JP2021071464A - Detection device and detection method - Google Patents

Abstract

To provide a detection device that can detect a vibration with a comparatively short cycle.SOLUTION: A detection device 1 includes an imaging unit 13 and a detection unit 14. The imaging unit photographs an object TG to output an image. The image includes a plurality of row pixel groups having different positions from each other in a first direction. Each of the row pixel groups has a plurality of different positions from each other in a second direction orthogonal to the first direction and includes a plurality of pixels that represent an amount of light incident in a processing period allocated to the row pixel group. In the output image, the detection unit acquires boundary positions, which are positions of a boundary among a plurality of areas in which amounts of light represented by pixels included in the image are different from each other and which are adjacent to each other for each of the row pixel groups, and detects a vibration of the object based on the plurality of acquired boundary positions.SELECTED DRAWING: Figure 1

Description

The present invention relates to a detection device and a detection method.

A detection device is known that includes an imaging unit that outputs an image by photographing an object and detects vibration of the object based on the output image. For example, the detection device described in Patent Document 1 detects the vibration of an object based on a plurality of images output by photographing the object at a plurality of different time points.

Japanese Unexamined Patent Publication No. 2003-156389

As a photographing method used by the imaging unit, a global shutter method and a rolling shutter method are known. In the global shutter method, all the pixels included in the image represent the amount of light incident in the same processing period. On the other hand, in the rolling shutter method, a plurality of row pixel groups included in the image and having a plurality of positions different from each other in the column direction represent the amount of light incident on each of the plurality of processing periods different from each other. Each row pixel group includes a plurality of pixels each having a plurality of positions different from each other in the row direction.

In the imaging unit using the global shutter method, if the imaging interval, which is the interval between two time points capable of continuous imaging, can be sufficiently shortened, vibration having a relatively short period can be detected. However, in the global shutter method, it is often difficult to shorten the shooting interval. Therefore, it is conceivable to use the rolling shutter method. However, the shooting interval tends to be longer in the rolling shutter method than in the global shutter method. Therefore, when the rolling shutter method is used in the above detection device, there is a problem that vibration having a relatively short period cannot be detected. It should be noted that this kind of problem can occur similarly when a method different from the rolling shutter method is used.

One of the objects of the present invention is to detect vibration having a relatively short period.

On one side, the detector
An image is output by photographing an object, and the image includes a plurality of row pixel groups having a plurality of positions different from each other in the first direction, and a plurality of processing periods different from each other in the plurality of row pixel groups. Are allotted, and each of the plurality of row pixel groups has a plurality of positions different from each other in the second direction orthogonal to the first direction, and in the processing period assigned to the row pixel group. An image pickup unit containing a plurality of pixels representing the amount of incident light,
In the output image, the amount of light represented by the pixels included in the image is different from each other, and the boundary position, which is the position of the boundary of a plurality of regions adjacent to each other, is acquired for each of the plurality of row pixel groups. Then, the detection unit that detects the vibration of the object based on the acquired plurality of boundary positions, and the detection unit.
To be equipped.

In one other aspect, the detection method
An image is output by photographing an object, and the image includes a plurality of row pixel groups having a plurality of positions different from each other in the first direction, and a plurality of processing periods different from each other in the plurality of row pixel groups. Are allotted, and each of the plurality of row pixel groups has a plurality of positions different from each other in the second direction orthogonal to the first direction, and in the processing period assigned to the row pixel group. Contains multiple pixels representing the amount of incident light
In the output image, the amount of light represented by the pixels included in the image is different from each other, and the boundary position, which is the position of the boundary of a plurality of regions adjacent to each other, is acquired for each of the plurality of row pixel groups. Then, the vibration of the object is detected based on the obtained plurality of boundary positions.
Including that.

Vibrations with a relatively short period can be detected.

It is a block diagram which shows the structure of the detection apparatus of 1st Embodiment. It is a figure which shows the structure of the marker part of 1st Embodiment. It is a block diagram which shows the structure of the detection part of 1st Embodiment. It is a block diagram which shows the function of the detection part of 1st Embodiment. It is a figure explaining the function which the detection part of 1st Embodiment detects the vibration in the x-axis direction when the object is vibrating. It is a figure explaining the function which the detection part of 1st Embodiment detects a vibration in a y-axis direction when an object is vibrating. It is a figure explaining the relationship between the distance between a marker part and an image pickup part, and the size of the area in a marker part in an image, when an object fluctuates in the z-axis direction. It is a figure explaining the function which the detection part of 1st Embodiment detects the vibration in the z-axis direction when an object is vibrating. It is a flowchart which shows the process for detecting the vibration in the x-axis direction by the detection apparatus of 1st Embodiment. It is a flowchart which shows the process for detecting the vibration in the y-axis direction by the detection apparatus of 1st Embodiment. It is a flowchart which shows the process for detecting the vibration in the z-axis direction by the detection apparatus of 1st Embodiment. It is a figure which shows the structure of the marker part of the 2nd Embodiment. It is a block diagram which shows the function of the detection part of 2nd Embodiment. It is a flowchart which shows the process for detecting the vibration in the x-axis direction by the detection apparatus of 2nd Embodiment. It is a figure which shows the structure of the marker part of the 1st modification of 2nd Embodiment. It is a figure which shows the structure of the marker part of 3rd Embodiment. It is a block diagram which shows the function of the detection part of 3rd Embodiment. It is a figure explaining the function which the detection part of 3rd Embodiment detects the vibration in the y-axis direction when the object is vibrating. It is a flowchart which shows the process for detecting the vibration in the y-axis direction by the detection apparatus of 3rd Embodiment. It is a flowchart which shows the process for detecting the vibration in the x-axis direction by the detection apparatus of 3rd Embodiment. It is a figure which shows the structure of the marker part of the 1st modification of 3rd Embodiment. It is a block diagram which shows the function of the detection part of the 1st modification of 3rd Embodiment.

For example, in the rolling shutter method, the entire image does not represent the amount of light incident in one processing period, but represents the amount of light incident in different processing periods for each row pixel group. Therefore, when an object moving at high speed is photographed, the amount of movement of the object differs for each row pixel group, so that the object in the image is distorted.

On the other hand, in the global shutter method, the entire image represents the amount of light incident in one processing period. Therefore, even when an object moving at high speed is photographed, the object in the image is not distorted. For this reason, when photographing an object moving at high speed, an imaging unit using a global shutter method is often used.

However, in the global shutter method, when an object moving sufficiently fast with respect to the shooting interval is photographed, the movement of the object cannot be detected. By the way, the shorter the shooting interval, the higher the manufacturing cost of the imaging unit tends to be. Therefore, in the global shutter method, it is difficult to detect the vibration of the object with high accuracy while suppressing the manufacturing cost.

By the way, the interval of the processing period for each row pixel group used in the rolling shutter method is relatively short. Therefore, as a result of diligent studies, the inventors have come up with a detection device that detects the vibration of an object based on an image showing the amount of light incident on each row pixel group in a different processing period.

Hereinafter, embodiments of the present invention relating to the detection device and the detection method will be described with reference to FIGS. 1 to 22.

<First Embodiment>
(Overview)
The detection device of the first embodiment includes an imaging unit and a detection unit.
The image pickup unit outputs an image by photographing the object.
The image includes a plurality of row pixel groups each having a plurality of positions different from each other in the first direction. A plurality of processing periods different from each other are assigned to the plurality of row pixel groups. Each of the plurality of row pixel groups has a plurality of positions different from each other in the second direction orthogonal to the first direction, and a plurality of light rays incident on the row pixel group during the processing period assigned to the row pixel group. Includes pixels.

In the output image, the detection unit sets the boundary position, which is the position of the boundary of a plurality of regions adjacent to each other while the amount of light represented by the pixels contained in the image is different from each other, with respect to each of the plurality of row pixel groups. And detect the vibration of the object based on the acquired multiple boundary positions.

According to this, the plurality of row pixel groups included in the image represent the amount of light incident on each of the plurality of processing periods different from each other. By the way, the interval between two consecutive processing periods is shorter than the imaging interval. Therefore, the plurality of boundary positions acquired for each of the plurality of row pixel groups can reflect the vibration having a sufficiently short period with respect to the shooting interval. As a result, according to the detection device, it is possible to detect vibration having a period sufficiently short with respect to the shooting interval.
Next, the detection device of the first embodiment will be described in more detail.

(Constitution)
As shown in FIG. 1, the detection device 1 of the first embodiment will be described using a right-handed Cartesian coordinate system having an x-axis, a y-axis, and a z-axis. Hereinafter, the same Cartesian coordinate system as in FIG. 1 is used in other figures.
The detection device 1 includes a marker unit 11, a light source unit 12, an imaging unit 13, and a detection unit 14.

The marker portion 11 has a sheet shape or a flat plate shape. In this example, the marker portion 11 forms a plane (in other words, an xy plane) orthogonal to the z-axis direction.
The marker portion 11 is fixed to the object TG. For example, the object TG is a machine, an instrument, an apparatus, an apparatus, or the like. For example, the object TG is a machine driven by power.
In this example, the marker portion 11 is fixed using an adhesive. The marker portion 11 may be removably fixed. Further, the marker portion 11 may be fixed by using a screw, a nail, or a fixing tool such as a bolt and a nut. In this example, the marker unit 11 is composed of one component. The marker unit 11 may be composed of a plurality of components. In this case, the plurality of components may be fixed at a plurality of positions different from each other.

As shown in FIG. 2, the marker portion 11 has a square shape. The marker portion 11 may have a shape different from the square shape (for example, a rectangular shape, a circular shape, an elliptical shape, a polygonal shape, or the like).

The marker unit 11 forms a plurality of regions adjacent to each other while the amount of reflected light is different from each other. In this example, the marker unit 11 has a first region 111 and a second region 112 adjacent to the first region 111 so as to be surrounded by the first region 111. The first region 111 reflects a larger amount of light than the second region 112.

In this example, the first region 111 is made of a retroreflective material. In this example, the second region 112 is made of a material different from the retroreflective material (eg, paper, cloth, resin, metal, ceramics, etc.). In this example, the second region 112 has a black color. The second region 112 may have a color different from black.
Further, the first region 111 does not have to be made of a retroreflective material. In this case, for example, the first region 111 may have white color, and the second region 112 may have a color different from white (for example, black).
Further, at least one of the first region 111 and the second region 112 may be the surface of the object TG. For example, the second region 112 may be the surface of the object TG.

The second region 112 is linear with a predetermined width. In this example, the second region 112 is N-shaped. The second region 112 includes a first straight line portion 1121, a second straight line portion 1122, and a third straight line portion 1123.

The first straight line portion 1121 extends in the y-axis direction at the end portion of the marker portion 11 in the negative direction of the x-axis. Therefore, the boundary between the first straight line portion 1121 of the second region 112 and the first region 111 forms a straight line extending in the y-axis direction.

The second straight line portion 1122 extends in the y-axis direction at the end portion of the marker portion 11 in the positive direction of the x-axis. Therefore, the boundary between the second straight line portion 1122 of the second region 112 and the first region 111 forms a straight line extending in the y-axis direction.

As described above, the boundary between the first straight line portion 1121 and the first region 111 of the second region 112 and the boundary between the second straight line portion 1122 and the first region 111 of the second region 112 are different from those in the x-axis direction. It forms a pair of straight lines that extend in the direction (in this example, the y-axis direction) and are parallel to each other.

The third straight line portion 1123 extends in a direction inclined with respect to the y-axis direction. Therefore, the boundary between the third straight line portion 1123 and the first region 111 of the second region 112 forms an inclined straight line which is a straight line inclined with respect to the y-axis direction. In this example, the third straight line portion 1123 includes the end portion of the first straight line portion 1121 in the positive direction of the y-axis and the end portion of the second straight line portion 1122 in the negative direction of the y-axis. Extend to connect.

The light source unit 12 generates light that irradiates the marker unit 11. In this example, the light source unit 12 includes an LED (Light Emitting Diode). The light source unit 12 may include a light source different from that of the LED. In this example, the light source unit 12 is located in the vicinity of the imaging unit 13. The number of light source units 12 included in the detection device 1 may be two or more. Further, the light source unit 12 may be located in the vicinity of the marker unit 11. Further, the marker unit 11 may include a light source unit 12. In this case, the first region 111 of the marker unit 11 may be configured to transmit light.

The imaging unit 13 outputs an image by photographing at least a part of the object TG including the marker unit 11. In this example, the image pickup unit 13 includes a solid-state image sensor (in other words, a CMOS image sensor) using CMOS (Complementary Metal-Oxide-Semiconductor).

The image output by the imaging unit 13 includes M row pixel groups having M positions different from each other in the y-axis direction. In this example, the M row pixel groups are located at equal intervals in the y-axis direction. M represents an integer of 2 or more (2048 in this example).
Each of the M row pixel groups includes N pixels having a plurality of positions different from each other in the x-axis direction orthogonal to the y-axis direction. N represents an integer of 2 or more (2048 in this example). In this example, the N pixels included in each row pixel group are located at equal intervals in the x-axis direction.

Therefore, in this example, the number of pixels of the image output by the imaging unit 13 is approximately 4 megapixels. For example, the number of pixels of the image output by the imaging unit 13 may be several megapixels.

Each of the M row pixel groups is assigned M different processing periods. In this example, in the M row pixel group, the time point at which the processing period starts is delayed by a predetermined delay time in the order in which the M row pixel groups are arranged in the positive direction of the y-axis. , M processing periods are assigned respectively.

In addition, in the M row pixel group, the time point at which the processing period starts is delayed by a predetermined delay time in the order in which the M row pixel group is arranged in the negative direction of the y-axis. Each processing period may be assigned. Further, in the M row pixel group, the time point at which the processing period starts is delayed by a predetermined delay time in an order different from the order in which the M row pixel groups are arranged in the y-axis direction. Each processing period may be assigned.

The N pixels included in each of the M row pixel groups represent the amount of light incident on the row pixel group during the processing period assigned to the row pixel group. In this example, each pixel has a value representing the brightness. In addition, each pixel may have a value representing lightness. Further, each pixel may have a value representing at least one intensity of red, blue, and green.
In this example, the method used by the imaging unit 13 to output an image may be expressed as a rolling shutter method.

In this example, the imaging interval, which is the interval between two time points during which the imaging unit 13 can continuously photograph, is approximately 33 milliseconds. In other words, the frame rate of the imaging unit 13 is 30 frames / sec. For example, the frame rate of the imaging unit 13 may be several tens of frames / sec.

The detection unit 14 acquires the boundary position for each of at least a part of the M row pixel group in the image output by the imaging unit 13. The boundary position is the position of the boundary of a plurality of regions adjacent to each other while the amount of light represented by the pixels included in the image is different from each other. The detection unit 14 detects the vibration of the object TG based on the acquired plurality of boundary positions. Details of the function of the detection unit 14 will be described later.

As shown in FIG. 3, the detection unit 14 includes a processing device 141, a storage device 142, and an output device 143 connected to each other via a bus BU.

The processing device 141 controls the storage device 142 and the output device 143 by executing the program stored in the storage device 142. As a result, the processing device 141 realizes the functions described later.

In this example, the processing device 141 is a CPU (Central Processing Unit). The processing device 141 may include an MPU (Micro Processing Unit) or a DSP (Digital Signal Processor) in place of the CPU or in addition to the CPU.

In this example, the storage device 142 includes a volatile memory and a non-volatile memory. For example, the storage device 142 includes at least one of a RAM (Random Access Memory), a ROM (Read Only Memory), a semiconductor memory, an organic memory, an HDD (Hard Disk Drive), and an SSD (Solid State Drive).

The output device 143 outputs information to the outside of the detection device 1. In this example, the output device 143 includes a display. The output device 143 may include a speaker.

(function)
As shown in FIG. 4, the functions of the detection unit 14 include the y-direction straight line boundary position acquisition unit 1401, the y-direction linear reference value acquisition unit 1402, the x-direction fluctuation amount acquisition unit 1403, and the spectrum analysis unit 1404. , Vibration detection unit 1405, inclined straight line boundary position acquisition unit 1406, inclined straight line reference value acquisition unit 1407, y-direction fluctuation amount acquisition unit 1408, parallel straight line boundary position acquisition unit 1409, distance reference value acquisition unit 1410. , Z-direction fluctuation amount acquisition unit 1411 and.

In this example, the detection unit 14 performs standardization processing and binarization processing on the image output by the imaging unit 13. In this example, the detection unit 14 refers to each unit 1401 to 1411 shown in FIG. 4 with respect to the image after the standardization processing and the binarization processing (in other words, the corrected image). Perform the function.

The standardization process is a process of standardizing the value represented by each of the MN pixels included in the image. In this example, the detection unit 14 acquires the maximum value among the MN values represented by the MN pixels, and the MN values represented by the MN pixels, respectively. Each of the above is standardized by multiplying the value obtained by dividing by the acquired maximum value by a predetermined coefficient (in this example, a value larger than 0, for example, 255). Normalization may be expressed as normalization.

In the binarization process, the detection unit 14 sets the value for each of the MN pixels included in the image when the value represented and standardized by the pixel is equal to or more than a predetermined correction threshold value. On the other hand, when the value represented and standardized by the pixel is smaller than the correction threshold value, the value is corrected to the second value. In this example, the first value is larger than the second value. The first value and the second value are predetermined.

In addition, instead of the binarization process, the detection unit 14 may perform an adaptive binarization process for determining a correction threshold value for each predetermined region included in the image. Further, the detection unit 14 does not have to perform at least one of the standardization process and the binarization process.

The y-direction linear boundary position acquisition unit 1401 includes both ends of the marker unit 11 in the y-axis direction (in this example, both ends of the first straight-line unit 1121 in the y-axis direction) from the M row pixel group. A group of P row pixels corresponding to the portion excluding the portion) is extracted. P represents an integer smaller than M.

The y-direction linear boundary position acquisition unit 1401 acquires the y-direction linear boundary position for each of the P row pixel groups based on the N pixels included in the row pixel group. In this example, the y-direction straight line boundary position corresponds to the position of the boundary between the end side of the first straight line portion 1121 in the negative direction of the x-axis and the second region 112.

The straight line boundary position in the y direction may correspond to the position of the boundary between the end side of the first straight line portion 1121 in the positive direction of the x-axis and the second region 112. Further, the y-direction straight line boundary position is the position of the boundary between the end side of the first straight line portion 1121 in the negative direction of the x-axis and the second region 112, and the x of the first straight line portion 1121. It may correspond to the position of the boundary between the end side in the positive direction of the axis and the second region 112, and the position of the midpoint. Further, the y-direction straight line boundary position acquisition unit 1401 may use the second straight line portion 1122 in place of the first straight line portion 1121 or in addition to the first straight line portion 1121.

In this example, the y-direction linear boundary position acquisition unit 1401 has a second value in which the second value of the N pixels included in the row pixel group is continuous with respect to each of the P row pixel groups. Within the boundary between the region and the first value region that is adjacent to the second value region at a position in the positive direction of the x-axis from the second value region and the first value is continuous, the x-axis The position of the first boundary from the end in the negative direction toward the positive direction of the x-axis is acquired as the linear boundary position in the y direction.

The y-direction straight line reference value acquisition unit 1402 acquires the y-direction straight line reference value for each of the P row pixel groups based on the P y-direction straight line boundary positions acquired by the y-direction straight line boundary position acquisition unit 1401. To do.

In this example, the y-direction linear reference value acquisition unit 1402 extends in the y-axis direction, and the total distance to the P y-direction linear boundary positions acquired by the y-direction linear boundary position acquisition unit 1401 is the minimum. Get a straight line as a reference straight line. The y-direction linear reference value acquisition unit 1402 extends in the y-axis direction, and the sum of the squares of the distances with respect to the P y-direction linear boundary positions acquired by the y-direction linear boundary position acquisition unit 1401 is the minimum. A straight line may be acquired as a reference straight line.

The y-direction straight line reference value acquisition unit 1402 determines the positions of the pixels intersecting the acquired reference line among the N pixels included in the row pixel group for each of the P row pixel groups. Obtained as a linear reference value in the y direction.

The x-direction fluctuation amount acquisition unit 1403 has P y-direction linear boundary positions acquired by the y-direction linear boundary position acquisition unit 1401 and P y-direction linear reference units acquired by the y-direction linear reference value acquisition unit 1402. Based on the value, the amount of variation in the x-direction for each of the P row pixel groups is acquired. In this example, the amount of fluctuation in the x direction is the difference in position between the y-direction straight line boundary position and the y-direction straight line reference value in the x-axis direction.

The spectrum analysis unit 1404 performs spectrum analysis on the P x-direction fluctuation amounts acquired by the x-direction fluctuation amount acquisition unit 1403. In this example, the position of each pixel in the y-axis direction corresponds to the time when the light on which the value represented by the pixel is based is incident on the imaging unit 13. Therefore, in this example, the change in the x-direction fluctuation amount in the y-axis direction corresponds to the change in the x-direction fluctuation amount with time.

For example, as shown in FIG. 5, it is assumed that the object TG vibrates in the x-axis direction. In this case, for example, the linear boundary position in the y direction is located on a wavy line that fluctuates in the x-axis direction as represented by the curve L11. Therefore, the change in the y-axis direction of the x-direction fluctuation amount, which is the difference between the y-direction straight line reference value on the reference straight line represented by the straight line L10 and the y-direction straight line boundary position in the x-axis direction, is a target. Reflects the vibration of the object TG in the x-axis direction with high accuracy.

Further, even when the object TG vibrates in the y-axis direction, the reference straight line extends in the y-axis direction, so that the vibration in the y-axis direction is not easily reflected in the amount of fluctuation in the x-direction. Therefore, the change in the amount of fluctuation in the x-direction in the y-axis direction can reflect the vibration of the object TG in the x-axis direction with high accuracy.

In this example, the spectrum analysis unit 1404 performs Fourier transform processing, wavelet transform processing, or the like on P x-direction fluctuation amounts, so that the x-direction fluctuation frequency is a frequency component of the change of the x-direction fluctuation amount with respect to time. Get the ingredients.

The vibration detection unit 1405 detects the vibration of the object TG in the x-axis direction based on the x-direction fluctuation frequency component acquired by the spectrum analysis unit 1404. For example, the vibration detection unit 1405 may detect that the object TG is vibrating in the x-axis direction when the intensity of the predetermined frequency component is equal to or higher than the predetermined threshold intensity.

When it is detected that the object TG is vibrating in the x-axis direction, the vibration detection unit 1405 outputs information indicating that the object TG is vibrating in the x-axis direction. The vibration detection unit 1405 may output the x-direction fluctuation frequency component acquired by the spectrum analysis unit 1404 in addition to or instead of the output of the information.

The inclined straight line boundary position acquisition unit 1406 is a portion of the marker unit 11 including both ends in the y-axis direction (in this example, both ends of the first straight line unit 1121 in the y-axis direction) from the group of M row pixels. ) Is excluded, and Q row pixel groups corresponding to the portion are extracted. Q represents an integer smaller than M. In this example, the integer represented by Q is equal to the integer represented by P. The integer represented by Q may be different from the integer represented by P.

The inclined straight line boundary position acquisition unit 1406 acquires the inclined straight line boundary position for each of the Q row pixel groups based on the N pixels included in the row pixel group. In this example, the inclined straight line boundary position corresponds to the position of the boundary between the end side of the third straight line portion 1123 in the negative direction of the x-axis and the second region 112.

The inclined straight line boundary position may correspond to the position of the boundary between the end side of the third straight line portion 1123 in the positive direction of the x-axis and the second region 112. Further, the inclined straight line boundary position is the position of the boundary between the end side in the negative direction of the x-axis of the third straight line portion 1123 and the second region 112, and the x-axis of the third straight line portion 1123. It may correspond to the position of the boundary between the end edge in the positive direction and the second region 112 and the position of the midpoint.

In this example, the inclined straight line boundary position acquisition unit 1406 has a second value region in which the second value of the N pixels included in the row pixel group is continuous with respect to each of the Q row pixel groups. And the negative of the x-axis within the boundary between the second value region and the first value region which is adjacent to the second value region at a position in the positive direction of the x-axis and the first value is continuous. The position of the second boundary from the end in the direction toward the positive direction of the x-axis is acquired as the inclined straight line boundary position.

The inclined straight line boundary position acquisition unit 1406 corrects the acquired Q inclined straight line boundary positions based on the x-direction fluctuation amount acquired by the x-direction fluctuation amount acquisition unit 1403. In this example, the inclined straight line boundary position acquisition unit 1406 obtains the acquired inclined straight line boundary position for each of the Q row pixel groups, and the x-direction fluctuation amount acquired by the x-direction fluctuation amount acquisition unit 1403 is used. The inclined straight line boundary position is corrected so as to move in the direction opposite to the expressed direction by the amount indicated by the x-direction fluctuation amount. In other words, the inclined straight line boundary position acquisition unit 1406 corrects the inclined straight line boundary position so as to cancel the amount of fluctuation in the x direction for each of the Q row pixel groups.

The inclined straight line reference value acquisition unit 1407 acquires the inclined straight line reference value for each of the Q row pixel groups based on the Q inclined straight line boundary positions corrected by the inclined straight line boundary position acquisition unit 1406.

In this example, the tilt straight line reference value acquisition unit 1407 extends in a predetermined tilt direction and draws a straight line having the minimum total distance to the Q tilt straight line boundary positions corrected by the tilt straight line boundary position acquisition unit 1406. Obtained as a reference straight line. The tilting direction is a direction of tilting with respect to the y-axis direction. In this example, the tilting direction is a direction that is tilted by 45 degrees with respect to the y-axis direction. The tilt straight line reference value acquisition unit 1407 extends in the tilt direction and uses a straight line having the smallest sum of squares of distances with respect to the Q tilt straight line boundary positions corrected by the tilt straight line boundary position acquisition unit 1406 as a reference straight line. May be obtained as.

The tilt straight line reference value acquisition unit 1407 tilts the positions of the pixels intersecting the acquired reference straight line among the N pixels included in the row pixel group with respect to each of the Q row pixel groups. Obtained as a straight line reference value.

The y-direction fluctuation amount acquisition unit 1408 includes Q tilt straight line boundary positions corrected by the tilt straight line boundary position acquisition unit 1406 and Q tilt straight line reference values acquired by the tilt straight reference value acquisition unit 1407. Based on this, the amount of fluctuation in the y direction for each of the Q row pixel groups is acquired.

In this example, the amount of fluctuation in the y direction is obtained by multiplying the difference in position between the inclined straight line boundary position and the inclined straight line reference value in the x-axis direction by the tangent of the angle formed by the acquired reference straight line and the x-axis direction. Value. In other words, the amount of fluctuation in the y direction corresponds to the amount of fluctuation in the x-axis direction of the Q inclined straight line boundary positions corrected by the inclined straight line boundary position acquisition unit 1406.

The spectrum analysis unit 1404 performs spectrum analysis on the Q y-direction fluctuation amounts acquired by the y-direction fluctuation amount acquisition unit 1408. In this example, the position of each pixel in the y-axis direction corresponds to the time when the light on which the value represented by the pixel is based is incident on the imaging unit 13. Therefore, in this example, the change in the y-direction fluctuation amount in the y-axis direction corresponds to the change in the y-direction fluctuation amount with time.

For example, as shown in FIG. 6, it is assumed that the object TG vibrates in the y-axis direction. In this case, for example, the inclined straight line boundary position is located on a wavy line that fluctuates in the x-axis direction as represented by the curve L21. Therefore, the change in the y-axis direction of the y-direction fluctuation amount representing the difference in position between the inclined straight line reference value on the reference straight line represented by the straight line L20 and the inclined straight line boundary position in the x-axis direction is the object TG. Reflects the vibration in the y-axis direction with high accuracy.

In this example, the spectrum analysis unit 1404 performs Fourier transform processing, wavelet transform processing, or the like on Q y-direction fluctuation amounts, so that the y-direction fluctuation frequency is a frequency component of the change of the y-direction fluctuation amount with respect to time. Get the ingredients.

The vibration detection unit 1405 detects the vibration of the object TG in the y-axis direction based on the y-direction fluctuation frequency component acquired by the spectrum analysis unit 1404. For example, the vibration detection unit 1405 may detect that the object TG is vibrating in the y-axis direction when the intensity of the predetermined frequency component is equal to or higher than the predetermined threshold intensity.

When it is detected that the object TG is vibrating in the y-axis direction, the vibration detection unit 1405 outputs information indicating that the object TG is vibrating in the y-axis direction. The vibration detection unit 1405 may output the y-direction fluctuation frequency component acquired by the spectrum analysis unit 1404 in addition to or instead of the output of the information.

The parallel straight line boundary position acquisition unit 1409 is a portion of the marker unit 11 including both ends in the y-axis direction (in this example, both ends of the first straight line unit 1121 in the y-axis direction) from the M row pixel group. ) Is extracted, and R row pixel groups corresponding to the portion excluding) are extracted. R represents an integer smaller than M. In this example, the integer represented by R is equal to the integer represented by P. The integer represented by R may be different from the integer represented by P.

The parallel straight line boundary position acquisition unit 1409 acquires a pair of parallel straight line boundary positions for each of the R row pixel groups based on the N pixels included in the row pixel group. In this example, the pair of parallel straight line boundary positions is the position of the boundary between the end side of the first straight line portion 1121 in the negative direction of the x-axis and the second region 112, and the second straight line portion 1122. It corresponds to the position of the boundary between the end edge in the negative direction of the x-axis and the second region 112.

The pair of parallel straight line boundary positions is the position of the boundary between the end side of the first straight line portion 1121 in the positive direction of the x-axis and the second region 112, and the second straight line portion 1122. , It may correspond to the position of the boundary between the end side in the positive direction of the x-axis and the second region 112.

Further, the pair of parallel straight line boundary positions are the positions of the boundaries between the end side of the first straight line portion 1121 in the negative direction of the x-axis and the second region 112, and the first straight line portion 1121. The position of the boundary between the end in the positive direction of the x-axis and the second region 112, the position of the midpoint, and the end of the second straight line portion 1122 in the negative direction of the x-axis, and the second Corresponds to the position of the boundary between the two regions 112, the end of the second straight line portion 1122 in the positive direction of the x-axis, the position of the boundary between the second region 112, and the position of the midpoint. You may.

In this example, the parallel straight line boundary position acquisition unit 1409 has a second value region in which the second value of the N pixels included in the row pixel group is continuous with respect to each of the R row pixel groups. And the negative of the x-axis within the boundary between the second value region and the first value region which is adjacent to the second value region at a position in the positive direction of the x-axis and the first value is continuous. The position of the first boundary and the position of the third boundary are acquired as a pair of parallel straight line boundary positions from the end in the direction toward the positive direction of the x-axis.

The parallel straight line boundary position acquisition unit 1409 acquires the acquired pair of parallel straight line boundary positions of the R group by the x-direction fluctuation amount acquired by the x-direction fluctuation amount acquisition unit 1403 and the y-direction fluctuation amount acquisition unit 1408. Correction is made based on the amount of fluctuation in the y direction.

In this example, the parallel straight line boundary position acquisition unit 1409 changes the acquired pair of parallel straight line boundary positions for each of the R row pixel groups in the x-direction, which is acquired by the x-direction fluctuation amount acquisition unit 1403. The pair of parallel straight line boundary positions are moved in the direction opposite to the direction indicated by the amount by the magnitude indicated by the x-direction fluctuation amount, and the y-direction fluctuation amount acquired by the y-direction fluctuation amount acquisition unit 1408 is set. Move by the amount represented by the value in the direction opposite to the direction represented by the value obtained by multiplying the margin of the angle formed by the reference straight line acquired by the y-direction fluctuation amount acquisition unit 1408 and the x-axis direction. , Correct the pair of parallel straight line boundary positions. In other words, in this example, the parallel straight line boundary position acquisition unit 1409 is the inclined straight line boundary from which the x-direction fluctuation amount and the y-direction fluctuation amount are acquired for each of the R row pixel groups. The pair of parallel straight line boundary positions are corrected so as to cancel the fluctuation amount of the position in the x-axis direction.

The distance reference value acquisition unit 1410 acquires the distance reference value for each of the R row pixel groups based on the pair of parallel straight line boundary positions of the R group corrected by the parallel straight line boundary position acquisition unit 1409.

In this example, the distance reference value acquisition unit 1410 determines the distance between the pair of parallel straight line boundary positions for each of the pair of parallel straight line boundary positions of the R group corrected by the parallel straight line boundary position acquisition unit 1409. get. The distance reference value acquisition unit 1410 acquires the average value of the acquired R distances as the distance reference value for each of the R row pixel groups.

The z-direction fluctuation amount acquisition unit 1411 acquires the distance between the pair of parallel straight line boundary positions for each of the pair of parallel straight line boundary positions of the R group corrected by the parallel straight line boundary position acquisition unit 1409. The z-direction fluctuation amount acquisition unit 1411 has the z-direction fluctuation amount for each of the R row pixel groups based on the acquired R distances and the distance reference value acquired by the distance reference value acquisition unit 1410. To get.

As shown in FIG. 7, it is assumed that the position of the marker portion 11 in the z-axis direction fluctuates from the reference position P0 to the fluctuating position P1. _{In this example, the distance h 1} between the fluctuating position P1 and the imaging unit 13 is longer than _{the distance h 0} between the reference position P0 and the imaging unit 13. Therefore, the size of the predetermined region VA in the marker portion 11 in the image is reduced by changing the position of the marker portion 11 from the reference position P0 to the fluctuation position P1.

_{The distance d 1} between the pair of parallel straight line boundary positions acquired when the position of the marker portion 11 is the variable position P1 is expressed by the mathematical formula 1. Here, d ₀ represents the distance between the pair of parallel straight line boundary positions acquired when the position of the marker portion 11 is the reference position P0.

z-direction variation amount Δz is the distance h ₀ between the reference position P0 and the imaging unit 13, which is the difference between the distance h _1, between the change position P1 and the imaging unit 13. Therefore, the amount of variation Δz in the z direction is expressed by the mathematical formula 2 using the mathematical formula 1.

In this example, the z-direction fluctuation amount acquisition unit 1411 is the distance between the marker unit 11 and the imaging unit 13 measured in a state where the object TG is not vibrating for each of the R row pixel groups. h _{0 , the distance reference value d 0} acquired by the distance reference value acquisition unit 1410 _{, the distance d 1} between the pair of parallel straight line boundary positions acquired for the row pixel group, the equation 2 and Based on, the amount of variation Δz in the z direction is acquired.

The spectrum analysis unit 1404 performs spectrum analysis on the R z-direction fluctuation amounts Δz acquired by the z-direction fluctuation amount acquisition unit 1411. In this example, the position of each pixel in the y-axis direction corresponds to the time when the light on which the value represented by the pixel is based is incident on the imaging unit 13. Therefore, in this example, the change in the z-direction fluctuation amount Δz in the y-axis direction corresponds to the change in the z-direction fluctuation amount Δz with respect to time.

For example, as shown in FIG. 8, it is assumed that the object TG vibrates in the z-axis direction. In this case, for example, the pair of parallel straight line boundary positions are on a pair of wavy lines in which _{the distance d 1 between the pair of parallel straight line boundary positions in the x-axis direction varies, as represented by the pair of curves L31.} To position. Therefore, the change in the z-direction fluctuation amount Δz obtained based on the distance reference value d _{0 , the distance d 1} between the pair of parallel straight line boundary positions, and the mathematical formula 2 in the y-axis direction is the object TG. Reflects the vibration in the z-axis direction with high accuracy.

In this example, the spectrum analysis unit 1404 performs Fourier transform processing, wavelet transform processing, or the like on R z-direction fluctuation amounts Δz, thereby performing z-direction fluctuation amount Δz, which is a frequency component of the change with time. Get the fluctuating frequency component.

The vibration detection unit 1405 detects the vibration of the object TG in the z-axis direction based on the z-direction fluctuation frequency component acquired by the spectrum analysis unit 1404. For example, the vibration detection unit 1405 may detect that the object TG is vibrating in the z-axis direction when the intensity of the predetermined frequency component is equal to or higher than the predetermined threshold intensity.

When it is detected that the object TG is vibrating in the z-axis direction, the vibration detection unit 1405 outputs information indicating that the object TG is vibrating in the z-axis direction. The vibration detection unit 1405 may output the z-direction fluctuation frequency component acquired by the spectrum analysis unit 1404 in addition to or instead of the output of the information.

In this example, the y-direction straight line boundary position, the inclined straight line boundary position, and the pair of parallel straight line boundary positions each correspond to the boundary position. In this example, the y-direction straight line reference value, the inclined straight line reference value, and the distance reference value each correspond to the reference value.

(motion)
Next, the operation of the detection device 1 will be described with reference to FIGS. 9 to 11.
The light source unit 12 generates light. The light generated by the light source unit 12 irradiates the marker unit 11. Next, the imaging unit 13 outputs an image by photographing at least a part of the object TG including the marker unit 11. Next, the detection unit 14 performs standardization processing and binarization processing on the image output by the imaging unit 13.

Next, the detection unit 14 executes the process represented by the flowchart in FIG. Hereinafter, the processing of FIG. 9 will be described.
The detection unit 14 acquires a linear boundary position in the y direction for each of the P row pixel groups based on the N pixels included in the row pixel group (step S101 in FIG. 9).

Next, the detection unit 14 acquires y-direction linear reference values for each of the P row pixel groups based on the P y-direction linear boundary positions acquired in step S101 (step S102 in FIG. 9). ..

Next, the detection unit 14 has P row pixels based on the P y-direction straight line boundary positions acquired in step S101 and the P y-direction linear reference values acquired in step S102. The amount of variation in the x-direction for each of the groups is acquired (step S103 in FIG. 9).

Next, the detection unit 14 acquires the x-direction fluctuation frequency component by performing Fourier transform processing, wavelet transform processing, or the like on the P x-direction fluctuation amounts acquired in step S103 (step S104 in FIG. 9). ).
Next, the detection unit 14 detects the vibration of the object TG in the x-axis direction based on the x-direction fluctuation frequency component acquired in step S104 (step S105 in FIG. 9).
In this way, the detection unit 14 executes the process shown in FIG.

Next, the detection unit 14 executes the process represented by the flowchart in FIG. Hereinafter, the processing of FIG. 10 will be described.
The detection unit 14 acquires the inclined straight line boundary position for each of the Q row pixel groups based on the N pixels included in the row pixel group (step S201 in FIG. 10).

Next, the detection unit 14 corrects the Q inclined straight line boundary positions acquired in step S201 based on the amount of fluctuation in the x direction acquired in step S103 of FIG. 9 (step S202 of FIG. 10).
Next, the detection unit 14 acquires the tilt straight line reference value for each of the Q row pixel groups based on the Q tilt straight line boundary positions corrected in step S202 (step S203 in FIG. 10).

Next, the detection unit 14 sets the Q row pixel group based on the Q tilt straight line boundary positions corrected in step S202 and the Q tilt straight line reference values acquired in step S203. The amount of fluctuation in the y direction for each is acquired (step S204 in FIG. 10).

Next, the detection unit 14 acquires the y-direction fluctuation frequency component by performing Fourier transform processing, wavelet transform processing, or the like on the Q y-direction fluctuation amounts acquired in step S204 (step S205 in FIG. 10). ).
Next, the detection unit 14 detects the vibration of the object TG in the y-axis direction based on the y-direction fluctuation frequency component acquired in step S205 (step S206 in FIG. 10).
In this way, the detection unit 14 executes the process shown in FIG.

Next, the detection unit 14 executes the process represented by the flowchart in FIG. Hereinafter, the processing of FIG. 11 will be described.
The detection unit 14 acquires a pair of parallel linear boundary positions for each of the R row pixel groups based on the N pixels included in the row pixel group (step S301 in FIG. 11).

Next, the detection unit 14 acquires the pair of parallel straight line boundary positions of the R group acquired in step S301 with the x-direction fluctuation amount acquired in step S103 of FIG. 9 and in step S202 of FIG. The correction is made based on the amount of fluctuation in the y-direction (step S302 in FIG. 11).
Next, the detection unit 14 acquires a distance reference value for each of the R row pixel groups based on the pair of parallel straight line boundary positions of the R group corrected in step S302 (step S303 in FIG. 11).

Next, the detection unit 14 has R row pixel groups based on the pair of parallel straight line boundary positions of the R set corrected in step S302 and the R distance reference values acquired in step S303. The amount of variation in the z-direction for each of the above is acquired (step S304 in FIG. 11).

Next, the detection unit 14 acquires the z-direction fluctuation frequency component by performing Fourier transform processing, wavelet transform processing, or the like on the R z-direction fluctuation amounts acquired in step S304 (step S305 in FIG. 11). ).
Next, the detection unit 14 detects the vibration of the object TG in the z-axis direction based on the z-direction fluctuation frequency component acquired in step S305 (step S306 in FIG. 11).
In this way, the detection unit 14 executes the process shown in FIG.

Next, when the detection unit 14 detects that the object TG vibrates in at least one of the x-axis direction, the y-axis direction, and the z-axis direction, the object TG vibrates in that direction. Outputs information indicating that the device is running.

As described above, the detection device 1 of the first embodiment includes an imaging unit 13 and a detection unit 14. The imaging unit 13 outputs an image by photographing the object TG. The image includes a plurality of row pixel groups each having a plurality of positions different from each other in the first direction (in this example, the y-axis direction). A plurality of processing periods different from each other are assigned to the plurality of row pixel groups. Each of the plurality of row pixel groups has a plurality of positions different from each other in the second direction (in this example, the x-axis direction) orthogonal to the first direction, and at the processing period assigned to the row pixel group. Includes a plurality of pixels representing the amount of incident light.

In the image output by the imaging unit 13, the detection unit 14 sets the boundary position, which is the position of the boundary of a plurality of regions adjacent to each other while the amount of light represented by the pixels included in the image is different from each other, in the plurality of rows. It is acquired for each of the pixel groups, and the vibration of the object TG is detected based on the acquired plurality of boundary positions.

According to this, the plurality of row pixel groups included in the image represent the amount of light incident on each of the plurality of processing periods different from each other. By the way, the interval between two consecutive processing periods is shorter than the imaging interval. Therefore, the plurality of boundary positions acquired for each of the plurality of row pixel groups can reflect the vibration having a sufficiently short period with respect to the shooting interval. As a result, according to the detection device 1, vibration having a period sufficiently short with respect to the photographing interval can be detected.

By the way, in order to detect vibration in an imaging unit using the global shutter method, it is known that the shooting interval needs to be 1/2 times or less of the period of the vibration according to the restriction of the Nyquist frequency. Therefore, for example, in order to detect vibration having a frequency of several kHz, an imaging unit having a frame rate close to 10,000 frames per second is required. Since it is often difficult to manufacture such an imaging unit, it is often expensive.

On the other hand, according to the detection device 1, for example, the vibration of the object TG can be detected by using an inexpensive and generally distributed industrial camera. Further, according to the detection device 1, for example, vibration having a frequency of several Hz to several kHz can be detected. Further, according to the detection device 1, for example, vibration having an amplitude of several tens of μm to several mm can be detected.

Further, for example, according to the detection device 1, the vibration of the object TG can be detected without wiring the object TG.

Further, in the detection device 1 of the first embodiment, the detection unit 14 acquires a reference value for at least a part of each of the plurality of row pixel groups based on at least a part of the acquired plurality of boundary positions. Then, the vibration of the object TG is detected based on the acquired reference value and the acquired plurality of boundary positions.

For example, when the position of the imaging unit 13 with respect to the object TG cannot be specified, the boundary position in the state where the object TG is not vibrating cannot be specified. Therefore, there is a possibility that the vibration of the object TG cannot be detected with high accuracy.
On the other hand, according to the detection device 1, it is possible to acquire a reference value for each row pixel group in a state where the object TG is not vibrating, based on the acquired plurality of boundary positions. Therefore, even when the position of the imaging unit 13 with respect to the object TG cannot be specified, the vibration of the object TG can be detected with high accuracy.
In the detection device 1 of the modified example of the first embodiment, the detection unit 14 may use a predetermined reference value without acquiring the reference value based on the acquired plurality of boundary positions.

Further, the detection device 1 of the first embodiment includes a marker unit 11 which is fixed to the object TG, has different amounts of reflected light, and forms a plurality of regions adjacent to each other. The imaging unit 13 outputs an image by photographing the marker unit 11. The detection unit 14 detects the vibration of the object TG.

The amount of light reflected by a plurality of regions adjacent to each other on the surface of the object TG may be substantially the same. In this case, there is a risk that the boundary position cannot be obtained with high accuracy.
On the other hand, according to the detection device 1, the marker portion 11 forming a plurality of regions in which the amount of reflected light is different from each other is fixed to the object TG. Therefore, the boundary position can be acquired with high accuracy by using the image output by photographing the marker unit 11. As a result, the vibration of the object TG can be detected with high accuracy.

Further, in the detection device 1 of the first embodiment, at least one of the plurality of regions of the marker unit 11 is made of a retroreflective material.

According to this, the amount of light reflected in the region made of the retroreflective material can be increased. As a result, since the boundary position can be acquired with high accuracy, the vibration of the object TG can be detected with high accuracy.

Further, in the detection device 1 of the first embodiment, the marker unit 11 forms a straight line in which at least a part of the boundary between the plurality of regions extends in the first direction (in this example, the y-axis direction). The detection unit 14 determines the object TG in the second direction based on the change in the first direction of the fluctuation amount in the second direction (in this example, the x-axis direction) of at least a part of the acquired plurality of boundary positions. Detects vibration.

The amount of fluctuation in the second direction of the boundary position of the boundary forming the straight line extending in the first direction reflects the vibration of the object TG in the second direction with high accuracy. Therefore, according to the detection device 1, the vibration of the object TG in the second direction can be detected with high accuracy.

Further, in the detection device 1 of the first embodiment, the marker unit 11 forms an inclined straight line in which at least a part of the boundary between the plurality of regions is a straight line inclined with respect to the first direction. The detection unit 14 detects the vibration of the object TG in the first direction based on the change in the first direction of the fluctuation amount in the second direction of at least a part of the acquired plurality of boundary positions.

The amount of fluctuation in the second direction of the boundary position of the boundary forming the inclined straight line reflects the vibration of the object TG in the first direction with high accuracy. Therefore, according to the detection device 1, the vibration of the object TG in the first direction can be detected with high accuracy.

Further, in the detection device 1 of the first embodiment, in the marker unit 11, at least a part of the boundary between the plurality of regions extends in a direction different from the second direction (in this example, the first direction) and is parallel to each other. Form a pair of straight lines. The detection unit 14 acquires two boundary positions for each of at least a part of the plurality of row pixel groups, acquires the distance between the acquired two boundary positions, and obtains the acquired distances of the plurality of distances. , The vibration of the object TG in the third direction (in this example, the z-axis direction) orthogonal to the first direction and the second direction is detected based on the change in the first direction.

The distance in the second direction between the boundary positions of the boundaries forming a pair of straight lines parallel to each other reflects the vibration of the object TG in the third direction with high accuracy. Therefore, according to the detection device 1, the vibration of the object TG in the third direction can be detected with high accuracy.

Further, in the detection device 1 of the first embodiment, the detection unit 14 corrects at least a part of the acquired plurality of boundary positions based on the amount of fluctuation in the second direction, and based on the corrected boundary position. , The vibration of the object TG in the first direction or the third direction is detected.

According to this, the vibration of the object TG in the first direction or the third direction can be detected with higher accuracy.

The detection device 1 of the modified example of the first embodiment does not have to include the light source unit 12.
Further, the detection device 1 of the modified example of the first embodiment does not have to include the marker unit 11. In this case, the detection device 1 detects the vibration of the object TG by using a part of the object TG (for example, a bolt, a nut, a tag, a plate, a design, etc.) instead of the marker portion 11. May be good.

Further, the detection device 1 of the modified example of the first embodiment detects the vibration of the object TG in at least one of the x-axis direction, the y-axis direction, and the z-axis direction, and also detects the vibration of the object TG in the other direction. It is not necessary to detect the vibration of the object TG.

Further, the detection device 1 of the modified example of the first embodiment includes a plurality of marker units 11 fixed to each of the plurality of object TGs, and the imaging unit 13 captures an image by photographing the plurality of marker units 11. It may be output and the detection unit 14 may detect the vibration of each of the plurality of object TGs. In this case, the plurality of object TGs may form a plurality of devices different from each other. Further, the plurality of object TGs may each constitute a plurality of constituent members constituting one device.

Further, the detection device 1 of the modified example of the first embodiment includes a plurality of imaging units 13, and each of the plurality of imaging units 13 outputs an image by photographing the marker unit 11, and the plurality of detection units 14 are present. The vibration of the object TG to which the marker portion 11 is fixed may be detected based on the image of.

Further, the detection device 1 of the modified example of the first embodiment may detect the vibration of the object TG based on a part of a plurality of pixels included in the image output by the imaging unit 13. For example, when the image includes a plurality of blocks composed of a plurality of row pixel groups and a plurality of different processing periods are assigned to the plurality of blocks, the detection unit 14 is based on the row pixel group representing each block. The vibration of the object TG may be detected.

<Second Embodiment>
Next, the detection device of the second embodiment will be described. The detection device of the second embodiment has a different shape of the area of the marker portion from the detection device of the first embodiment. Hereinafter, the differences will be mainly described. In the description of the second embodiment, those having the same reference numerals as those used in the first embodiment are the same or substantially the same.

(Constitution)
As shown in FIG. 12, the marker portion 11A of the second embodiment has a square shape. The marker portion 11A may have a shape different from the square shape (for example, a rectangular shape, a circular shape, an elliptical shape, a polygonal shape, or the like).
In this example, the marker portion 11A extends in a direction in which both ends in the x-axis direction are inclined by an angle smaller than 45 degrees with respect to the y-axis direction. In the marker portion 11A, both ends in the x-axis direction may extend in the y-axis direction.
In this example, the marker portion 11A is composed of one component. The marker portion 11A may be composed of a plurality of components. In this case, the plurality of components may be fixed at a plurality of positions different from each other.

The marker portion 11A forms a plurality of regions adjacent to each other while the amount of reflected light is different from each other. In this example, the marker portion 11A is surrounded by the first region 111A, the second region 112A adjacent to the first region 111A so as to be surrounded by the first region 111A, and the second region 112A so as to be surrounded by the second region 112A. It has a third region 113A adjacent to the third region 113A and a fourth region 114A adjacent to the third region 113A so as to be surrounded by the third region 113A.

In this example, the amount of reflected light in the first region 111A is equal to that in the third region 113A. The amount of reflected light in the second region 112A is equal to that in the fourth region 114A. The amount of reflected light in the first region 111A and the third region 113A is smaller than that in the second region 112A and the fourth region 114A.

In this example, the second region 112A and the fourth region 114A are made of a retroreflective material. In this example, the first region 111A and the third region 113A are made of a material different from the retroreflective material (for example, paper, cloth, resin, metal, ceramics, etc.). In this example, the first region 111A and the third region 113A have a black color. The first region 111A and the third region 113A may have a color different from black.

Further, the second region 112A and the fourth region 114A do not have to be made of a retroreflective material. In this case, for example, the second region 112A and the fourth region 114A have white color, and the first region 111A and the third region 113A have a color different from white (for example, black). May be good.
Further, at least one of the first region 111A, the second region 112A, the third region 113A, and the fourth region 114A may be the surface of the object TG. For example, the first region 111A and the third region 113A may be the surfaces of the object TG.

The second region 112A has a circular shape having a center in the central portion of the marker portion 11A. Therefore, the boundary between the second region 112A and the first region 111A forms a circle.

The third region 113A has a strip shape having a predetermined width. The third region 113A extends along a rectangle located inside the second region 112A. In this example, the four ends of the third region 113A are parallel to the four ends of the marker portion 11A, respectively.

The third region 113A is a pair of side portions 1131A forming both end sides in the y-axis direction of the third region 113A and a pair forming both end sides of the third region 113A in the x-axis direction. The side portion 1132A and the like.

In this example, each of the pair of side portions 1131A and the pair of side portions 1132A extends in a direction inclined with respect to the y-axis direction. Therefore, the boundary between the third region 113A and the second region 112A or the fourth region 114A forms an inclined straight line which is a straight line inclined with respect to the y-axis direction.

Further, in this example, the pair of side portions 1131A extend in parallel with each other. Therefore, the boundary between the pair of side portions 1131A and the second region 112A or the fourth region 114A extends in a direction different from the x-axis direction and forms a pair of straight lines parallel to each other.

Similarly, the pair of sides 1132A extend parallel to each other. Therefore, the boundary between the pair of side portions 1132A and the second region 112A or the fourth region 114A extends in a direction different from the x-axis direction and forms a pair of straight lines parallel to each other.

In this example, the width of each of the pair of side portions 1131A is slightly larger than the width of each of the pair of side portions 1132A. The width of each of the pair of side portions 1131A may be equal to or less than the width of each of the pair of side portions 1132A.

The third region 113A has a pair of connecting portions 1133A that connect the pair of corner portions and the first region at a pair of rectangular corners. Each of the pair of connecting portions 1133A has a strip shape having a predetermined width.

The fourth region 114A has a rectangular shape located inside the third region 113A. In this example, the four ends of the fourth region 114A are parallel to the four ends of the marker portion 11A, respectively.

(function)
As shown in FIG. 13, the functions of the detection unit 14A of the second embodiment include the y-direction straight line boundary position acquisition unit 1401 and the y-direction straight line reference value acquisition among the functions of the detection unit 14 of the first embodiment. First implementation, except that the circular boundary position acquisition unit 1412A, the central straight line reference value acquisition unit 1413A, and the x-direction fluctuation amount acquisition unit 1414A are provided in place of the unit 1402 and the x-direction fluctuation amount acquisition unit 1403. This is the same as the function of the form detection unit 14.

The circular boundary position acquisition unit 1412A is a portion of the marker unit 11 where only the first region 111A of the marker unit 11A exists (in this example, the portion of the marker unit 11A where only the first region 111A exists) from the group of M row pixels. ) Is excluded, and P row pixel groups corresponding to the portions are extracted. P represents an integer smaller than M.

The circular boundary position acquisition unit 1412A acquires a pair of circular boundary positions for each of the P row pixel groups based on the N pixels included in the row pixel group. In this example, the pair of circular boundary positions correspond to the positions of the boundaries between the second region 112A and the first region 111A.

In this example, the circular boundary position acquisition unit 1412A sets the first value region in which the first value is continuous among the N pixels included in the row pixel group for each of the P row pixel groups. , In the negative direction of the x-axis within the boundary between the first value region and the second value region which is adjacent to the first value region at a position in the positive direction of the x-axis and the second value is continuous. The position of the first boundary from the end to the positive direction of the x-axis, the first-value region where the first values are continuous, and the first-value region in the negative direction of the x-axis from the first-value region. Within the boundary between the second value region, which is adjacent at the position of and the second value is continuous, the position of the first boundary from the end in the positive direction of the x-axis toward the negative direction of the x-axis. Obtained as a pair of circular boundary positions.

The center line reference value acquisition unit 1413A acquires the center line reference value for each of the P row pixel groups based on the pair of circle boundary positions of the P set acquired by the circle boundary position acquisition unit 1412A.

In this example, the central straight line reference value acquisition unit 1413A acquires the midpoint position, which is the midpoint of the pair of circular boundary positions acquired by the circular boundary position acquisition unit 1412A, for each of the P row pixel groups. To do. The central straight line reference value acquisition unit 1413A extends in the y-axis direction and acquires a straight line having the minimum total distance to the acquired P midpoint positions as a reference straight line. The central straight line reference value acquisition unit 1413A may extend in the y-axis direction and acquire a straight line having the smallest sum of squares of distances with respect to the acquired P midpoint positions as a reference straight line.

The center straight line reference value acquisition unit 1413A centers on each of the P row pixel groups with the positions of the pixels intersecting the acquired reference straight line among the N pixels included in the row pixel group. Obtained as a straight line reference value.

The x-direction fluctuation amount acquisition unit 1414A includes a pair of P-group circular boundary positions acquired by the circular boundary position acquisition unit 1412A and P central straight line reference values acquired by the central straight line reference value acquisition unit 1413A. Based on this, the amount of variation in the x-direction for each of the P row pixel groups is acquired. In this example, the amount of fluctuation in the x-direction is the difference between the midpoint position, which is the midpoint of the pair of circular boundary positions, and the center line reference value in the x-axis direction.

The spectrum analysis unit 1404 performs spectrum analysis on the P x-direction fluctuation amounts acquired by the x-direction fluctuation amount acquisition unit 1414A. In this example, the position of each pixel in the y-axis direction corresponds to the time when the light on which the value represented by the pixel is based is incident on the imaging unit 13. Therefore, in this example, the change in the x-direction fluctuation amount in the y-axis direction corresponds to the change in the x-direction fluctuation amount with time.

For example, as shown in FIG. 5, it is assumed that the object TG vibrates in the x-axis direction. In this case, for example, the midpoint position is located on a wavy line that fluctuates in the x-axis direction, as represented by the curve L11. Therefore, the change in the x-direction fluctuation amount, which is the difference between the center straight line reference value on the reference straight line represented by the straight line L10 and the midpoint position in the x-axis direction, is the change in the y-axis direction of the object TG. Reflects vibration in the x-axis direction with high accuracy.

Further, even when the object TG vibrates in the y-axis direction, the reference straight line extends in the y-axis direction, so that the vibration in the y-axis direction is not easily reflected in the amount of fluctuation in the x-direction. Further, even when the object TG vibrates in the z-axis direction, the reference straight line does not easily move with the vibration, so that the vibration in the z-axis direction is not easily reflected in the amount of fluctuation in the x-direction. Therefore, the change in the amount of fluctuation in the x-direction in the y-axis direction can reflect the vibration of the object TG in the x-axis direction with high accuracy.

In this example, the circle boundary position corresponds to the boundary position. In this example, the central straight line reference value corresponds to the reference value.

(motion)
Next, the operation of the detection device 1 of the second embodiment will be described with reference to FIG.
The detection device 1 of the second embodiment operates in the same manner as the detection device 1 of the first embodiment except that the process of FIG. 14 is executed instead of the process of FIG.

Hereinafter, the processing of FIG. 14 will be described.
The detection unit 14A acquires a pair of circular boundary positions for each of the P row pixel groups based on the N pixels included in the row pixel group (step S101A in FIG. 14).

Next, the detection unit 14A acquires the center straight line reference value for each of the P row pixel groups based on the pair of circular boundary positions of the P set acquired in step S101A (step S102A in FIG. 14).

Next, the detection unit 14A has P row pixel groups based on the pair of circle boundary positions of the P set acquired in step S101A and the P center straight line reference values acquired in step S102A. The amount of variation in the x-direction for each of the above is acquired (step S103A in FIG. 14).

Next, the detection unit 14A acquires the x-direction fluctuation frequency component by performing Fourier transform processing, wavelet transform processing, or the like on the P x-direction fluctuation amounts acquired in step S103A (step S104A in FIG. 14). ).
Next, the detection unit 14A detects the vibration of the object TG in the x-axis direction based on the x-direction fluctuation frequency component acquired in step S104A (step S105A in FIG. 14).
In this way, the detection unit 14A executes the process shown in FIG.

As described above, according to the detection device 1 of the second embodiment, the same operations and effects as those of the detection device 1 of the first embodiment are exhibited.
Further, in the detection device 1 of the second embodiment, in the marker unit 11A, at least a part of the boundary of the plurality of regions forms at least a part of the circle. The detection unit 14A acquires two boundary positions for each of at least a part of the plurality of row pixel groups, acquires the midpoint position which is the midpoint of the acquired two boundary positions, and acquires the midpoint position. Vibration of the object TG in the second direction based on the change in the amount of fluctuation of the plurality of midpoint positions in the second direction (x-axis direction in this example) in the first direction (y-axis direction in this example). Is detected.

For example, when the boundary of a plurality of regions forms a straight line and the boundary position of the boundary forming the straight line is acquired, the row pixel whose boundary position can be acquired as the straight line approaches the state of being parallel to the second direction. The number of groups is reduced. Therefore, the vibration of the object TG in the second direction may not be detected with high accuracy.

On the other hand, according to the detection device 1, the midpoint position can be acquired for each of the plurality of row pixel groups regardless of the rotation angle of the marker unit 11A with respect to the object TG. As a result, for example, the difference in the positional relationship between the marker unit 11 and the imaging unit 13, which may occur due to the fixing of the marker unit 11 or the installation of the imaging unit 13, affects the accuracy of the vibration detection of the object TG. The effect can be suppressed. Further, the amount of fluctuation in the second direction of the plurality of midpoint positions acquired for each of the plurality of row pixel groups reflects the vibration of the object TG in the second direction with high accuracy. Therefore, the vibration of the object TG in the second direction can be detected with high accuracy.

<First modification of the second embodiment>
Next, the detection device of the first modification of the second embodiment will be described. The detection device of the first modification of the second embodiment has a different shape of the area of the marker portion from the detection device of the second embodiment. Hereinafter, the differences will be mainly described. In the description of the first modification of the second embodiment, those having the same reference numerals as those used in the second embodiment are the same or substantially the same.

(Constitution)
As shown in FIG. 15, the marker portion 11B of the first modification of the second embodiment has a circular shape.
In this example, the marker portion 11B is composed of one component. The marker portion 11B may be composed of a plurality of components. In this case, the plurality of components may be fixed at a plurality of positions different from each other.
The marker portion 11B forms a plurality of regions adjacent to each other while the amount of reflected light is different from each other. In this example, the marker portion 11B has a first region 111B and a pair of second regions 112B adjacent to the first region 111B so as to be surrounded by the first region 111B.

Further, the marker portion 11B is fixed to the object TG to form an outer region of the marker portion 11B. Therefore, in this example, the boundary between the first region 111B and the region outside the marker portion 11B forms a circle.

In this example, the amount of reflected light in the first region 111B is smaller than that in the second region 112B. Further, in this example, the amount of reflected light in the first region 111B is smaller than that in the region outside the marker portion 11B.

In this example, the pair of second regions 112B are made of retroreflective material. In this example, the first region 111B is made of a material different from the retroreflective material (for example, paper, cloth, resin, metal, ceramics, etc.). In this example, the first region 111B has a black color. The first region 111B may have a color different from black.

Further, the pair of second regions 112B may not be made of a retroreflective material. In this case, for example, the pair of second regions 112B may have white color, and the first region 111B may have a color different from white (for example, black). In this example, the outer region of the marker portion 11B is made of a glossy metal. In this example, the region outside the marker portion 11B is the surface of the object TG.
At least one of the pair of second regions 112B may be the surface of the object TG.

Each of the pair of second regions 112B is strip-shaped with a predetermined width. Each of the pair of second regions 112B is L-shaped. The pair of second regions 112B extend along a rectangle located inside the first region 111B. The pair of second regions 112B extend in a direction in which both ends in the x-axis direction are inclined by an angle smaller than 45 degrees with respect to the y-axis direction.

The pair of second regions 112B includes a pair of side portions 1121B forming both end sides in the y-axis direction of the pair of second regions 112B and both ends of the pair of second regions 112B in the x-axis direction. It has a pair of side portions 1122B, each of which forms a side.

In this example, each of the pair of side portions 1121B and the pair of side portions 1122B extends in a direction inclined with respect to the y-axis direction. Therefore, the boundary between each of the pair of side portions 1121B and the pair of side portions 1122B and the first region 111B forms an inclined straight line which is a straight line inclined with respect to the y-axis direction.

Further, in this example, the pair of side portions 1121B extend in parallel with each other. Therefore, the boundary between the pair of side portions 1121B and the first region 111B extends in a direction different from the x-axis direction and forms a pair of straight lines parallel to each other.

Similarly, the pair of sides 1122B extend parallel to each other. Therefore, the boundary between the pair of side portions 1122B and the first region 111B extends in a direction different from the x-axis direction and forms a pair of straight lines parallel to each other.

In this example, the width of each of the pair of side portions 1121B is slightly larger than the width of each of the pair of side portions 1122B. The width of each of the pair of side portions 1121B may be equal to or less than the width of each of the pair of side portions 1122B.

As described above, according to the detection device 1 of the first modification of the second embodiment, the same operations and effects as those of the detection device 1 of the second embodiment are exhibited.

<Third Embodiment>
Next, the detection device of the third embodiment will be described. The detection device of the third embodiment has a different shape of the area of the marker portion from the detection device of the first embodiment. Hereinafter, the differences will be mainly described. In the description of the third embodiment, those having the same reference numerals as those used in the first embodiment are the same or substantially the same.

(Constitution)
As shown in FIG. 16, the marker portion 11C of the third embodiment has a rectangular shape. The marker portion 11C may have a shape different from the rectangular shape (for example, a square shape, a circular shape, an elliptical shape, a polygonal shape, or the like).
In this example, the marker unit 11C is composed of one component. The marker unit 11C may be composed of a plurality of components. In this case, the plurality of components may be fixed at a plurality of positions different from each other.
In this example, the marker portion 11C extends in a direction in which both ends in the x-axis direction are inclined by an angle smaller than 45 degrees with respect to the y-axis direction. In the marker portion 11C, both ends in the x-axis direction may extend in the y-axis direction.

The marker unit 11C forms a plurality of regions adjacent to each other while the amount of reflected light is different from each other. In this example, the marker portion 11C has a first region 111C and a second region 112C adjacent to the first region 111C so as to be surrounded by the first region 111C. The first region 111C has a larger amount of reflected light than the second region 112C.

In this example, the first region 111C is made of a retroreflective material. In this example, the second region 112C is made of a material different from the retroreflective material (eg, paper, cloth, resin, metal, ceramics, etc.). In this example, the second region 112C has a black color. The second region 112C may have a color different from black.
Further, the first region 111C does not have to be made of a retroreflective material. In this case, for example, the first region 111C may have white color, and the second region 112C may have a color different from white (for example, black).
Further, at least one of the first region 111C and the second region 112C may be the surface of the object TG. For example, the second region 112C may be the surface of the object TG.

The second region 112C is linear with a predetermined width. In this example, the second region 112C is N-shaped. The second region 112C includes a first straight line portion 1121C, a second straight line portion 1122C, and a third straight line portion 1123C.

In the first straight line portion 1121C, the distance between the end portion of the marker portion 11C in the negative direction of the x-axis and the end portion of the marker portion 11C in the negative direction of the x-axis is the positive axis of the y-axis. It extends in a direction that is inclined with respect to the edge so as to become longer in the direction. Therefore, the boundary between the first straight line portion 1121C and the first region 111C of the second region 112C forms a first straight line extending in a direction different from the x-axis direction.

In the second straight line portion 1122C, the distance between the central portion of the marker portion 11C in the x-axis direction and the end edge of the marker portion 11C in the negative direction of the x-axis is in the negative direction of the y-axis. It extends in a direction that is inclined with respect to the edge so as to become longer toward the end. Therefore, the boundary between the second straight line portion 1122C and the first region 111C of the second region 112C forms a second straight line extending in a direction different from the x-axis direction and extending in a direction different from the first straight line.

In this example, the second straight line portion 1122C has the end portion of the first straight line portion 1121C in the positive direction of the y-axis and the end portion of the third straight line portion 1123C in the negative direction of the y-axis. Extend to connect.

In the third straight line portion 1123C, the distance between the end portion of the marker portion 11C in the positive direction of the x-axis and the end portion of the marker portion 11C in the negative direction of the x-axis is the positive axis of the y-axis. It extends in a direction that is inclined with respect to the edge so as to become longer in the direction. Therefore, the boundary between the third straight line portion 1123C and the first region 111C of the second region 112C forms a third straight line extending in a direction different from the x-axis direction.

In this example, the third straight line portion 1123C is parallel to the first straight line portion 1121C. Therefore, the third straight line is parallel to the first straight line.
As described above, the boundary between the first straight line portion 1121C and the first region 111C and the boundary between the third straight line portion 1123C and the first region 111C are in a direction different from the x-axis direction (in this example, in the y-axis direction). It extends in the direction of slight inclination) and forms a pair of straight lines that are parallel to each other.

(function)
As shown in FIG. 17, the functions of the detection unit 14C of the third embodiment include the y-direction straight line boundary position acquisition unit 1401 and the y-direction straight line reference value acquisition among the functions of the detection unit 14 of the first embodiment. In place of the unit 1402, the x-direction fluctuation amount acquisition unit 1403, the inclined straight line boundary position acquisition unit 1406, the inclined straight line reference value acquisition unit 1407, and the y-direction fluctuation amount acquisition unit 1408, the linear boundary position group acquisition unit 1421C, the distance ratio Detection of the first embodiment except that the reference value acquisition unit 1422C, the y-direction fluctuation amount acquisition unit 1423C, the linear boundary position acquisition unit 1424C, the linear reference value acquisition unit 1425C, and the x-direction fluctuation amount acquisition unit 1426C are provided. It is the same as the function of the part 14.

The straight line boundary position group acquisition unit 1421C is a portion of the marker unit 11C excluding both ends in the y-axis direction from the M row pixel group (in this example, the first straight line portion of the marker unit 11C). A group of P row pixels corresponding to 1121C, a portion in which all of the second straight line portion 1122C and the third straight line portion 1123C are present) is extracted. P represents an integer smaller than M.

The linear boundary position group acquisition unit 1421C acquires a linear boundary position group for each of the P row pixel groups based on N pixels included in the row pixel group. In this example, the straight line boundary position group includes the first straight line boundary position, the second straight line boundary position, and the third straight line boundary position.

In this example, the first straight line boundary position corresponds to the position of the boundary between the end side of the first straight line portion 1121C in the negative direction of the x-axis and the first region 111C. The second straight line boundary position corresponds to the position of the boundary between the end side of the second straight line portion 1122C in the negative direction of the x-axis and the first region 111C. The third straight line boundary position corresponds to the position of the boundary between the end side of the third straight line portion 1123C in the negative direction of the x-axis and the first region 111C.

The first straight line boundary position may correspond to the position of the boundary between the end side of the first straight line portion 1121C in the positive direction of the x-axis and the first region 111C. The second straight line boundary position may correspond to the position of the boundary between the end side of the second straight line portion 1122C in the positive direction of the x-axis and the first region 111C. The third straight line boundary position may correspond to the position of the boundary between the end side of the third straight line portion 1123C in the positive direction of the x-axis and the first region 111C.

Further, the first straight line boundary position is the position of the boundary between the end side of the first straight line portion 1121C in the negative direction of the x-axis and the first region 111C, and the x of the first straight line portion 1121C. It may correspond to the position of the midpoint between the end side in the positive direction of the axis and the position of the boundary between the first region 111C. The second straight line boundary position is the position of the boundary between the end side in the negative direction of the x-axis of the second straight line portion 1122C and the first region 111C, and the x-axis of the second straight line portion 1122C. It may correspond to the position of the boundary between the end side in the positive direction and the first region 111C and the position of the midpoint. The third straight line boundary position is the position of the boundary between the end side in the negative direction of the x-axis of the third straight line portion 1123C and the first region 111C, and the x-axis of the third straight line portion 1123C. It may correspond to the position of the boundary between the end side in the positive direction and the first region 111C and the position of the midpoint.

In this example, the linear boundary position group acquisition unit 1421C has a second value region in which the second value of the N pixels included in the row pixel group is continuous with respect to each of the P row pixel groups. And the negative of the x-axis within the boundary between the second value region and the first value region which is adjacent to the second value region at a position in the positive direction of the x-axis and the first value is continuous. The positions of the first to third boundaries are acquired as the first straight line boundary position, the second straight line boundary position, and the third straight line boundary position from the end in the direction toward the positive direction of the x-axis.

The distance ratio reference value acquisition unit 1422C acquires the distance ratio reference value for each of the P row pixel groups based on the P linear boundary position group acquired by the linear boundary position group acquisition unit 1421C.

In this example, the distance ratio reference value acquisition unit 1422C acquires a straight line having the smallest total distance with respect to the P first straight line boundary positions acquired by the straight line boundary position group acquisition unit 1421C as the first reference straight line. The distance ratio reference value acquisition unit 1422C acquires a straight line having the smallest sum of squares of distances with respect to the P first straight line boundary positions acquired by the straight line boundary position group acquisition unit 1421C as the first reference straight line. May be good.

The distance ratio reference value acquisition unit 1422C determines the positions of the pixels intersecting the acquired first reference line among the N pixels included in the row pixel group for each of the P row pixel groups. , Obtained as the first straight line reference value.

The distance ratio reference value acquisition unit 1422C acquires the second straight line reference value based on the second straight line boundary position by using the same method as the method of acquiring the first straight line reference value based on the first straight line boundary position. At the same time, the third straight line reference value is acquired based on the third straight line boundary position.

The distance ratio reference value acquisition unit 1422C transfers the distance ratio to each of the P row pixel groups based on the acquired first straight line reference value, second straight line reference value, and third straight line reference value. Get the reference value.

The distance ratio reference value is the first distance between the first position pair consisting of two straight line reference values among the first straight line reference value, the second straight line reference value, and the third straight line reference value, and the first. The ratio of the 1st straight line reference value, the 2nd straight line reference value, and the 3rd straight line reference value to the 2nd distance between the 2nd position pair consisting of 2 straight line reference values different from the 1st position pair. Is.

In this example, the distance ratio reference value is the second straight line reference value and the third straight line reference value of the first distance between the first straight line reference value and the first position pair consisting of the second straight line reference value. A ratio to a second distance between a second position pair consisting of. The first position pair may consist of the first straight line reference value and the second straight line reference value, and the second position pair may consist of the first straight line reference value and the third straight line reference value. .. Further, the first position pair may be composed of the first straight line reference value and the third straight line reference value, and the second position pair may be composed of the second straight line reference value and the third straight line reference value. ..

The y-direction fluctuation amount acquisition unit 1423C includes P linear boundary position groups acquired by the linear boundary position group acquisition unit 1421C and P distance ratio reference values acquired by the distance ratio reference value acquisition unit 1422C. Based on this, the amount of fluctuation in the y direction for each of the P row pixel groups is acquired. In this example, the y-direction fluctuation amount acquisition unit 1423C has the first linear boundary position, the second linear boundary position, and the linear boundary position acquired by the linear boundary position group acquisition unit 1421C for each of the P row pixel groups. , The distance ratio is acquired based on the third straight line boundary position.

In this example, the distance ratio is the first straight line boundary position, the second straight line boundary position, and the third straight line boundary position instead of the first straight line reference value, the second straight line reference value, and the third straight line reference value. Is obtained using the same method as the distance ratio reference value, except that. Therefore, in this example, the distance ratio is the second straight line boundary position and the third straight line boundary position of the first distance between the first straight line boundary position and the first position pair consisting of the second straight line boundary position. The ratio to the second distance between the second position pairs consisting of.

The y-direction fluctuation amount acquisition unit 1423C has a distance ratio reference value having a value closest to the distance ratio acquired for the row pixel group and the row pixel group for each of the P row pixel groups. The difference between the acquired row pixel group and the position in the y-axis direction is acquired as the amount of fluctuation in the y-direction.

The spectrum analysis unit 1404 performs spectrum analysis on P Y-direction fluctuation amounts acquired by the y-direction fluctuation amount acquisition unit 1423C. In this example, the position of each pixel in the y-axis direction corresponds to the time when the light on which the value represented by the pixel is based is incident on the imaging unit 13. Therefore, in this example, the change in the y-direction fluctuation amount in the y-axis direction corresponds to the change in the y-direction fluctuation amount with time.

For example, as shown in FIG. 18, it is assumed that the object TG vibrates in the y-axis direction. In this case, for example, each of the first straight line boundary position, the second straight line boundary position, and the third straight line boundary position is located on a wavy line that fluctuates in the x-axis direction as represented by the curve L41. Further, each of the first straight line reference value, the second straight line reference value, and the third straight line reference value is located on three reference straight lines as represented by the straight line L40.

Therefore, a distance ratio reference value having a value closest to a certain row pixel group and the distance ratio based on the first straight line boundary position, the second straight line boundary position, and the third straight line boundary position with respect to the row pixel group is acquired. The change in the y-axis direction of the y-direction fluctuation amount representing the difference in position between the row pixel group and the y-axis direction reflects the vibration of the object TG in the y-axis direction with high accuracy.

In this example, the distance ratio γ is expressed by Equation 3. Here, s ₁ represents the first distance between the first straight line boundary position and the first position pair consisting of the second straight line boundary position. s ₂ represents the second distance between the second straight line boundary position and the second position pair consisting of the third straight line boundary position.

Further, even when the object TG vibrates in the x-axis direction, the distance ratio does not easily change with the vibration, so that the vibration in the x-axis direction is not easily reflected in the amount of fluctuation in the y-direction. Further, even when the object TG vibrates in the z-axis direction, the distance ratio does not easily change with the vibration, so that the vibration in the z-axis direction is not easily reflected in the amount of fluctuation in the y-direction. Therefore, the change in the y-direction fluctuation amount in the y-axis direction can reflect the vibration of the object TG in the y-axis direction with high accuracy.

The linear boundary position acquisition unit 1424C is a portion of the marker unit 11C that includes both ends in the y-axis direction (in this example, a portion including both ends of the first linear unit 1121C in the y-axis direction) from the group of M row pixels. The Q row pixel group corresponding to the portion excluding is extracted. Q represents an integer smaller than M. In this example, the integer represented by Q is equal to the integer represented by P. The integer represented by Q may be different from the integer represented by P.

The linear boundary position acquisition unit 1424C acquires the linear boundary position for each of the Q row pixel groups based on the N pixels included in the row pixel group. In this example, the straight line boundary position corresponds to the position of the boundary between the end side of the first straight line portion 1121C in the negative direction of the x-axis and the second region 112C.

The straight line boundary position may correspond to the position of the boundary between the end side of the first straight line portion 1121C in the positive direction of the x-axis and the second region 112C. Further, the linear boundary position is the position of the boundary between the end side in the negative direction of the x-axis of the first straight portion 1121C and the second region 112C, and the x-axis of the first straight portion 1121C. It may correspond to the position of the boundary between the end edge in the positive direction and the second region 112C, and the position of the midpoint.

The detection unit 14C may acquire the amount of fluctuation in the x direction by using the second straight line portion 1122C or the third straight line portion 1123C instead of the first straight line portion 1121C. Further, the detection unit 14C may acquire the amount of fluctuation in the x direction by using at least two of the first straight line unit 1121C, the second straight line unit 1122C, and the third straight line unit 1123C. In this case, for example, the detection unit 14C acquires the position of the boundary with respect to at least two of the first straight line portion 1121C, the second straight line portion 1122C, and the third straight line portion 1123C, and the acquired plurality of positions. The average value of may be acquired as a linear boundary position.

In this example, the linear boundary position acquisition unit 1424C sets the second value region in which the second value is continuous among the N pixels included in the row pixel group for each of the Q row pixel groups. , In the negative direction of the x-axis within the boundary between the second value region and the first value region which is adjacent to the second value region at a position in the positive direction of the x-axis and the first value is continuous. The position of the first boundary from the end in the positive direction of the x-axis is acquired as the linear boundary position.

The linear boundary position acquisition unit 1424C corrects the acquired Q linear boundary positions based on the y-direction fluctuation amount acquired by the y-direction fluctuation amount acquisition unit 1423C.

In this example, the straight line boundary position acquisition unit 1424C acquires a provisional reference straight line based on the acquired Q straight line boundary positions. In this example, the straight line boundary position acquisition unit 1424C acquires a straight line having the minimum total distance with respect to the acquired Q straight line boundary positions as a provisional reference straight line. The straight line boundary position acquisition unit 1424C may acquire a straight line having the smallest sum of squares of distances with respect to the acquired Q straight line boundary positions as a provisional reference straight line.

Further, in this example, the linear boundary position acquisition unit 1424C sets the acquired linear boundary position for each of the Q row pixel groups to the y-direction fluctuation amount acquired by the y-direction fluctuation amount acquisition unit 1423C. , The straight line boundary position is moved in the direction opposite to the direction represented by the value obtained by multiplying the margin of the angle formed by the acquired provisional reference straight line and the x-axis direction by the magnitude represented by the value. to correct. In other words, in this example, the linear boundary position acquisition unit 1424C corrects the linear boundary position for each of the Q row pixel groups so as to cancel the amount of fluctuation in the y direction.

The linear reference value acquisition unit 1425C acquires linear reference values for each of the Q row pixel groups based on the Q linear boundary positions corrected by the linear boundary position acquisition unit 1424C.

In this example, the straight line reference value acquisition unit 1425C acquires a straight line having the minimum total distance to the Q straight line boundary positions corrected by the straight line boundary position acquisition unit 1424C as a reference straight line. The straight line reference value acquisition unit 1425C may acquire a straight line having the smallest sum of squares of distances with respect to the Q straight line boundary positions corrected by the straight line boundary position acquisition unit 1424C as a reference straight line.

The linear reference value acquisition unit 1425C sets the position of the pixel intersecting the acquired reference straight line among the N pixels included in the row pixel group with respect to each of the Q row pixel groups as a linear reference. Get as a value.

The x-direction fluctuation amount acquisition unit 1426C has Q based on the Q linear boundary positions corrected by the linear boundary position acquisition unit 1424C and the Q linear reference values acquired by the linear reference value acquisition unit 1425C. The amount of variation in the x-direction for each of the row pixel groups is acquired.

In this example, the amount of fluctuation in the x direction is the difference in position between the straight line boundary position and the straight line reference value in the x-axis direction. In other words, the amount of fluctuation in the x-direction corresponds to the amount of fluctuation in the x-axis direction of the Q linear boundary positions corrected by the linear boundary position acquisition unit 1424C.

The spectrum analysis unit 1404 performs spectrum analysis on the Q x-direction fluctuation amounts acquired by the x-direction fluctuation amount acquisition unit 1426C. In this example, the position of each pixel in the y-axis direction corresponds to the time when the light on which the value represented by the pixel is based is incident on the imaging unit 13. Therefore, in this example, the change in the x-direction fluctuation amount in the y-axis direction corresponds to the change in the x-direction fluctuation amount with time.

In this example, each of the first straight line boundary position, the second straight line boundary position, the third straight line boundary position, and the straight line boundary position corresponds to the boundary position. In this example, each of the distance ratio reference value and the linear reference value corresponds to the reference value.

(motion)
Next, the operation of the detection device 1 of the third embodiment will be described with reference to FIGS. 19 and 20.
The detection device 1 of the third embodiment detects the first embodiment except that the process of FIG. 19 and the process of FIG. 20 are executed instead of the process of FIG. 9 and the process of FIG. It operates in the same manner as the device 1.

Hereinafter, the processing of FIG. 19 will be described.
The detection unit 14C acquires a linear boundary position group for each of the P row pixel groups based on the N pixels included in the row pixel group (step S401 in FIG. 19).

Next, the detection unit 14C acquires a distance ratio reference value for each of the P row pixel groups based on the P linear boundary position group acquired in step S401 (step S402 in FIG. 19).

Next, the detection unit 14C sets the P row pixel group based on the P linear boundary position group acquired in step S401 and the P distance ratio reference value acquired in step S402. The amount of fluctuation in the y direction for each is acquired (step S403 in FIG. 19).

Next, the detection unit 14C acquires the y-direction fluctuation frequency component by performing Fourier transform processing, wavelet transform processing, or the like on the P y-direction fluctuation amounts acquired in step S403 (step S404 in FIG. 19). ).
Next, the detection unit 14C detects the vibration of the object TG in the y-axis direction based on the y-direction fluctuation frequency component acquired in step S404 (step S405 in FIG. 19).
In this way, the detection unit 14C executes the process shown in FIG.

Next, the process of FIG. 20 will be described.
The detection unit 14C acquires a linear boundary position for each of the Q row pixel groups based on the N pixels included in the row pixel group (step S501 in FIG. 20).

Next, the detection unit 14C corrects the Q linear boundary positions acquired in step S501 based on the y-direction fluctuation amount acquired in step S403 of FIG. 19 (step S502 of FIG. 20).
Next, the detection unit 14C acquires a linear reference value for each of the Q row pixel groups based on the Q linear boundary positions corrected in step S502 (step S503 in FIG. 20).

Next, the detection unit 14C with respect to each of the Q row pixel groups based on the Q linear boundary positions corrected in step S502 and the Q linear reference values acquired in step S503. The amount of fluctuation in the x direction is acquired (step S504 in FIG. 20).

Next, the detection unit 14C acquires the x-direction fluctuation frequency component by performing Fourier transform processing, wavelet transform processing, or the like on the Q x-direction fluctuation amounts acquired in step S504 (step S505 in FIG. 20). ).
Next, the detection unit 14C detects the vibration of the object TG in the x-axis direction based on the x-direction fluctuation frequency component acquired in step S505 (step S506 in FIG. 20).
In this way, the detection unit 14C executes the process of FIG. 20.

As described above, the detection device 1 of the third embodiment includes an imaging unit 13 and a detection unit 14C. The imaging unit 13 outputs an image by photographing the object TG. The image includes a plurality of row pixel groups each having a plurality of positions different from each other in the first direction (in this example, the y-axis direction). A plurality of processing periods different from each other are assigned to the plurality of row pixel groups. Each of the plurality of row pixel groups has a plurality of positions different from each other in the second direction (in this example, the x-axis direction) orthogonal to the first direction, and at the processing period assigned to the row pixel group. Includes a plurality of pixels representing the amount of incident light.

In the image output by the imaging unit 13, the detection unit 14C sets the boundary position, which is the position of the boundary of a plurality of regions adjacent to each other while the amount of light represented by the pixels included in the image is different from each other, in the plurality of rows. It is acquired for each of the pixel groups, and the vibration of the object TG is detected based on the acquired plurality of boundary positions.

According to this, the plurality of row pixel groups included in the image represent the amount of light incident on each of the plurality of processing periods different from each other. By the way, the interval between two consecutive processing periods is shorter than the imaging interval. Therefore, the plurality of boundary positions acquired for each of the plurality of row pixel groups can reflect the vibration having a sufficiently short period with respect to the shooting interval. As a result, according to the detection device 1, vibration having a period sufficiently short with respect to the photographing interval can be detected.

Further, in the detection device 1 of the third embodiment, the detection unit 14C acquires a reference value for at least a part of each of the plurality of row pixel groups based on at least a part of the acquired plurality of boundary positions. Then, the vibration of the object TG is detected based on the acquired reference value and the acquired plurality of boundary positions.

For example, when the position of the imaging unit 13 with respect to the object TG cannot be specified, the boundary position in the state where the object TG is not vibrating cannot be specified. Therefore, there is a possibility that the vibration of the object TG cannot be detected with high accuracy.
On the other hand, according to the detection device 1, it is possible to acquire a reference value for each row pixel group in a state where the object TG is not vibrating, based on the acquired plurality of boundary positions. Therefore, even when the position of the imaging unit 13 with respect to the object TG cannot be specified, the vibration of the object TG can be detected with high accuracy.
In the detection device 1 of the modified example of the third embodiment, the detection unit 14C may use a predetermined reference value without acquiring the reference value based on the acquired plurality of boundary positions.

Further, the detection device 1 of the third embodiment includes a marker unit 11C which is fixed to the object TG, has different amounts of reflected light, and forms a plurality of regions adjacent to each other. The imaging unit 13 outputs an image by photographing the marker unit 11C. The detection unit 14C detects the vibration of the object TG.

The amount of light reflected by a plurality of regions adjacent to each other on the surface of the object TG may be substantially the same. In this case, there is a risk that the boundary position cannot be obtained with high accuracy.
On the other hand, according to the detection device 1, the marker portion 11C forming a plurality of regions in which the amount of reflected light is different from each other is fixed to the object TG. Therefore, the boundary position can be acquired with high accuracy by using the image output by photographing the marker unit 11C. As a result, the vibration of the object TG can be detected with high accuracy.

Further, in the detection device 1 of the third embodiment, at least one of the plurality of regions of the marker unit 11C is made of a retroreflective material.

According to this, the amount of light reflected in the region made of the retroreflective material can be increased. As a result, since the boundary position can be acquired with high accuracy, the vibration of the object TG can be detected with high accuracy.

Further, in the detection device 1 of the third embodiment, the marker unit 11C has a first straight line in which at least a part of the boundary between the plurality of regions extends in a direction different from the second direction (x-axis direction in this example). , A second straight line extending in a direction different from the second direction and extending in a direction different from the first straight line, and a third straight line extending in a direction different from the second direction are formed.

The detection unit 14C acquires three boundary positions for each of at least a part of the plurality of row pixel groups, and is between a first position pair consisting of two boundary positions among the acquired three boundary positions. The distance ratio, which is the ratio between the first distance of the above and the second distance between the second position pair consisting of two boundary positions different from the first position pair among the acquired three boundary positions, is acquired. , The vibration of the object TG in the first direction is detected based on the change in the acquired plurality of distance ratios in the first direction (in this example, the y-axis direction).

The distance ratio, which is the ratio of the distances in the second direction between the boundary positions of the boundaries forming the three straight lines, reflects the vibration of the object TG in the first direction with high accuracy. Therefore, according to the detection device 1, the vibration of the object TG in the first direction can be detected with high accuracy.

Further, in the marker portion 11C of the detection device 1 of the third embodiment, the third straight line is parallel to the first straight line. The detection unit 14C is the distance between the two boundary positions corresponding to the first straight line and the third straight line of the three acquired boundary positions for each of at least a part of the plurality of row pixel groups. Is acquired, and based on the changes in the acquired distances in the first direction, the vibration of the object TG in the third direction (in this example, the z-axis direction) orthogonal to the first direction and the second direction is performed. To detect.

The distance in the second direction between the boundary positions of the boundaries forming a pair of straight lines parallel to each other reflects the vibration of the object TG in the third direction with high accuracy. Therefore, according to the detection device 1, the vibration of the object TG in the third direction can be detected with high accuracy.

Further, in the detection device 1 of the third embodiment, the detection unit 14C corrects at least a part of the acquired plurality of boundary positions based on the acquired plurality of distance ratios, and is based on the corrected boundary positions. The vibration of the object TG in the second direction or the third direction is detected.

According to this, the vibration of the object TG in the second direction or the third direction can be detected with higher accuracy.

The detection device 1 of the modified example of the third embodiment does not have to include the light source unit 12.
Further, the detection device 1 of the modified example of the third embodiment does not have to include the marker unit 11C. In this case, the detection device 1 detects the vibration of the object TG by using a part of the object TG (for example, a bolt, a nut, a tag, a plate, a design, etc.) instead of the marker portion 11C. May be good.

Further, the detection device 1 of the modified example of the third embodiment detects the vibration of the object TG in at least one of the x-axis direction, the y-axis direction, and the z-axis direction, and also detects the vibration of the object TG in the other direction. It is not necessary to detect the vibration of the object TG.

Further, the detection device 1 of the modified example of the third embodiment includes a plurality of marker units 11C fixed to each of the plurality of object TGs, and the imaging unit 13 captures the plurality of marker units 11C to capture an image. It may be output and the detection unit 14C may detect the vibration of each of the plurality of object TGs. In this case, the plurality of object TGs may form a plurality of devices different from each other. Further, the plurality of object TGs may each constitute a plurality of constituent members constituting one device.

Further, the detection device 1 of the modified example of the third embodiment includes a plurality of imaging units 13, and each of the plurality of imaging units 13 outputs an image by photographing the marker unit 11C, and the plurality of detection units 14C are present. The vibration of the object TG to which the marker portion 11C is fixed may be detected based on the image of.

Further, the detection device 1 of the modified example of the third embodiment may detect the vibration of the object TG based on a part of a plurality of pixels included in the image output by the imaging unit 13. For example, when the image includes a plurality of blocks composed of a plurality of row pixel groups and a plurality of different processing periods are assigned to the plurality of blocks, the detection unit 14C is based on the row pixel group representing each block. The vibration of the object TG may be detected.

<First modification of the third embodiment>
Next, the detection device of the first modification of the third embodiment will be described. The detection device of the first modification of the third embodiment has a different shape of the area of the marker portion from the detection device of the third embodiment. Hereinafter, the differences will be mainly described. In the description of the first modification of the third embodiment, those having the same reference numerals as those used in the third embodiment are the same or substantially the same.

(Constitution)
As shown in FIG. 21, the marker portion 11D of the first modification of the third embodiment has a circular shape.
In this example, the marker unit 11D is composed of one component. The marker unit 11D may be composed of a plurality of components. In this case, the plurality of components may be fixed at a plurality of positions different from each other.
The marker unit 11D forms a plurality of regions adjacent to each other while the amount of reflected light is different from each other. In this example, the marker unit 11D has a first region 111D and a second region 112D adjacent to the first region 111D so as to be surrounded by the first region 111D.

Further, the marker portion 11D forms an outer region of the marker portion 11D by being fixed to the object TG. Therefore, in this example, the boundary between the first region 111D and the region outside the marker portion 11D forms a circle.

In this example, the amount of reflected light in the first region 111D is smaller than that in the second region 112D. Further, in this example, the amount of reflected light in the first region 111D is smaller than that in the region outside the marker portion 11D.

In this example, the second region 112D is made of a retroreflective material. In this example, the first region 111D is made of a material different from the retroreflective material (for example, paper, cloth, resin, metal, ceramics, etc.). In this example, the first region 111D has a black color. The first region 111D may have a color different from black.

Further, the second region 112D does not have to be made of a retroreflective material. In this case, for example, the second region 112D may have white color, and the first region 111D may have a color different from white (for example, black). In this example, the outer region of the marker portion 11D is made of a glossy metal. In this example, the region outside the marker portion 11D is the surface of the object TG.
The second region 112D may be the surface of the object TG.

The second region 112D is a linear shape having a predetermined width. In this example, the second region 112D is N-shaped. In this example, the second region 112D has the same shape as the second region 112C of the third embodiment. The second region 112D includes a first straight line portion 1121D, a second straight line portion 1122D, and a third straight line portion 1123D, similarly to the second region 112C of the third embodiment.

(function)
As shown in FIG. 22, the functions of the detection unit 14D of the first modification of the third embodiment are the inclination straight line boundary position acquisition unit 1406 and the inclination straight line among the functions of the detection unit 14A of the second embodiment. Instead of the reference value acquisition unit 1407 and the y-direction fluctuation amount acquisition unit 1408, the linear boundary position group acquisition unit 1421C, the distance ratio reference value acquisition unit 1422C, and y of the detection units 14C of the third embodiment. The function is the same as that of the detection unit 14A of the second embodiment except that the direction fluctuation amount acquisition unit 1423C is provided.

As described above, according to the detection device 1 of the first modification of the third embodiment, the same operations and effects as those of the detection device 1 of the third embodiment are exhibited.
Further, in the detection device 1 of the first modification of the third embodiment, in the marker unit 11D, at least a part of the boundary of the plurality of regions forms at least a part of the circle. The detection unit 14D acquires two boundary positions for each of at least a part of the plurality of row pixel groups, acquires the midpoint position which is the midpoint of the acquired two boundary positions, and acquires the midpoint position. Vibration of the object TG in the second direction based on the change in the amount of fluctuation of the plurality of midpoint positions in the second direction (x-axis direction in this example) in the first direction (y-axis direction in this example). Is detected.

According to this, the midpoint position can be acquired for each of the plurality of row pixel groups regardless of the rotation angle of the marker unit 11D with respect to the object TG. Further, the amount of fluctuation in the second direction of the plurality of midpoint positions acquired for each of the plurality of row pixel groups reflects the vibration of the object TG in the second direction with high accuracy. Therefore, the vibration of the object TG in the second direction can be detected with high accuracy.

The present invention is not limited to the above-described embodiment. For example, various modifications that can be understood by those skilled in the art may be made to the above-described embodiments without departing from the spirit of the present invention.

1 Detection device 11, 11A to 11D Marker part 111, 111A to 111D First area 112, 112A to 112D Second area 1121, 1121C, 1121D First straight line part 1121B Side part 1122, 1122C, 1122D Second straight line part 1122B Side part 1123, 1123C, 1123D 3rd straight line part 1131A, 1132A Side part 1133A Connecting part 113A 3rd area 114A 4th area 12 Light source part 13 Imaging part 14, 14A, 14C, 14D Detection part 141 Processing device 142 Storage device 143 Output device 1401 y-direction straight line boundary position acquisition unit 1402 y-direction straight line reference value acquisition unit 1403 x-direction fluctuation amount acquisition unit 1404 Spectrum analysis unit 1405 Vibration detection unit 1406 Inclined straight line boundary position acquisition unit 1407 Inclination straight line reference value acquisition unit 1408 y-direction fluctuation amount acquisition unit Part 1409 Parallel straight line boundary position acquisition part 1410 Distance reference value acquisition part 1411 z direction fluctuation amount acquisition part 1412A Circular boundary position acquisition part 1413A Central straight line reference value acquisition part 1414A x direction fluctuation amount acquisition part 1421C Straight line boundary position group acquisition part 1422C Distance Ratio reference value acquisition unit 1423C y direction fluctuation amount acquisition unit 1424C Straight line boundary position acquisition unit 1425C Straight line reference value acquisition unit 1426C x direction fluctuation amount acquisition unit BU bus P0 Reference position P1 Fluctuation position TG Object

Claims (14)

An image is output by photographing an object, and the image includes a plurality of row pixel groups having a plurality of positions different from each other in the first direction, and a plurality of processing periods different from each other in the plurality of row pixel groups. Are allotted, and each of the plurality of row pixel groups has a plurality of positions different from each other in the second direction orthogonal to the first direction, and in the processing period assigned to the row pixel group. An image pickup unit containing a plurality of pixels representing the amount of incident light,
In the output image, the amount of light represented by the pixels included in the image is different from each other, and the boundary position, which is the position of the boundary of a plurality of regions adjacent to each other, is acquired for each of the plurality of row pixel groups. Then, the detection unit that detects the vibration of the object based on the acquired plurality of boundary positions, and the detection unit.
A detection device. The detection device according to claim 1.
The detection unit acquires a reference value for each of at least a part of the plurality of row pixel groups based on at least a part of the acquired plurality of boundary positions, and obtains the acquired reference value and the acquired reference value. A detection device that detects the vibration based on the acquired plurality of boundary positions. The detection device according to claim 1 or 2.
It is provided with a marker portion which is fixed to an object and has different amounts of reflected light and forms a plurality of regions adjacent to each other.
The imaging unit outputs the image by photographing the marker unit.
The detection unit is a detection device that detects the vibration of the object. The detection device according to claim 3.
The marker unit is a detection device in which at least one of the plurality of regions is made of a retroreflective material. The detection device according to claim 3 or 4.
In the marker portion, at least a part of the boundary of the plurality of regions forms a straight line extending in the first direction.
The detection unit detects the vibration in the second direction based on the change in the first direction of the fluctuation amount in the second direction of at least a part of the acquired plurality of boundary positions. .. The detection device according to claim 3 or 4.
In the marker portion, at least a part of the boundary of the plurality of regions forms at least a part of a circle.
The detection unit acquires two boundary positions for each of at least a part of the plurality of row pixel groups, acquires the midpoint position which is the midpoint of the acquired two boundary positions, and obtains the midpoint position. A detection device that detects the vibration in the second direction based on the change in the acquired fluctuation amount of the plurality of midpoint positions in the second direction in the first direction. The detection device according to any one of claims 3 to 6.
The marker portion forms an inclined straight line in which at least a part of the boundary of the plurality of regions is a straight line inclined with respect to the first direction.
The detection unit detects the vibration in the first direction based on the change in the first direction of the fluctuation amount in the second direction of at least a part of the acquired plurality of boundary positions. .. The detection device according to any one of claims 3 to 7.
The marker portion forms a pair of straight lines in which at least a part of the boundary of the plurality of regions extends in a direction different from the second direction and is parallel to each other.
The detection unit acquires two boundary positions for each of at least a part of the plurality of row pixel groups, acquires the distance between the acquired two boundary positions, and obtains the acquired plurality of boundary positions. A detection device that detects the vibration in the first direction and the third direction orthogonal to the second direction based on the change in the distance in the first direction. The detection device according to any one of claims 5 to 8.
The detection unit corrects at least a part of the acquired plurality of boundary positions based on the amount of fluctuation in the second direction, and based on the corrected boundary position, the first direction or the third direction. A detection device that detects the vibration in a direction. The detection device according to claim 3 or 4.
In the marker portion, at least a part of the boundary of the plurality of regions extends in a direction different from the second direction and a direction different from the first straight line. A second straight line extending in and a third straight line extending in a direction different from the second direction are formed.
The detection unit acquires three boundary positions for each of at least a part of the plurality of row pixel groups, and is a first position pair consisting of two boundary positions out of the acquired three boundary positions. Distance between the first distance between and the second distance between the second position pair consisting of two boundary positions different from the first position pair among the three acquired boundary positions. A detection device that acquires a ratio and detects the vibration in the first direction based on a change in the acquired distance ratio in the first direction. The detection device according to claim 10.
The third straight line is parallel to the first straight line and is parallel to the first straight line.
The detection unit has, for each of at least a part of the plurality of row pixel groups, two boundary positions corresponding to the first straight line and the third straight line among the three acquired boundary positions. The detection, which acquires the distance between the two, and detects the vibration in the third direction orthogonal to the first direction and the second direction based on the change in the acquired plurality of distances in the first direction. apparatus. The detection device according to claim 10 or 11.
In the marker portion, at least a part of the boundary of the plurality of regions forms at least a part of a circle.
The detection unit acquires two boundary positions for each of at least a part of the plurality of row pixel groups, acquires the midpoint position which is the midpoint of the acquired two boundary positions, and obtains the midpoint position. A detection device that detects the vibration in the second direction based on the change in the acquired fluctuation amount of the plurality of midpoint positions in the second direction in the first direction. The detection device according to any one of claims 10 to 12.
The detection unit corrects at least a part of the acquired plurality of boundary positions based on the acquired plurality of distance ratios, and based on the corrected boundary position, the second direction or the first. A detection device that detects the vibration in three directions. An image is output by photographing an object, and the image includes a plurality of row pixel groups having a plurality of positions different from each other in the first direction, and a plurality of processing periods different from each other in the plurality of row pixel groups. Are allotted, and each of the plurality of row pixel groups has a plurality of positions different from each other in the second direction orthogonal to the first direction, and in the processing period assigned to the row pixel group. Contains multiple pixels representing the amount of incident light
In the output image, the amount of light represented by the pixels included in the image is different from each other, and the boundary position, which is the position of the boundary of a plurality of regions adjacent to each other, is acquired for each of the plurality of row pixel groups. Then, the vibration of the object is detected based on the acquired plurality of boundary positions.
Detection methods, including that.

Images