JP2000162220A - Speed measuring system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To obtain a speed measuring system with which a speed is measured easily and with high accuracy. SOLUTION: An imaging device 10a and an imaging device 10b, which use linear sensors are arranged at a prescribed interval along the advance direction on the side part of an object to be measured. Sensor arrangement directions are made nearly at right angles with respect to the advance direction of the object to be measured. An imaged signal DMa and an imaged signal DMb from the devices 10a, 10b are written sequentially in a ring buffer 31a and in a ring buffer 31b at a control device 30. The written signals DMa, DMb are read out. The influence of flickers is removed by a flicker removal circuit 32a and a flicker removal circuit 32b. In a correlation retrieval circuit 33, either of a signal DMCa and a signal DMCb is shifted in the direction of a time base, a correlation coefficient is calculated, and a shift amount whose correlation coefficient becomes maximum is retrieved. On the basis of the retrieved shift amount, the time required for moving the object to be measured by the prescribed interval is found. The prescribed interval is divided by the time, so that the speed of the object to be measured is calculated. The speed can be easily measured from the side part of the object to be measured.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to a speed measurement system. Specifically, first and second imaging devices that generate a photographing signal of a measurement target using a one-dimensional image sensor are arranged at predetermined intervals along a traveling direction on a side of the measurement target, and the first and second imaging devices are arranged. The speed of the measurement target is calculated by obtaining the time required for the measurement target to move at a predetermined interval by using the imaging signal generated by the second imaging device.

[0002]

2. Description of the Related Art Conventionally, in a speed measurement system for measuring the speed of a moving body, measurement using the Doppler effect, measurement using an optical sensor, measurement using a pass sensor, or measurement using an image sensor is performed. ing.

[0003]

By the way, in the speed measurement using the Doppler effect, a sensor must be installed facing the traveling direction of the object to be measured. The speed cannot be measured correctly. Further, in the measurement using an optical sensor such as a laser displacement meter, the speed cannot be measured and the apparatus is expensive unless the measurement range is narrow and the trajectory of the measurement target is not accurately positioned. In the case where the passage sensor is used, if the object to be measured is long in the traveling direction, when acceleration or deceleration occurs while passing through the sensor portion, the speed cannot be measured correctly. Furthermore, the movement of the measurement target is detected by photographing the measurement target using an area sensor that is a two-dimensional image sensor, dividing the obtained captured image into a plurality of blocks, and performing block matching processing between frames. However, when calculating the speed based on the moving distance, sometimes the speed cannot be measured correctly due to the parallax caused by the depth.

Accordingly, the present invention provides a speed measuring system capable of easily and accurately measuring a speed.

[0005]

A speed measurement system according to the present invention generates a photographing signal of a measurement target using a one-dimensional image sensor, and at a predetermined interval along a traveling direction on a side of the measurement target. Using the arranged first and second imaging devices and the imaging signals generated by the first and second imaging devices, determine the time required for the measurement target to move at a predetermined interval, and And a control device for calculating and outputting the speed.

The control device includes a signal storage unit for storing a first image pickup signal from the first image pickup device and a second image pickup signal from the second image pickup device, and a first and a second signal from the signal storage unit. 2 is read out, the second image pickup signal is shifted in the time axis direction with respect to the first image pickup signal,
A correlation search means for searching for a shift amount at which the correlation between the imaging signal of the second imaging signal and the second imaging signal is maximized, and a time required for the measurement object to move a predetermined interval, and a search result obtained by the search means It has a speed calculating means for calculating the speed of the measurement target using time, and a measurement control means for controlling the operations of the first and second imaging devices, the signal storing means, the searching means and the speed calculating means. Further, there is provided flicker removing means for removing flicker components from the first and second image pickup signals using a comb-shaped filter, and the search means includes first and second flicker components removed by the flicker removing means. The shift amount that maximizes the correlation is searched for using the imaging signal.

In the present invention, the first and second image pickup devices using a one-dimensional image sensor, for example, a linear sensor, are arranged at predetermined intervals along the traveling direction to the side of the object to be measured, and the sensor of the linear sensor is used. They are arranged so that the arrangement direction is substantially perpendicular to the traveling direction of the measurement target. The imaging signals generated by the first and second imaging devices are sequentially stored in the signal storage means of the control device, and
The first and second imaging signals are read from the signal storage means. Either the read first imaging signal or the second imaging signal is shifted in the time axis direction, and a correlation coefficient between the first imaging signal and the second imaging signal is calculated. The time required for the measurement object to move a predetermined interval is obtained based on the shift amount at which the correlation coefficient becomes maximum. By dividing the predetermined interval by the time required to move the predetermined interval, the speed of the measurement target is calculated and output.

[0008]

Next, an embodiment of the present invention will be described in detail with reference to the drawings. FIG. 1 shows the overall configuration of the speed measuring device. The speed measuring device includes two imaging devices 10a and 10b, and the respective imaging devices 10a and 10b.
The control device 30 calculates the speed using the imaging signals DMa and DMb obtained in the step (1). The two imaging devices 10a and 10b are arranged at an interval D in the traveling direction of the measurement target.

FIG. 2 is a diagram showing the configuration of the imaging device and the control device. The two imaging devices 10a and 10b have the same configuration, and only the imaging device 10a will be described, and the description of the imaging device 10b will be omitted.

Light incident through a lens 11a of an image pickup device 10a is applied to a linear sensor 12a which is a one-dimensional image sensor. Here, in the imaging device 10a, the sensor arrangement direction of the linear sensor 12a (the arrangement direction of the photodiodes described later) is substantially perpendicular to the traveling direction of the measurement target, that is, the measurement target is in the X direction as shown in FIG. When it moves to, it is installed so as to be substantially in the Y direction. Further, the imaging device 10a and the imaging device 10b are installed so that the sensor arrangement directions are substantially parallel.

For this reason, the exposure is performed in a point-like manner as shown in FIG. 3A in the X direction which is the traveling direction of the measurement object, and is linearly exposed in the Y direction as shown in FIG. 3B. Is performed, the parallax is reduced with respect to the traveling direction of the measurement target, and the influence of unevenness on the surface of the measurement target and a change in depth can be reduced. In addition, since the imaging range in the Y direction is wide, it is possible to measure the speed without accurately positioning the trajectory of the measurement target.

FIG. 4 shows the configuration of the linear sensor.
In this linear sensor 12a, a plurality of photodiodes 1
21-1 to 121-n are linearly arranged adjacent to each other, and each photodiode generates a charge corresponding to the irradiated light. The generated charges are read out to the CCD shift register 123 via the readout gate 122, and
The reading of this charge is controlled by the read signal ROG supplied to the read gate 122.

In the CCD shift register 123, charges read out from the respective photodiodes are sequentially transferred in the direction of arrow A based on the two-phase driving pulses φ1 and φ2, and are stored in the output capacitor 124. Therefore, the output capacitor 124
Terminal voltage Vc changes according to the amount of accumulated electric charge.

A sample and hold circuit 125 and a transistor 126 are connected to the output capacitor 124. The sample and hold circuit 125 samples and holds the terminal voltage Vc based on the sample and hold control signal HLD, and holds the sampled and held voltage. A signal indicating the voltage level is generated and output as the imaging signal Sa.

The transistor 126 has a reset signal RS
T is supplied and when the transistor 126 is turned on by the reset signal RST, the output capacitor 1
The charge stored in 24 is released.

FIG. 5 shows the operation of the linear sensor. When the read signal ROG shown in FIG. 5A is set to the signal level “LRS” at time t 1, the charges generated by each photodiode 121 are read out to the CCD shift register 123 via the read gate 122. When the signal level is changed to "LRE" at the time point t2, the reading of the electric charge is completed.

When the supply of the two-phase driving pulses φ1 and φ2 shown in FIGS. 5B and 5C to the CCD shift register 123 is started at time t3, the charges read from the photodiodes 121 are sequentially transferred by the CCD shift register 123. The data is transferred and stored in the output capacitor 124. For this reason, the voltage level of the terminal voltage Vc of the output capacitor 124 changes according to the accumulated charge amount as shown in FIG. 5E.

Here, in order to obtain an image signal Sa of a desired area of the linear sensor 12a, for example, the image signal Sa based on the electric charge generated by each of the photodiodes 121-a to 121-b shown in FIG. If it is obtained, the signal level of the reset signal RST shown in FIG. 5D is set to the low level "L" at the time t4 when the charge read from the photodiode 121- (a-1) is output from the CCD shift register 123. When the transistor 126 is turned on, the electric charge stored in the output capacitor 124 is released.

Thereafter, the electric charge read from the photodiode 121-a is stored in the output capacitor 124, and the electric charge read from each photodiode is subsequently stored in the output capacitor 124.

The last photodiode 121 in the desired area
When the electric charge read out from -b is stored in the output capacitor 124, the sample hold control signal HL shown in FIG.
When the signal level of D becomes low level "L" at time t5, the terminal voltage Vc of the capacitor is sampled and held, and a signal indicating the voltage level of the terminal voltage Vc at this time is output as the imaging signal Sa. Is done.

The read signal ROG and the two-phase drive pulse φ
1, φ2, the reset signal RST, and the sample hold control signal HLD are supplied from a measurement control unit 36 of the control device 30, which will be described later.

The image signal Sa generated by the linear sensor 12a is supplied to the A / D converter 13a. The A / D converter 13a converts the image signal Sa into a digital signal and supplies it as an image signal DMa to the control device 30 shown in FIG.

Similarly, the linear sensor 12b of the image pickup device 10b generates the image pickup signal Sb, converts the image pickup signal Sb into the image pickup signal DMa by the A / D converter 13b, and supplies it to the control device 30.

The imaging devices 10a and 10b sequentially generate imaging signals DMa and DMb at intervals of a predetermined time τ and supply the signals to the control device 30, and the control device 30 receives the imaging signals DMa from the imaging device 10a. The image data DMb from the image pickup device 10b is sequentially written to the ring buffer 31b.

Flicker elimination circuits 32a and 32b are connected to the ring buffers 31a and 31b. In this flicker elimination circuit, when the exposure time is short when measuring the speed, the flicker of the image signal DM causes the flicker of illumination. In order to prevent the level from fluctuating and making measurement impossible, the influence of flicker is removed from the imaging signal DM.

Here, the removal of the influence of flicker will be described with reference to FIG. FIG. 6A shows the spectrum of the imaging signal DM which is not affected by flicker. Assuming that the period of the flicker is “T”, the spectrum of the flicker signal FL has components at positions that are integral multiples of the fundamental frequency 1 / T as shown in FIG. 6B. For this reason, the frequency band of the imaging signal DM is lower than the frequency band of the flicker signal FL, and the flicker signal F
When L acts multiplicatively on the image signal DM, the spectrum of the image signal DM affected by flicker has a periodic shape around the spectrum of the flicker signal FL as shown in FIG. 6C. Appears.

For this reason, the frequency n / T (n is a positive integer)
By applying a filter having a comb characteristic as shown in FIG. 6D having a gain at the position of, a signal DFE having a shape similar to the flicker spectrum can be obtained. The signal DFE is obtained by applying a gain only to the DC component of the imaging signal. By dividing the image signal DM affected by flicker by the signal DFE obtained in this way, it is possible to obtain an image signal DMC from which the effect of flicker has been removed. As a filter, for example, IIR (infinite impu
If an lse response) type filter is used, the filter characteristics can be made steep.

FIG. 7 is a diagram showing a configuration of the flicker removing circuit 32a. The flicker removal circuits 32a and 32b have the same configuration, and a description of the flicker removal circuit 32b will be omitted.

The image pickup signal DMa read from the ring buffer 31a is supplied to a coefficient multiplier 321 constituting an IIR filter and a divider 325.

The IIR type filter includes a coefficient multiplier 321,
324, an adder 322, and a delay unit 323. The coefficient multiplier 321 adds a coefficient (1
-G) and supplies it to the adder 322 as a signal DKA. The adder 322 is supplied with a signal DFG from a coefficient multiplier 324 described later, and outputs a signal DKA and a signal DFG.
Are added to generate a signal DFE. This signal DFE is
It is supplied to the delay unit 323 and the divider 325.

The delay unit 323 delays the signal DFE by one cycle of the flicker frequency and supplies the delayed signal DFE to the coefficient multiplier 324 as a delayed signal DFF. The coefficient multiplier 324 multiplies the delayed signal DFF supplied from the delay unit 323 by a coefficient G to generate a signal DFG. This signal DFG
Is supplied to the adder 322 described above.

The signal DFE obtained from the adder 322 in this manner is a signal having a shape similar to the flicker spectrum, and this signal DFE is supplied to a divider 325,
By multiplying the imaging signal DMa by the signal DFE, it is possible to obtain an imaging signal DMCa from which the influence of flicker has been removed. Similarly, the flicker removing circuit 32b can remove the influence of flicker and obtain the image signal DMCb.

Here, assuming that the spatial frequency in the traveling direction of the surface pattern of the object to be measured is “FS” and the speed of the object to be measured is “SV”, the frequency band W of the image signal DM
B is “WB” = “FS” × “SV”. If the frequency band WB of the image signal DM is not smaller than the flicker frequency, the filter is spatially filtered like an optical filter to remove a pattern having a high spatial frequency, thereby reducing the influence of flicker by the above-described method. Can be removed.

If the period "T" of the flicker is not clear, the frequency component is analyzed by performing a fast Fourier transform using the image pickup signal DM, and the frequency pitch is estimated from the complex spectrum to determine the frequency pitch. be able to. By performing the above-described processing based on the estimated flicker frequency, the influence of flicker can be removed.

The image pickup signals DMCa and DMCb output from the flicker elimination circuits 32a and 32b are applied to the correlation search circuit 3
3 is supplied. In the correlation search circuit 33, the correlation between the imaging signal DMCa and the imaging signal DMCb is checked by statistical processing.

Here, assuming that the data sequence of the image signal DMCa from the flicker elimination circuit 32a is {ai} and the data sequence of the image signal DMCb from the flicker elimination circuit 32b is {bi}, the length of the data sequence (number of samples) Is N, the average value AVGa of the data sequence {ai} and the average value AVGb of the data sequence {bi} are calculated using the equations (1) and (2).

[0037]

(Equation 1)

The standard deviation σa of the data sequence {ai} and the standard deviation σb of the data sequence {bi} are calculated by using equations (3) and (4). Further, the covariance σab is calculated using Expression (5).

[0039]

(Equation 2)

Using the standard deviation σa, standard deviation σb, and covariance σab calculated in this way, the correlation coefficient R can be calculated as shown in equation (6).

[0041]

(Equation 3)

Here, a shift amount Wsh at which the correlation coefficient R becomes the maximum is searched for while shifting one of the data sequence {ai} or the data sequence {bi} in the time direction with reference to the other. When the shift amount Wsh at which the correlation coefficient R becomes the maximum is found, the data sequence time difference τsh is obtained by multiplying the shift amount Wsh by the predetermined time τ. The data sequence time difference τsh is the time required for the measurement target to move by the interval D, and a signal TSH indicating the data sequence time difference τsh is supplied to the speed calculation circuit 34. If the data sequence {ai} and the data sequence {bi} have a time difference (hereinafter referred to as a “read time difference”) before shifting the other in the time direction, the shift amount Wsh and the predetermined time τ are determined. Time obtained by multiplication, data sequence {ai} and data sequence {bi}
Data string time difference τsh
It is said.

The speed calculation circuit 34 includes two imaging devices 1
The information indicating the interval D between 0a and 10b is supplied from the measurement control unit 36 or the two imaging devices 10a and 10b
When the interval is fixed at a predetermined interval in advance, the interval D is stored in advance in the speed calculating circuit 34, and the speed calculating circuit 34 divides the interval D by a data sequence time difference τsh based on the signal TSH. , The speed of the object to be measured can be calculated. Further, in the speed calculation circuit 34,
A speed signal MS indicating the calculated speed is generated and supplied to the speed output unit 35.

The speed output unit 35 is constituted by using a display element, a printer, or the like, and displays or prints out the speed measured based on the supplied speed signal MS. Further, information indicating the measured speed is supplied from the speed output unit 35 to an external data processing device.

An operation unit 37 is connected to the measurement control unit 36. When the interval between the two imaging devices 10a and 10b is not fixed at a predetermined interval in advance, the two imaging devices 10
The interval D between a and 10b is input using the operation unit 37.
Further, operations such as setting of a desired region for the linear sensors 12a and 12b and a measurement operation of speed measurement are performed using the operation unit 37.

In the measurement control unit 36, the read signal RO
G, two-phase drive pulses φ1, φ2, a reset signal RST, and a sample-and-hold control signal HLD are generated and supplied to the linear sensors 12a and 12b. Also, the ring buffer 31
a, 31b, control of readout of the imaging signals DMa, DMb from the flicker removal circuits 32a, 32b, processing of removing the influence of flicker, and search of the time shift amount at which the correlation coefficient R is maximized by the correlation search circuit 33. The control signal CT for performing the control is generated and supplied to each circuit. Further, a signal indicating the interval D input using the operation unit 37 is generated to generate the speed calculation circuit 34.
Is also performed.

Next, the operation of the speed measuring device will be described. First, the imaging devices 10a and 10b are arranged at an interval D with respect to the traveling direction of the measurement target, and the areas of the photodiodes for generating the imaging signals Sa and Sb are set by the linear sensors 12a and 12b. In this region setting, the region is set so as to include the region of the photodiode where the image of the measurement target is formed. Note that the imaging devices 10a and 10b
In the case where the intervals are constant, they are arranged so as to have a predetermined interval. If the distance between the imaging devices 10a and 10b can be freely set, the distance between the imaging devices is input using the operation unit 37.

When the object to be measured is photographed by the imaging devices 10a and 10b, imaging signals DMCa and DMCb are generated at predetermined time intervals τ, and the ring buffer 31 of the control device 30 is generated.
a and 31b are sequentially written. In the control device 30, the speed is calculated using the imaging signals DMCa and DMCb written in the ring buffers 31a and 31b.

FIG. 8 is a flowchart showing the operation of the control device 30. In step ST1, the imaging signal DMC written in the ring buffers 31a and 31b
The data strings of a and DMCb are read, and the process proceeds to step ST2. Note that the imaging signals D are output from the ring buffers 31a and 31b.
When reading MCa and DMCb, the ring buffer 31
The two imaging signals DMCa and DMCb depend on the reading position of the imaging signal DMCa read from a and the reading position of the imaging signal DMCb read from the ring buffer 31b.
Is apparent. For example, when the measurement target is photographed, the imaging signals DMCa and DMCb are sequentially written from the same address position of the ring buffers 31a and 31b, and the imaging signals DMCa and D are read from the same address position.
When reading MCb, the read time difference between the read image signals DMCa and DMCb is “0”. Also,
If the read image signal DMCa and the image signal DMCb have different address positions, the difference between the addresses is equal to the predetermined time τ.
Is the read time difference.

In step ST2, the correlation coefficient maximum value Rma
x and the shift amount Wsh are initialized, and the process proceeds to step ST3.

In step ST3, the imaging signals DMCa,
The correlation coefficient R is calculated using the DMCb data sequence, and the process proceeds to step ST4.

In step ST4, the calculated correlation coefficient R is compared with the correlation coefficient maximum value Rmax. Here, when the calculated correlation coefficient R is larger than the correlation coefficient maximum value Rmax, the process proceeds to step ST5, where the calculated correlation coefficient R is stored as the correlation coefficient maximum value Rmax. The shift amount Wsh is stored, and the process proceeds to step ST6. If the calculated correlation coefficient R is not larger than the maximum correlation coefficient value Rmax, the process proceeds to step ST6.

In step ST6, the imaging signal is shifted. Here, for example, when the image signal Sa is used as a reference, the data sequence of the image signal Sb that is not a reference is shifted, and the process proceeds to step ST7.

In step ST7, it is determined whether or not the shift amount of the image signal Sb is within a preset shift amount range. The shift amount range is set based on the specification of the detection speed range, and regulates the shift amount so that a speed exceeding the detection speed range is not detected. Here, when the shift amount is within the preset shift amount range, the process returns to step ST3, and when the shift amount exceeds the preset range, the process proceeds to step ST8.

In step ST8, the data string time difference τsh is calculated by adding the time obtained by multiplying the shift amount Wsh when the correlation coefficient R is maximized by the predetermined time τ to the read time difference in step ST1. Is done. For this reason, for example, the elapsed time from when the measurement target is photographed by the imaging device 10a to when the correlation coefficient is maximized, that is, when the measurement target is photographed by the imaging device 10b, is calculated as the data sequence time difference τsh. Proceed to step ST9.

H in step ST9, the imaging devices 10a, 1
By dividing the interval D of 0b by the data sequence time difference τsh, an average speed from when the measurement target is photographed by the imaging device 10a to when it is photographed by the imaging device 10b is calculated, and the process proceeds to step ST10.

In step ST10, the calculated average speed is displayed or printed out, and the process returns to step ST1.

As described above, according to the above-described embodiment,
By using the linear sensor, the speed of the object to be measured can be easily and correctly measured from the side. Also, for example, even if the measurement object moving in the horizontal direction shakes in the vertical direction, the average speed can be measured correctly. Furthermore, in the case where the speed is measured under illumination with flicker, the speed can be correctly measured by using the flicker removing circuits 32a and 32b.

[0059]

According to the present invention, the first and second imaging devices for generating a photographing signal of a measurement target using a one-dimensional image sensor are provided at predetermined intervals along the traveling direction to the side of the measurement target. And the speed of the measurement target is calculated by determining the time required for the measurement target to move at a predetermined interval using the imaging signals generated by the first and second imaging devices. For this reason, the speed can be measured with an easy and inexpensive configuration from the side of the measurement object without installing a sensor directly in the traveling direction of the measurement object.

The first and second imaging devices are arranged so that the sensor array direction of the one-dimensional image sensor is in a direction orthogonal to the traveling direction of the object to be measured. For this reason, it is possible to accurately measure the speed while reducing the influence of the unevenness and the change in the depth of the surface of the measurement target, and to accurately measure the speed without accurately positioning the trajectory of the measurement target.

Further, the second image signal is shifted in the time axis direction with respect to the first image signal, and a shift amount at which the correlation between the first image signal and the second image signal is maximized is searched. Since the time required for the measurement target to move at the predetermined interval is obtained, the speed can be measured even when the distance between the measurement target and the imaging device changes.

Further, the shift means having the maximum correlation is searched by the search means using the first and second image pickup signals from which the flicker component has been removed by the flicker removal means. Can measure the speed correctly.

[Brief description of the drawings]

FIG. 1 is a diagram showing a configuration of a speed measurement system.

FIG. 2 is a diagram illustrating a configuration of an imaging device and a control device.

FIG. 3 is a diagram illustrating an installation state of an imaging device.

FIG. 4 is a diagram illustrating a configuration of a linear sensor.

FIG. 5 is a diagram for explaining a driving method of the linear sensor.

FIG. 6 is a diagram for explaining the principle of flicker removal.

FIG. 7 is a diagram illustrating a configuration of a flicker removal circuit.

FIG. 8 is a flowchart showing the operation of the control device.

[Explanation of symbols]

10a, 10b ... imaging device, 11a, 11b ... lens, 12a, 12b ... linear sensor, 13a, 13b ...
..A / D converters, 30 ... control devices, 31
a, 31b ring buffer, 32a, 32b flicker removal circuit, 33 correlation search circuit, 34
Speed calculation circuit, 35: Speed output unit, 36: Measurement control unit, 37: Operation unit, 321, 324: Coefficient multiplier, 322: Adder, 323: Delay unit , 3
25 ... divider

Claims (4)

[Claims] 1. A first and a second imaging device that generate a photographing signal of a measurement target using a one-dimensional image sensor and are arranged at predetermined intervals along a traveling direction on a side of the measurement target, A control for calculating and outputting the speed of the measurement target by calculating the time required for the measurement target to move through the predetermined interval using the imaging signals generated by the first and second imaging devices. A speed measurement system comprising a device. 2. The apparatus according to claim 1, wherein the first and second imaging devices are arranged such that a sensor array direction of the one-dimensional image sensor is a direction orthogonal to a traveling direction of the measurement target. Item 2. The speed measurement system according to Item 1. 3. The control device, comprising: a signal storage unit configured to store a first imaging signal from the first imaging device and a second imaging signal from the second imaging device; and The first and second imaging signals are read out, the second imaging signal is shifted in the time axis direction with respect to the first imaging signal, and the first imaging signal and the second imaging signal are shifted. Correlation search means for searching for the shift amount at which the correlation is maximized, and calculating the time required for the measurement object to move through the predetermined interval, and for the measurement object using the time obtained by the search means A speed calculation unit for calculating a speed, the first and second imaging devices, and a measurement control unit for controlling operations of the signal storage unit, the search unit, and the speed calculation unit. Speed measurement system described in Item 1 Beam. 4. The control device includes flicker removing means for removing flicker components from the first and second image pickup signals using a comb-shaped characteristic filter. The speed measurement system according to claim 1, wherein a shift amount having a maximum correlation is searched using the first and second imaging signals from which components have been removed.