JP2015025657A - Speed measurement device and method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a speed measurement device and method which can measure a ground speed more accurately than possible in prior art, and which is simple in configuration.SOLUTION: Provided is a speed measurement device and method for measuring a speed value of a moving body when the moving body moves at a prescribed speed in a prescribed advance direction, wherein the speed measurement device, having a main scanning direction image-capturing line in a direction substantially parallel to the advance direction of the moving body, captures the image of an object fixedly placed to ground facing the moving body repeatedly in a sub-scanning direction at prescribed time intervals and outputs an image signal, calculates a concentration gradient of the image including a texture included in the image signal when the moving body moves and formed in shape of a stripe based on the object, converts a slope angle into the speed value of the moving body using a relation of the slope angle being proportional to the speed value, and outputs the converted value.

Description

  The present invention relates to a speed measuring apparatus and method for measuring the speed of a moving body such as a vehicle.

In general, it is difficult to accurately obtain the ground speed from a moving body such as a vehicle, and in many cases, the ground speed is calculated by accumulating wheel rotation pulses and converting the accumulated pulses into speeds (for example, (See Patent Document 1). For example, the speed calculation method disclosed in Patent Document 1 is characterized in that the speed V _{n of the} vehicle 1 is calculated from speed pulses generated by the rotation of an axle that rotates the wheels in order to calculate an accurate speed. Specifically, when no speed pulse is generated in at least one of the plurality of speed calculation time intervals TS obtained by equally dividing the pulse count time interval TZ, the number and arrangement of speed calculation time intervals TS in which no speed pulse is generated; Based on the speed pulse number in the speed calculation time interval TS at which the speed pulse is generated, any one of the predetermined speed, the speed V _n−1 calculated previously, and the speed calculation value V ′ _n is set as the speed V _n. The speed calculation value V ′ _n is calculated based on the number P of speed pulses generated within the pulse count time interval TZ and the time interval T from the first speed pulse to the last speed pulse within the pulse count time interval TZ. .

JP 2005-283525 A

  However, in the case of a vehicle, phenomena such as idling and sliding frequently occur, and there is a problem that an accurate vehicle speed cannot be obtained. There are also proposals for techniques to compensate for idling by using pulses from multiple axles, and methods such as using free wheels only for speed measurement. There was a problem that became large.

  On the other hand, as existing technologies for measuring ground speed in a non-contact manner, there are laser and microwave Doppler velocimeters and spatial filter velocimeters. (1) In the case of microwaves, laws and regulations are strict in each country. (2) In the laser and spatial filter type, there is a problem in that an accurate speed cannot be calculated because the gap between sleepers is a complete space and the objective distance fluctuates extremely.

  An object of the present invention is to solve the above-described problems, and to provide a speed measuring device and method that can measure the ground speed more accurately than the prior art and that has a simple configuration.

A speed measuring device according to a first invention is a speed measuring device that measures a speed value of the moving body when the moving body moves in a predetermined traveling direction at a predetermined speed,
An object having an imaging line in the main scanning direction in a direction substantially parallel to the traveling direction of the moving body, and an object fixedly placed on the ground facing the moving body at a predetermined time interval. Imaging means for repeatedly imaging in the sub-scanning direction not parallel to the scanning direction and outputting an image signal;
Based on the image signal output from the imaging means, a density gradient of an image including a stripe-shaped texture included in the image signal and based on the object is calculated when the moving body moves. Speed value conversion means for converting the gradient angle into a speed value of the moving body using a relationship proportional to the speed value and outputting the speed value is provided.

  In the speed measuring device, the sub-scanning direction is set to be substantially orthogonal to the main scanning direction, and the speed value converting means includes a first direction corresponding to the main scanning direction and the sub-scanning direction. Calculating a gradient angle based on a concentration gradient in the second direction with respect to a concentration gradient in the first direction, and calculating in a second direction corresponding to the direction and perpendicular to the first direction. And

  In the speed measurement device, after filtering the image signal output from the imaging unit so as to remove a predetermined noise component different from the image component related to the striped texture based on the object, the speed Further, a filter means for outputting to the value conversion means is further provided.

  Further, in the speed measurement device, the speed value conversion unit calculates an average value of the density gradient related to the image signal when calculating the density gradient of the image including the striped texture based on the object, and The gradient angle is calculated based on an average value of the concentration gradient.

  Still further, in the speed measurement device, the speed value conversion means changes the measurement accuracy of the speed value by changing the resolution of the image signal with respect to the image signal output from the imaging means. Features.

  Further, in the speed measurement device, the speed value conversion means generates a plurality of image signals having different resolutions, and selects an image signal that can obtain a predetermined speed measurement accuracy from the plurality of image signals. A density gradient of an image including the stripe-shaped texture is calculated based on the selected image signal, and converted to a velocity value of the moving body.

A speed measuring method according to a second invention is a speed measuring method for measuring a speed value of the moving body when the moving body moves in a predetermined traveling direction at a predetermined speed,
The imaging means has an imaging line in the main scanning direction in a direction substantially parallel to the traveling direction of the moving body, and the imaging means is configured to detect an object fixedly placed on the ground facing the moving body. A step of repeatedly imaging in a sub-scanning direction not parallel to the main scanning direction at a predetermined time interval and outputting an image signal;
Based on the image signal output from the imaging means, the control device calculates the density gradient of the image including the striped texture included in the image signal and based on the object when the moving body moves, Converting the gradient angle into a velocity value of the moving body using a relationship in which the gradient angle is proportional to the velocity value, and outputting the velocity value.

  In the speed measurement method, the sub-scanning direction is set to be substantially orthogonal to the main scanning direction, and the step of converting to the speed value and outputting is performed by the control device corresponding to the main scanning direction. And the second direction corresponding to the sub-scanning direction and perpendicular to the first direction, and the density gradient in the second direction with respect to the density gradient in the first direction is calculated. And calculating a gradient angle based thereon.

  In the speed measurement method, the control is performed after filtering the image signal output from the imaging unit so as to remove a predetermined noise component different from the image component related to the striped texture based on the object. The method further includes a step of outputting to the apparatus.

  Further, in the speed measurement method, the step of converting to the speed value and outputting the speed value relates to the image signal when the control device calculates a density gradient of an image including a striped texture based on the object. The method includes calculating an average value of the concentration gradient and calculating the gradient angle based on the average value of the concentration gradient.

  Furthermore, in the speed measurement method, the control device performs a decimation filter process on the image signal output from the imaging unit to change the resolution of the image signal, thereby increasing the measurement accuracy of the speed value. The method further includes a step of changing.

  Further, in the speed measurement method, the step of converting to the speed value and outputting the speed value includes generating a plurality of image signals having different resolutions and performing a predetermined speed measurement among the plurality of image signals. Selecting an image signal capable of obtaining accuracy, calculating a density gradient of an image including the stripe-shaped texture based on the selected image signal, and converting the calculated gradient into a velocity value of the moving body, To do.

  Therefore, according to the speed measuring apparatus and method of the present invention, it is possible to provide a speed measuring apparatus and method that can measure the ground speed more accurately than the prior art and that has a simple configuration. Therefore, the speed measurement apparatus and method according to the present invention explained in the column of the problem to be solved by the invention “(1) In the case of microwaves, laws and regulations are severe in each country. (2) In laser and spatial filter type In the bridge, the space between sleepers is a complete space, and the problem that the accurate speed cannot be calculated because the objective distance fluctuates extremely can be solved. For example, by measuring the ground speed of the vehicle, the position of the vehicle can be measured more accurately in the vehicle inspection system, and the inspection accuracy is greatly improved. Further, in the running state of the vehicle, it is possible to accurately detect and measure the idling / sliding state together with the conventional speedometer.

It is a perspective view which shows the external appearance of the speed measuring device which concerns on one Embodiment of this invention. It is a block diagram which shows the structure of the image processing apparatus 3 of FIG. It is a figure which shows an example of the captured image by the speed measuring device of FIG. 1, Comprising: (a) is equipped with the speed measuring device of FIG. It is a top view which shows visual field A1 of the line sensor camera 1 which images the sleeper 22, and (b) is a top view which shows an example of the image taken in by the line sensor camera 1 of FIG. SOBEL used a gradient angle calculation circuit 14 of FIG. 2 is a diagram showing an example of an operator _{f x.} SOBEL used a gradient angle calculation circuit 14 of FIG. 2 is a diagram showing an example of an operator _{f y.} 2 is a modification of the gradient angle calculation circuit 14 shown in FIG. 2, in which the density on the line is integrated while changing the starting point and angle on the image, and the direction with the largest difference between white and black is detected as the fringe angle. It is a figure of the image which shows a method. It is a modification of the image processing apparatus 3 of FIG. 1, Comprising: In each case obtained by changing speed, the image of the resolution for high speed, the image of the resolution for medium speed, and the resolution for low speed obtained by changing the resolution It is a figure which shows an image.

  Hereinafter, embodiments according to the present invention will be described with reference to the drawings. In addition, in each following embodiment, the same code | symbol is attached | subjected about the same component.

  FIG. 1 is a perspective view showing an appearance of a speed measuring apparatus according to an embodiment of the present invention, and FIG. 2 is a block diagram showing a configuration of the image processing apparatus of FIG. 3 is a diagram showing an example of an image captured by the speed measurement device of FIG. 1. FIG. 3 (a) shows a pair of rails when the speed measurement device of FIG. It is a top view which shows visual field A1 of the line sensor camera 1 which images the some sleepers 22 on 21a, 21b, FIG.3 (b) is a top view which shows an example of the image taken in by the line sensor camera 1 of FIG. It is.

  In FIG. 1, the speed measurement device according to the present embodiment is installed, for example, under the floor of a moving body that is a railway vehicle, and a plurality of sensor elements are juxtaposed as shown in FIG. The line sensor camera 1 that images the plurality of sleepers 22 on the pair of rails 21a and 21b, the laser line projector 2 that illuminates when the imaging target is dark, and the image data captured by the line sensor camera 1 And an image processing device 3 that calculates a speed value of the railway vehicle and outputs a speed output signal 5 to the speed indicator 4. An image line of the line sensor camera 1 in the main scanning direction (referred to as a juxtaposition direction of a plurality of center elements) is arranged substantially in parallel with the moving direction of the moving body, and the image signal in a fixed cycle (at predetermined time intervals). Repeatedly acquire digital data. Here, output signals from each center element per sampling are arranged in the X direction corresponding to the main scanning direction to obtain sampling data, and the sampling data sequentially corresponds to the sub scanning direction substantially orthogonal to the main scanning direction. Images are arranged in the -Y direction (time direction in FIG. 3). When the moving body has a relative movement with respect to the ground, as shown in FIG. 3B, for example, the image pattern of the sleeper 22 is an image including a striped texture (pattern) inclined obliquely in proportion to the speed. Become a pattern.

  In the present embodiment, the main scanning direction and the sub-scanning direction are substantially perpendicular to each other. However, the present invention is not limited to this, and the above-described stripes may be used as long as the direction is not at least parallel to the main scanning direction. A shape image can be obtained. Further, if the above cycle is associated with the data interval in the Y direction in FIG. 3, it is not always necessary to perform sampling at a constant cycle. For example, even when different cycles are used such that the data interval in the Y direction is 1 pixel when the time interval is 0.1 second and the data interval in the Y direction is 2 pixels when the time interval is 0.2 seconds, A shape image can be obtained.

  In FIG. 1, A <b> 1 indicates a field area of the line sensor camera 1, and A <b> 2 indicates a laser light illumination area of the laser line projector 2. Note that the laser line projector 2 may not be equipped when the imaging target is bright, or may be turned off. Further, since the visual field area A1 of the line sensor camera 1 is line-shaped, the laser line projector 2 can be configured compactly using an illumination device such as a laser slit projector.

  FIG. 2 is a block diagram showing the configuration of the image processing apparatus 3 of FIG. In FIG. 2, the image processing device 3 includes an image capturing circuit 11 having a buffer memory 11m, a low-pass filter (LPF) 12, a differential filter 13, a gradient angle calculation circuit 14, and a velocity value conversion circuit 15. It is configured with. Note that the constituent elements 11 to 15 of the image processing apparatus 3 may be collectively configured by a digital control device such as a digital signal processor.

In FIG. 2, an image capturing circuit 11 converts an image signal from the line sensor camera 1 into image digital data by sampling at a predetermined sampling frequency such as 30 kHz using a predetermined internal clock, and converts the image signal into a frame. Each time it is taken into the buffer memory 11m and temporarily stored, it is output to the low-pass filter 12 for each frame. Next, the low-pass filter 12 is an example of a noise cut filter for suppressing noise from an input signal, and outputs the signal after noise suppression to the differential filter 13. Here, the low-pass filter 12 removes at least a predetermined noise component different from the image component related to the striped texture based on the object. Furthermore, differential filter 13, for example, by using a SOBEL operator _{f x} of X direction parallel to the vehicle traveling direction as shown in FIG. 3, a SOBEL operator _{f y} in the Y direction corresponding to the time direction, described in detail later differentiation By executing the filter processing, a density gradient that becomes a peak at each pixel of the input image is locally calculated in the X direction and the Y direction, and is output to the gradient angle calculation circuit 14. The gradient angle calculation circuit 14 calculates the average value of the density gradients in all the pixels by adding the density gradients in the X direction and Y direction of each pixel that are input, and calculates the gradient of the entire image from the average value. The angle is calculated using the following equation and output to the speed value conversion circuit 15.

  Next, the speed value conversion circuit 15 obtains in advance a relationship with the speed of the moving body corresponding to the gradient angle θ in the apparatus (a proportional relation in which the gradient angle increases as the speed of the moving body increases). A coefficient (proportional constant) (or a conversion table may be stored in an internal memory in the speed value conversion circuit 15, and a speed value corresponding to the gradient angle θ input using the relationship is calculated, A speed output signal indicating the calculated speed value is output to the speed indicator 4.

で強度を計算することができる。また、

でエッジの方向を計算することができる。そして、本実施形態では、勾配角度θを上記式（１）を用いて計算することができる。 4A is a diagram showing an example of a SOBEL operator _{f x} used a gradient angle calculation circuit 14 of FIG. 2, FIG. 4B is a diagram showing an example of a SOBEL operator _{f y} used in the gradient angle calculation circuit 14 of FIG. Here, SOBEL operator is an operator for edge extraction, a method for determining the intensity gradient at a local product-sum operation (differential value), for example, with respect to f _y of f _x and 4B of Figure 4A And

The intensity can be calculated with Also,

The edge direction can be calculated with In this embodiment, the gradient angle θ can be calculated using the above equation (1).

In the above embodiments, the differential filter 13 and the gradient angle calculation circuit 14, the total differential processing (a SOBEL operator _{f x} image data after treatment, after treatment SOBEL operator _{f y} in the Y direction corresponding to the time direction A process of adding and synthesizing the image data) With respect to the image data after being subjected to density projection while changing the gradient angle θ, by evaluating the variance of the obtained density distribution, the angle at which the variance value is maximized is selected. Can also be converted to a speed value.

In the above embodiments, SOBEL operator f _x in a gradient angle calculation circuit _14, but is seeking a concentration gradient with f _y, the present invention is not limited to this, as shown in FIG. 5, starting on the image The direction with the largest difference between white and black may be detected as the fringe angle by integrating the density on the line while changing the point and angle. Processing of the direction, SOBEL operator _f x, may be performed after the pre-treatment or treatment of _{f y,} each SOBEL operator _f x, may be used after processing only one of _{f y.}

  In the embodiment of FIG. 2, the decimation process is not performed, and the low-pass filter 12 and the differential filter 13 which are examples of the noise cut filter are used. However, the present invention is not limited thereto, As shown below, the decimation process may be executed a plurality of times using a decimation filter. For example, 1000 × 1000 image data is converted in resolution to, for example, 1000 × 250 image data or 250 × 1000 image data. At this time, the low-pass filter 12 necessary for avoiding aliasing is applied and then re-sampling (decimation). The low-pass filter 12 used here is different from the filter when the decimation process is not performed, in order to remove noise such as aliasing noise and limit the band, among the frequency components of the input image digital data, Only frequency components with a frequency equal to or lower than half the frequency are low-pass filtered and output. Thereby, it is possible to obtain a high accuracy by expanding the detection range and integrating the results. That is, since the speed value is proportional to the gradient angle θ, the resolution of the image data is set to the predetermined first resolution in the predetermined first speed value range, and the predetermined second value different from the first speed value range is set. By making the resolution of the image data higher than the first resolution in the velocity value range, the accuracy of the velocity value measurement in the second velocity value range becomes the accuracy of the velocity value measurement in the first velocity value range. Can be high. In other words, the measurement accuracy of the velocity value of the moving object can be changed by changing the resolution of the image data.

  In the above method, the decimation process is performed to change the resolution. However, the present invention is not limited to this, and the resolution may be changed by changing the sampling speed or by using various other methods.

  FIG. 6 is a modification of the image processing apparatus 3 of FIG. 1, and includes high-speed resolution images, medium-speed resolution images obtained by changing the resolution in each case obtained by changing the speed, and It is a figure which shows the image of the resolution for low speeds. FIG. 6A shows the images V11 to V13 of case 1 at the lowest speed, and at the lowest speed, a predetermined measurement accuracy can be obtained by performing speed measurement using the images V12 and V13. FIG. 6B shows the images V21 to V23 of the case 2 at the medium speed faster than that in FIG. 6A. Even if the images approach the medium speed, the images V22 and V23 are selected and the speed is measured. The problem of “too steep” does not occur, and the speed measurement can be performed by selecting one of the images V22 and V23 that are convenient for measurement. Further, FIG. 6C shows the images V31 to V33 of the case 3 at high speed, which is faster than that of FIG. 6B. If the inclination of the striped pattern exceeds a predetermined threshold value, for example, the images V31, V31, A predetermined measurement accuracy can be obtained by selecting V32 and measuring the speed. Furthermore, FIG. 6 (d) shows the images V41 to V43 of the case 4 at the highest speed that is faster than that of FIG. 6 (c). A predetermined measurement accuracy can be obtained by using. Therefore, the image processing device 3 generates a plurality of image signals having different resolutions, and has a predetermined speed measurement accuracy (or a speed measurement accuracy equal to or higher than a predetermined threshold) among the plurality of image signals. The obtained image signal may be selected, and a density gradient of an image including the stripe-shaped texture may be calculated based on the selected image signal and converted into a velocity value of the moving body.

  In the above embodiment, as shown in FIG. 3, the ground speed of the railway vehicle is calculated based on the image of the sleepers 22. For example, when the sleepers 22 are arranged at intervals of 60 cm, it is preferable to select the imaging line length of the line sensor camera 1 so that at least two sleepers 22 can be imaged. Here, the present invention is not limited to this, and the ground speed of the moving body may be calculated based on an image of an object under the floor of the moving body, such as a tipping machine, regardless of the sleepers 22. In other words, the moving body is not limited to a railway vehicle, and may be a moving body such as a vehicle. The moving body is striped based on an object that is fixedly placed on the ground, for example, under the floor or on the side surface. And a velocity value can be obtained based on the gradient angle θ.

  In the above embodiment, the illuminated image is captured by the line sensor camera 1. However, the present invention is not limited to this, and the image is captured by the line sensor camera 1 by illuminating with light having a predetermined wavelength. The captured image signal may be filtered using a band-pass filter that passes only the wavelength, and the image processing may be performed. Thereby, noise due to other light wavelengths can be removed, and an accurate ground speed can be calculated.

  In the above embodiment, the low-pass filter 12 is used. However, when the noise component included in the image digital data from the image capturing circuit 11 is relatively small, the low-pass filter 12 may not be inserted. Good.

  In the above embodiment, the line sensor camera 1 is used. However, the present invention is not limited to this, and may be a so-called linear array sensor in which a plurality of detection elements are arranged in a line. A partial scan (partial line to several lines) of a central scanning line of a simple area sensor may be partially acquired (partial scan), and may be realized at low cost.

  As described above, according to the embodiment of the present invention, the imaging line of the line sensor camera 1 is arranged substantially parallel to the moving direction of the moving body, and an image is acquired at a constant cycle. When there is relative movement, the image pattern becomes a slanting pattern in proportion to the speed. By converting this inclination into the speed, the speed value of the moving object is highly accurate and simple compared to the conventional technology. Can be measured with a simple configuration. In the present embodiment, (1) radio waves are not used and not affected by laws and regulations, and (2) the field of view can be widened by selecting a lens of the line sensor camera 1, for example, and the objective distance is the period of the sleepers 22. Measurement is possible even with a fluctuating bridge, solving the above-mentioned problems of the prior art.

  As described above in detail, therefore, according to the speed measuring apparatus and method according to the present invention, it is possible to accurately measure the ground speed as compared with the prior art, and the speed measuring apparatus and method are simple in configuration. Can be provided. For example, by measuring the ground speed of the vehicle, the position of the vehicle can be measured more accurately in the vehicle inspection system, and the inspection accuracy is greatly improved. Further, in the running state of the vehicle, it is possible to accurately detect and measure the idling / sliding state together with the conventional speedometer.

1 ... Line sensor camera,
2 ... Laser line projector,
3 ... Image processing device,
4 ... Speed indicator,
5. Speed output signal,
11 ... Image capture circuit,
11m ... buffer memory,
12 ... Low-pass filter (LPF),
13 ... Differential filter,
14: Gradient angle calculation circuit,
15 ... Speed value conversion circuit,
21a, 21b ... rails,
22 ... sleepers,
A1: Field sensor camera viewing area,
A2 ... Illumination area by laser,
V11 to V43 ... images.

Claims (12)

A speed measuring device that measures a speed value of the moving body when the moving body moves in a predetermined traveling direction at a predetermined speed,
An object having an imaging line in the main scanning direction in a direction substantially parallel to the traveling direction of the moving body, and an object fixedly placed on the ground facing the moving body at a predetermined time interval. Imaging means for repeatedly imaging in the sub-scanning direction not parallel to the scanning direction and outputting an image signal;
Based on the image signal output from the imaging means, a density gradient of an image including a stripe-shaped texture included in the image signal and based on the object is calculated when the moving body moves. A speed measuring device comprising: a speed value converting means for converting the gradient angle into a speed value of the moving body using a relationship proportional to the speed value and outputting the speed value.   The sub-scanning direction is set so as to be substantially orthogonal to the main scanning direction, and the speed value converting means corresponds to the first direction corresponding to the main scanning direction, the sub-scanning direction, and the The gradient angle is calculated based on the concentration gradient in the second direction with respect to the concentration gradient in the first direction, and is calculated in a second direction orthogonal to the first direction. Speed measuring device.   Filter means for filtering the image signal outputted from the image pickup means so as to remove a predetermined noise component different from the image component related to the striped texture based on the object, and then outputting to the velocity value converting means The speed measuring device according to claim 1, further comprising:   When calculating the density gradient of the image including the stripe-shaped texture based on the object, the speed value conversion unit calculates an average value of the density gradient related to the image signal, and based on the average value of the density gradient The velocity measuring device according to any one of claims 1 to 3, wherein the gradient angle is calculated.   The speed value conversion means changes the measurement accuracy of the speed value by changing the resolution of the image signal with respect to the image signal output from the imaging means. The speed measuring device according to any one of the above.   The speed value converting means generates a plurality of image signals having different resolutions, selects an image signal that can obtain a predetermined speed measurement accuracy from the plurality of image signals, and based on the selected image signal 6. The speed measuring apparatus according to claim 5, wherein a density gradient of an image including the striped texture is calculated and converted to a speed value of the moving body. A speed measuring method for measuring a speed value of the moving body when the moving body moves in a predetermined traveling direction at a predetermined speed,
The imaging means has an imaging line in the main scanning direction in a direction substantially parallel to the traveling direction of the moving body, and the imaging means is configured to detect an object fixedly placed on the ground facing the moving body. A step of repeatedly imaging in a sub-scanning direction not parallel to the main scanning direction at a predetermined time interval and outputting an image signal;
Based on the image signal output from the imaging means, the control device calculates the density gradient of the image including the striped texture included in the image signal and based on the object when the moving body moves, And a step of converting the gradient angle into a velocity value of the moving body using a relationship in which the gradient angle is proportional to the velocity value, and outputting the velocity value.   The sub-scanning direction is set so as to be substantially orthogonal to the main scanning direction, and the step of converting into the speed value and outputting is performed by the control device with a first direction corresponding to the main scanning direction. And a second direction corresponding to the sub-scanning direction and perpendicular to the first direction, and calculating a gradient angle based on the density gradient in the second direction with respect to the density gradient in the first direction. The speed measuring method according to claim 7, further comprising:   A step of filtering the image signal output from the imaging unit to remove a predetermined noise component different from the image component related to the striped texture based on the object, and then outputting the filtered noise component to the control device; The speed measurement method according to claim 7 or 8, wherein   The step of converting to the speed value and outputting the step calculates the average value of the density gradient related to the image signal when the control device calculates the density gradient of the image including the striped texture based on the object. The speed measurement method according to claim 7, further comprising calculating the gradient angle based on an average value of the concentration gradient.   The control device further includes a step of changing the measurement accuracy of the speed value by performing decimation filter processing on the image signal output from the imaging unit to change the resolution of the image signal. The speed measurement method according to any one of claims 7 to 10.   In the step of converting to the speed value and outputting, the control device generates a plurality of image signals having different resolutions and selects an image signal capable of obtaining a predetermined speed measurement accuracy from the plurality of image signals. The velocity measurement according to claim 11, further comprising: calculating a density gradient of an image including the stripe-shaped texture based on the selected image signal and converting the gradient into a velocity value of the moving body. Method.

Images