, S
As an example, a case of handling a high-resolution image (hereinafter referred to as an HD (high-definition) image), which is a high-definition method, for example, 1125 lines, is used as the second image.

The images respectively captured by the first image pickup device or the second image pickup device are recorded in the recording section of the image pickup device 11 or 12, respectively. Here, the configurations of the image pickup devices 11 and 12 will be described with reference to FIG.

The image pickup device 11 includes a first image pickup device section 11a having a first number of pixels for picking up the subject, and an SD image corresponding to the first image taken by the first image pickup device section 11a. The SD image storage unit 11b that stores the image and a camera adjustment unit 11c that adjusts, for example, a physical position, an angle, a zoom ratio, and the like according to a control signal supplied from a marker position determination unit 14 described later.

The image pickup device 12 has a second image pickup device section 12a having a second number of pixels for picking up the image of the subject.
And an HD image storage unit 12b that stores an HD image corresponding to the second image captured by the second image sensor unit 12a,
A camera adjustment unit 12c that adjusts the camera according to the control signal supplied from the marker position determination unit 14.

SD photographed by the image pickup devices 11 and 12
The image and the HD image are the SD image storage unit 11b and the H image, respectively.
It is input to the marker detection unit 13 via the D image storage unit 12b. The marker detection unit 13 uses the SD for the same subject.
A marker indicating the reference pixel position is detected from the image data of the image and the HD image. This marker is recorded together when the subject is photographed. The marker detected by the marker detection unit 13 is supplied to the marker position determination unit 14. The marker position determination unit 14 determines the marker position based on the detection result of the marker detection unit 13.

In this way, the marker position is determined in such a manner that the target area when the image feature amount is obtained has different image formats, for example, if one is an SD signal and the other is an HD signal, one frame / This is because the image content will be different if the entire one field is targeted.

The marker will be described below with reference to FIGS. 4 and 5.

As the marker, for example, a marker area 20 as shown in FIG. 4 in which a black line crosses a white background is used. As shown in (a) and (b) of FIG. 5, this marker is set at four locations M1 in the horizontal and vertical directions on the subject position.
Attached to M2, M3 and M4, the markers M1, M2 and M3
And images including M4 are captured. Each marker is arranged in the image frame 21 above the subject. Markers M1 and M2
Are arranged at the same height in the horizontal direction, and the markers M3 and M4 are arranged in the same column, that is, in the vertical direction.

In the HD image, for example, as shown in FIG. 5A, w1 is set between the intersections of the horizontal markers M1 and M2, and h1 is set between the intersections of the vertical markers M3 and M4. Further, in the SD image, for example, as shown in FIG. 5B, the distance between the intersections of the horizontal markers M1 and M2 is w2,
The distance between the intersections of the vertical markers M3 and M4 is h2, and the angle of view format is different from that of the HD image, so that the positions are set to different positions.

In this way, the same subject 16 is imaged so that the horizontality of the marker including the marker in the horizontal direction and the verticality of the marker in the vertical direction can be obtained. If the intersection of the cross marker is detected and the effective area is extracted from the coordinates, the two images become the same image.

As described above, the image pickup apparatus 1 is used by using the marker.
The first image signal and the second image signal obtained by imaging the same subject 16 by 1 or 12 are input terminals 1a, 1b.
Is supplied to the image timing coincidence detection apparatus of this embodiment.

The average luminance value calculation circuit 2 calculates the average luminance value m _S of the SD image signal which is the first image signal and the average luminance value m _H of the HD image signal which is the second image signal. If the number of pixels in the effective area of the SD image signal is N _S ,
The average brightness value m _S is

[0036]

[Equation 1]

[0037] Here, Y _Si is the i-th luminance value within the number of effective area pixels N _S of the SD image signal, and is the same in the following.

When the number of pixels in the effective area of the HD image signal is N _H , the average luminance value m _H is

[0039]

[Equation 2]

It becomes Here, Y _Hi is the i-th luminance value within the number N _H of effective area pixels of the HD image signal, and is the same in the following.

The average luminance values m _S and m _H calculated by the average luminance value calculation circuit 2 are supplied to the time series value storage unit 3,
The time-series value storage unit 3 holds the time-series data for a fixed time.

The standard deviation calculating circuit 4 calculates the standard deviation σ _S of the SD image signal and the standard deviation σ _H of the HD image signal. When the number of pixels in the effective area of the SD image signal is N _S , the standard deviation σ _S is

[0043]

[Equation 3]

It becomes

Similarly, when the number of pixels in the effective area of the HD image signal is N _H , the standard deviation σ _H is

[0046]

[Equation 4]

It becomes

The standard deviations σ _S and σ _H calculated by the standard deviation calculating circuit 4 are supplied to the time-series value storage unit 5 and stored in the time-series value storage unit 5 as time-series data for the matching time. To be done.

The interframe difference calculation circuit 6 calculates the interframe difference Δ _S of the SD image signal and the inter frame difference Δ _H of the HD image signal. Assuming that the number of pixels in the effective area of the SD image signal is N _S , the inter-frame difference Δ _S is

[0050]

[Equation 5]

It becomes Here, Y _Si ^(t) represents the i-th luminance value of the t frame of the SD image signal, and
_Si ^(t-1) represents the i-th luminance value of the (t-1) frame of the SD image signal.

Similarly, assuming that the number of pixels in the effective area of the HD image signal is N _H , the inter-frame difference Δ _H is

[0053]

[Equation 6]

It becomes Here, Y _Hi ^(t) represents the i-th luminance value of the t frame of the HD image signal, and
_Hi ^(t-1) represents the i-th luminance value of the (t-1) frame of the HD image signal.

The inter-frame differences Δ _S and Δ _H calculated by the inter-frame difference calculation circuit 6 are supplied to the time-series value storage unit 7, and the time-series value storage unit 7 stores the time-series data for a fixed time. Is held as.

The motion vector calculation circuit 8 calculates the motion vector (v _Sx , v _Sy ) of the SD image signal and the motion vector (v _Hx , v _Hy ) of the HD image signal. These motion vectors are obtained by full search of the block matching method.

The motion vectors (v _Sx , v _Sy ) and (v _Hx , v _Hy ) calculated by the motion vector calculation circuit 8 are supplied to the time series value storage unit 9 and are stored in the time series value storage unit 9. Are stored as time series data for a certain time.

The time-direction matching circuit 10 uses the time-series data of various characteristic amounts held in the time-series value storage units 3, 5, 7, and 9 to perform timing coincidence detection by matching. The timing coincidence detection processing of the time direction matching circuit 10 will be described with reference to the flowchart of FIG.

First, in step S1, time series fluctuations are calculated from the time series data of each feature quantity stored in the time series value storage units 3, 5, 7 and 9. For example, the time-series fluctuation of the SD image signal is as shown in FIG. Further, for example, the time-series fluctuation of the HD image signal is as shown in FIG.

Next, in step S2, a data string for a certain time within a constant time width T in the time series variation of the SD image signal as shown in FIG. 7A is set as a target data string, A range in which more data than the target data string is taken within the time width T in the time series variation of the HD image signal as shown in FIG. 7B is set as the search range S _H , and the search range S _{From H} , reference the same number of data strings as the target data string.

In step S3, it is determined whether or not the search for the same number of data strings as the target data string has been completed within the search range S _H. For the search here, the processing shown in steps S4 and S5 is performed each time, and the optimum position is determined. Then, when the search is finally ended, the process proceeds to step S6 to output the timing matching image.

If it is determined that the search is not completed in step S3, the process proceeds to step S4, and the evaluation by the correlation coefficient is performed.

For example, the average luminance values m _S and m _H shown in the above equations (1) and (2) are higher than the levels D ₁ and D ₂ as shown in (A) and (B) of FIG. Otherwise, if the waveforms are almost the same, then matching can be achieved by only comparing the shapes of the waveforms. The same applies to the standard deviation σ and the interframe difference Δ. However, when the average luminance m, the standard deviation σ, and the inter-frame difference Δ change with time as shown in FIGS. 8A and 8B, the characteristics of the SD image signal and the HD image signal The difference makes it difficult to match.
Therefore, in this embodiment, each feature amount is normalized and evaluated. As an evaluation function that can be normalized, a correlation coefficient is used in step S4. This evaluation function NCCF (i)
Is

[0064]

[Equation 7]

It becomes Here, N represents the number of pixels in the effective area. Further, X _S is the average luminance value m _S of the SD image signal, a standard deviation sigma _S, the difference delta _S between frames, X _H is the average luminance value m _H of the HD image signal, a standard deviation sigma _H, frame difference delta _H Is represented. Therefore, the evaluation function NCC shown in equation (7)
F (i) is applied to the average value, standard deviation, and feature amount of inter-frame difference.

Note that another evaluation function is required to evaluate the motion vector. The motion vector normalization is performed by the ratio of the number of horizontal and vertical pixels of each effective area. That is, this evaluation function VNRM is

[0067]

[Equation 8]

It becomes Here, M is the ratio of the number of horizontal and vertical pixels in the effective area. Further, v _x ^(t) represents a motion vector in the x direction in the t frame of the SD image signal or the HD image signal, and v _x ^(t-1) in the (t-1) frame of the SD image signal or the HD image signal. It represents a motion vector in the x direction. Further, v _y ^(t) is ^t of the SD image signal or the HD image signal
Represents the motion vector in the y direction in the frame, v _y
^(t-1) represents the y-direction motion vector in the (t-1) frame of the SD image signal or the HD image signal.

Then, in order to improve the accuracy of the timing coincidence detection, the above-mentioned characteristic amounts, that is, the average luminance value m, the standard deviation σ, the interframe difference Δ and the motion vector v are comprehensively determined. The evaluation values of the average brightness value m, the standard deviation σ, and the interframe difference Δ have a maximum value of 1, and the closer to 1, the more optimal the solution. For example, when the evaluation values of the average brightness value m, the standard deviation σ, and the inter-frame difference Δ are as shown in FIG. 8A, the evaluation value approaches 1 and is said to be correlated.
On the other hand, in the case as shown in FIG. 8B, the evaluation value is 1
It is said that there is no correlation because it moves away from On the other hand, the motion vector v becomes optimal as the evaluation value is smaller. Therefore, in FIG.
Between (A) of FIG. 9 and (B) of FIG.
The evaluation value of the motion vector v of the image signal becomes smaller, and it is determined that the two match. On the other hand, between (A) of FIG. 10 and (B) of FIG. 10, the evaluation value of the motion vector v of the SD image signal and the HD image signal becomes large, and it is determined that they do not match. For example, the motion vector of one pixel of the SD image signal of FIG. 9A is (2, 3), and the motion vector of the pixel of FIG. 9B of the HD image signal corresponding to this pixel is also ( If 2, 3), then V in the above equation (8)
The NRM becomes 0. On the other hand, the motion vector of one pixel of the SD image signal of FIG. 10A is (1, 2), the motion vector of the pixel of the HD image corresponding to this pixel is (0, 0), and When the motion vector of the pixel of the unsupported HD image is (-1, -2), VNRM
Becomes 10.

Here, the above-described feature amount, that is, the average luminance value m, the standard deviation σ, the inter-frame difference Δ, and the motion vector v are comprehensively determined by normalizing the motion vector v and other feature amounts. I need.

The average brightness value m, the standard deviation σ, the interframe difference Δ, and the evaluation value of the motion vector v are respectively E _m ,
If E _S , E _f and E _v , then the overall evaluation value ESTM is

[0072]

[Equation 9]

It becomes However, E _v (MIN) is evaluated by the minimum value in the appearing E _v .

Then, the process proceeds to step S5, the optimum position is determined, and steps S3 to S5 are repeated until it is determined that the search end in step S3 is YES.

If the search end judgment in step S3 is YES, the process proceeds to step S6, in which the image recording device 12 is set within the search range S _H so that the position where the above expression (9) becomes maximum becomes the start point of the timing coincidence. Read out HD image data,
Images with the same timing are output.

As described above, the image timing coincidence detecting apparatus of the present embodiment determines the characteristic amount of an image when the SD image and the HD image captured by the image capturing apparatus having different input systems cannot be time-matched by the time code. It is possible to detect a timed position by extracting and holding the time variation as time series data and matching the time series data with each other.

In the present invention, the image formats do not have to match, but the field frequencies may be different when the formats are different. Therefore, the field frequencies are almost the same within a short period of time. It is based on assumptions.

[0078]

According to the image timing coincidence detecting method of the present invention, the first image output by the first image sensor having the first number of pixels and the second image sensor having the second number of pixels are output. An image timing coincidence detecting method for detecting coincidence in time direction with a second image to be output, comprising: an average value, a standard deviation of unit images of the first image and the second image,
Since at least one of the time difference and the amount of motion is used to detect the coincidence of the timings in the time direction between the images, it is possible to perform imaging using a plurality of image capturing devices while eliminating the time code recording. It is possible to detect the coincidence of the timing in the time direction of the captured image.

Further, in the image timing coincidence detection method according to the present invention, the first image output by the first image sensor having the first number of pixels and the second image sensor having the second number of pixels are output. An image timing coincidence detecting method for detecting coincidence in time direction with a second image to be output, comprising: an average value, a standard deviation, a time difference, and a motion amount of a unit image of the first image and the second image. Since the evaluation using the correlation coefficient for the time-series change value is performed, and the coincidence of the timing in the time direction between the images is detected by comprehensively judging the evaluation value,
It is possible to detect the coincidence of the timings of the images in the time direction without the need to record the time code.

Further, in the image timing coincidence detecting apparatus according to the present invention, the first image output by the first image sensor having the first number of pixels and the second image sensor having the second number of pixels are provided. In an image timing coincidence detection apparatus for detecting coincidence in time direction with a second image to be output, an average value, a standard deviation, a time difference, and a motion amount of a unit image of the first image and the second image are detected. Since the evaluation using the correlation coefficient for the time-series change value is performed, and the coincidence of the timing in the time direction between the images is detected by comprehensively judging the evaluation value,
It is possible to detect the coincidence of the timings of the images in the time direction without the need to record the time code.

[Brief description of drawings]

FIG. 1 is a block diagram showing a schematic configuration of an image timing coincidence detection apparatus according to an exemplary embodiment of the present invention.

FIG. 2 is a layout diagram of two image pickup devices that supply images to the image timing coincidence detection device according to the present exemplary embodiment.

FIG. 3 is a configuration diagram of two image pickup devices that supply images to the image timing coincidence detection device according to the present exemplary embodiment.

FIG. 4 is a schematic diagram of a marker used when the two image capturing devices shown in FIG. 3 capture an image of a subject.

FIG. 5 is a diagram for explaining setting of a target area using a marker.

FIG. 6 is a flowchart showing processing of a time direction matching circuit of the image timing coincidence detection apparatus of the present exemplary embodiment shown in FIG.

FIG. 7 is a diagram showing time-series fluctuations of an SD image signal and an HD image signal.

FIG. 8 is a diagram showing time characteristics of average luminance, standard deviation, and inter-frame difference, which are characteristic amounts of SD and HD images.

FIG. 9 is a diagram for explaining a case where an evaluation value of a motion vector, which is one of the feature quantities, is small.

FIG. 10 is a diagram for explaining a case where the evaluation value of a motion vector, which is one of the feature quantities, is large.

[Explanation of symbols]

2 Average brightness calculation circuit 4 Standard deviation calculation circuit 6-frame difference calculation circuit 8 Motion vector calculation circuit 3, 5, 7, 9 time series value storage 10 hour direction matching circuit

─────────────────────────────────────────────────── ─── Continuation of front page (58) Fields surveyed (Int.Cl. ⁷ , DB name) H04N 5/232

Claims (10)

(57) [Claims] 1. A coincidence in time direction between a first image output by a first image sensor having a first number of pixels and a second image output by a second image sensor having a second number of pixels. In the image timing coincidence detection method for detecting an image, at least one of an average value, a standard deviation, a time difference, and a motion amount in the unit image of the first image and the second image.
An image timing coincidence detection method, characterized in that the coincidence of the timing in the time direction between the images is detected using two pieces of information. 2. The information includes the first image and the second image.
The image timing coincidence detection method according to claim 1, wherein the method is at least one of an average value, a standard deviation, a time difference, and a time-series change value of a motion amount in the unit image. 3. The method according to claim 1, wherein the average value, standard deviation, time difference, and motion amount information is evaluated using a correlation coefficient. 4. A time-direction agreement between a first image output by a first image sensor having a first number of pixels and a second image output by a second image sensor having a second number of pixels. In the method for detecting the timing coincidence of images, the evaluation using the correlation coefficient for the time series change value of the average value, standard deviation, time difference, and motion amount in the unit image of the first image and the second image. And detecting the coincidence of the timing in the time direction between the images by comprehensively judging the evaluation value. 5. A time-direction agreement between a first image output by a first image sensor having a first number of pixels and a second image output by a second image sensor having a second number of pixels. In the image timing coincidence detection apparatus for detecting the image, at least one of the average value, the standard deviation, the time difference, and the motion amount in the unit image of the first image and the second image.
An image timing coincidence detection apparatus, characterized in that the coincidence of the timings in the time direction between the images is detected using two pieces of information. 6. The information includes the first image and the second image.
6. The image timing coincidence detection device according to claim 5, wherein the image timing coincidence detection device is at least one of an average value, a standard deviation, a time difference, and a time-series change value of a motion amount in the unit image. 7. The image timing coincidence detecting apparatus according to claim 5, wherein the average value, standard deviation, time difference, and motion amount information are evaluated using a correlation coefficient. 8. A time-direction agreement between a first image output by a first image sensor having a first number of pixels and a second image output by a second image sensor having a second number of pixels. In an image timing coincidence detection apparatus for detecting an image, an evaluation using a correlation coefficient for a time series change value of an average value, a standard deviation, a time difference, and a motion amount in a unit image of the first image and the second image. And an image timing coincidence detecting apparatus which detects the coincidence of timings in the time direction between the images by comprehensively judging the evaluation value. 9. The image timing coincidence detection apparatus according to claim 5, wherein the first and second image pickup devices pick up the same subject. 10. The image timing coincidence detecting apparatus according to claim 9, wherein the subject is provided with a marker indicating a reference pixel position.