JPH05116626A - Obstacle detecting device for railway rolling stock - Google Patents

Abstract

PURPOSE:To detect an obstacle far away from the braking distance and measure the distance up to the obstacle so as to obtain high safety and prevent complication of a system. CONSTITUTION:An infrared camera 11 to photograph the front of a running railway rolling stock and a laser distance measuring sensor 19 to measure the distance up to an obstacle existing in the course are provided. The image photographed by the infrared camera 11 is converted to digital data by means of an A/D convertor 12 and memorized in an image memory 13. Further, a distance sensor in kilometers 15, to detect the present position of the running rolling stock and a route data memory 16 to memorize the route data are provided. Existence of an obstacle in the image stored in the image memory 13 is judged by means of a computing device 14 based on the route data stored in the memory 16 and the position detected signal from the distance sensor in kilometers 15, and the judged result and the distance up to the obstacle measured by means of the laser distance measuring sensor 19 are output to an output device 17 so as to inform a driver.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an obstacle detection device for a railway vehicle for detecting an obstacle existing on the route of a running railway vehicle.

[0002]

2. Description of the Related Art When operating trains, trains, and other railway vehicles,
In order to ensure the safety of the vehicle, it is possible to photograph the front of the traveling vehicle and determine the presence or absence of obstacles on the route from the photographed image. In this case, the problem is the distance La between obstacles. When the braking distance of the vehicle is L, if L <La, the collision can be avoided. However, if L> La, a collision will occur even if an obstacle can be detected. Therefore, the obstacle is meaningless unless it is detected beyond the front L (braking distance) of the vehicle. Furthermore, considering the margin ΔL for the driver to judge, (L + Δ
It is necessary to detect obstacles beyond L). The following two methods are conceivable as a method of determining the position of the (L + ΔL) front rail within the screen of the captured image.

(1) The distance between two rails is constant.
Using this feature, the number of pixels between rails in front of (L + ΔL) is calculated in advance from the angle of view of the camera.
Find the position of the rail in front of (L + ΔL) in the screen.
However, this method is effective only when the track is straight, and the accurate position cannot be calculated when the track is curved. Therefore, the following method can be considered. (2) The distance is calculated by triangulation using two cameras, and the position of the rail in front of (L + ΔL) is determined.

If the position of the front rail (L + ΔL) is known on the screen, an area (obstacle search area) for inspecting for the presence or absence of an obstacle is set around the position, and image processing is performed within the same area. The presence or absence of obstacles is determined by things.

[0005]

In the above methods (1) and (2), it is necessary to identify and recognize the position of the rail within the screen by performing image processing and the like. Moreover,
The position in front of (L + ΔL) is obtained by the methods (1) and (2). However, in (1), it is impossible to determine the position when the track is curved, and in (2), the distance between the cameras cannot be equal to or larger than the lateral width of the vehicle body. There is a problem in accuracy because the distance between the cameras cannot be taken sufficiently.
Further, since two cameras are used, the whole system becomes complicated, the time required for calculation is prolonged, and the cost is increased.

Further, the obstacle detected by performing the image processing in the obstacle search area is in front of the vehicle (L
It is an obstacle existing between + ΔL) and the point at infinity (theoretically). The driver's response changes according to the distance La to the obstacle. If La is slightly larger than (L + ΔL), it is necessary to brake suddenly. If La is considerably larger than (L + ΔL), you can ring the whistle and watch the situation.
As a method of knowing the distance to the obstacle, it is possible to calculate from the size of the obstacle on the screen and the angle of view of the camera, but the size of the obstacle varies widely and this method is impossible. is there.

The present invention has been made in view of the above points, and it is possible to accurately determine the presence or absence of an obstacle at a distance beyond the braking distance (L + ΔL), and to inform the driver by measuring the distance to the obstacle. An object is to provide an obstacle detection device for a railway vehicle.

[0008]

An obstacle detection device for a railway vehicle according to the present invention is an image input device installed so as to capture the front of a traveling railway vehicle, and an image captured by this image input device. An image memory for storing, a position detection means for detecting the current position of the traveling vehicle, and a vehicle for which the position detection means detects how the rail looks in the image captured by the image input device. A route data memory stored in advance as route data corresponding to the position, and obstacles in the image stored in the image memory based on the route data stored in the memory and the position detection signal from the position detecting means. Determination means for determining whether there is a vehicle, a vibration angle detection means for detecting the vibration of the vehicle, a distance measurement means for measuring the distance to an obstacle, and a screen obtained by the image input device. And the calculation means for calculating the direction of the obstacle from the vibration angle obtained by the vibration angle detection means, and the distance measuring direction of the distance measuring means based on the direction information obtained by this calculation means. A distance measuring direction control means for controlling and driving, and an informing means for informing a driver of presence or absence of an obstacle determined by the determining means and a distance to the obstacle measured by the distance measuring means. Is.

[0009]

[Operation] The route data in which the laying condition of the railway is converted is stored in advance in the route data memory, and the current position of the running vehicle (for example, the distance from the reference point) is detected by the position detecting means.
Is detected and the stored contents of the route data memory are read, it is possible to confirm the laying condition of the track in front of the vehicle. That is, when three-dimensional coordinates with the current position as the origin are set, the coordinates of the position p of the rail ahead of the arbitrary distance R can be uniquely determined. An image input device installed so as to photograph the front of the traveling vehicle, for example, an infrared camera, can be used to find where on the screen the point p is seen by mathematical coordinate conversion calculation. Therefore, it is possible to know where the rail ahead of the arbitrary distance R can be seen in the screen. Therefore, by setting an obstacle search area of a finite size around the position of the rail in the screen when R = (L + ΔL), the presence or absence of obstacles can be determined by performing image processing in the area. judge. Furthermore, the direction of the obstacle is calculated from the position of the obstacle on the screen and the output of the vibration angle detection sensor, and the distance to the obstacle is measured by controlling and driving the distance measuring direction of the distance measuring means. Then, the driver is informed of the presence or absence of an obstacle based on the above determination result and the measured distance to the obstacle.

By using the route data as described above, even if the front line is straight or curved, (L
+ ΔL) It is possible to accurately calculate the position of the front rail in the screen. Note that the distance (L + ΔL) is arbitrary and can be set to a braking distance according to the speed of the vehicle. Further, since it is not necessary to identify and recognize the position of the rail on the screen by performing image processing or the like, the calculation time can be greatly reduced. Further, since the distance to the obstacle is known by performing the distance measurement, the driver can take appropriate measures according to the distance to the obstacle.

[0011]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings. FIG. 1 shows the configuration of an obstacle detection device for a railway vehicle according to an embodiment of the present invention, which is equipped with an infrared camera 11 as an image pickup means. This infrared camera 11
Is attached to the head position of the vehicle 10 as shown in FIG. 2 and photographs the front of the vehicle. The video signal taken by this camera is converted into an A / D (analog / digital) converter 12
Is converted into digital data. And this A / D
The output signal from the converter 12 is written in the image memory 13 based on a command from the arithmetic unit 14.

The arithmetic unit 14 has a kilometer (distance from a reference point) signal from a kilometer sensor 15 for detecting the current position of the vehicle, route data stored in advance in the route data memory 16, and a gyro sensor. The vibration angle from 18 and the distance from the laser distance measuring sensor 19 to the obstacle are supplied. The above-mentioned kilometer sensor 15 is attached to the wheel of the vehicle as shown in FIG. 2, and outputs distance data corresponding to the number of revolutions thereof, for example.

Further, the route data memory 16 stores route data showing a line shape corresponding to a kilometer as shown in FIGS. 3 (a) and 3 (b). In the example shown in FIG. 3, there is a right (R) curve with a radius of 50 m and a crossing angle of 90 ° from a distance of 0 to 100 km, followed by a straight line of 400 m, and a radius of 100 from a distance of 500 to 700.
It is a left (L) curve with m and an intersection angle of 60 °, then 100
A straight line of m continues. In FIG. 3 above, BTC (Bi
ginning of Transition Curve) is the starting point of the relaxation curve, BC
C (Biginning of Circular Curve) is the starting point of the circular curve, E
CC (END of Circular Curve) is the end point of the circular curve, ETC
(END of Transition Curve) indicates the end point of the relaxation curve. Although the route data shown in FIG. 3 is for the horizontal direction, the route data for the vertical direction is also recorded in the route data memory 16 in the same way, for example, the slope of the line corresponding to about kilometer.

The gyro sensor 18 and the laser distance measuring sensor 19 are installed near the infrared camera 11 as shown in FIG. 2, and the gyro sensor 18 outputs the yawing angle and the pitching angle of the vehicle at the camera mounting position. ..

The arithmetic unit 14 in FIG. 1 is roughly divided into four parts, that is, a binarizing means 14a, an obstacle search area setting means 14b, and an obstacle presence / absence determining means 14c.
And a laser beam direction setting means 14d.
The binarizing unit 14 a obtains image data from the image memory 13, binarizes the image data by various image processing algorithms, and records the binary image in the image memory 13. The obstacle search area setting unit 14b uses the kilometer signal from the kilometer sensor 15, the route data from the route data memory 16, and the binary image from the image memory 13 to perform an inspection for an obstacle (obstacle). Set the search area) within the screen. The obstacle presence / absence determining unit 14c determines the presence / absence of an obstacle by performing various image processes in the area (obstacle search area). The laser beam direction setting means 14d is
The direction of the obstacle is calculated from the position of the obstacle on the screen and the output of the gyro sensor 18, and the calculated direction is output to the laser distance measuring sensor 19.

The laser distance measuring sensor 19 comprises a laser beam direction setting means 19a and a laser distance measuring means 19b. The laser beam direction setting means 19a brakes the direction of the laser beam based on the direction of the obstacle outputted by the laser beam direction setting means 14d. The laser distance measuring means 19b is the laser beam direction control driving means 1 described above.
Laser distance measurement is performed in the direction determined by 9a, and the distance to the obstacle is output to the arithmetic unit 14. Then, the presence / absence of the obstacle and the distance to the obstacle obtained by the arithmetic unit 14 are sent to the output device 17 to notify the driver.

FIG. 4 shows a flow of operations in the arithmetic unit 14. In step 101, the distance in kilometers corresponding to the current position of the vehicle is set. Next step 102
Then, the output of the infrared camera 11 is digitized by the A / D converter 12 and stored in the image memory 13. Step 10
In 3, the image data stored in the image memory 13 is binarized by various known image processing algorithms to extract the rail portion as a binary image, and the binary image is recorded in the image memory 13.

In step 104, the kilometer corresponding to the present position of the vehicle is read from the kilometer sensor 15. In step 105, the route data corresponding to this kilometer is read from the route data memory 16. In step 106, an obstacle search area is set in the screen according to a method described later using the binary image and the route data. Step 10
At 7, the presence / absence of an obstacle is determined by performing various types of image processing described later in the obstacle search area. Then, in the next step 108, it is judged whether or not the obstacle is detected in the above step 107, and if the obstacle is detected, the process proceeds to step 109, the direction of the laser beam is set by the method described later, and the step 110 is performed. To measure the distance to the obstacle by laser measurement. In step 111, output display is executed to inform the driver of the existence of the obstacle and the distance to the obstacle. In step 112, it is determined whether or not the operation has ended. If the operation has ended, this process ends, and if the operation has not ended, the process returns to step 102 to repeat the above operation.

Next, a method for setting the obstacle search area in step 106 will be described. FIG. 5 shows an example of the processing flow of the above method. Step 201 corresponds to steps 104 and 105. By this operation, the laying condition of the track ahead in the traveling direction can be known. Therefore, in step 202, 3 as shown in FIG.
The dimensional coordinate system 20 is set, and the coordinates (x, y, z) of the center point p21 of the left and right rails 1500 m ahead are calculated, for example. Here, 1500 m is an example of the sum (L + ΔL) of the braking distance L and the margin ΔL for the driver to judge, and can be set to a value corresponding to the speed of the vehicle. In step 203, when the infrared camera 11 photographs the point p21, the position (m, n) 24 where the point p is visible in the image 23 is calculated as shown in FIG.

FIG. 7 shows a digital image stored in the image memory 13 after being photographed by the infrared camera 11.
The coordinates (m, n) 24 are given by a mathematical coordinate conversion calculation using the camera mounting position, the angle, and the angle of view of the camera that can be uniquely determined by the three-dimensional coordinate system 20 as parameters. In step 204, the obstacle search area 25 is set around the position (m, n) 24 where the point p is visible on the screen. The size of the obstacle search area is arbitrary, but if it is too large, it is meaningless, and if it is too small, the area to be inspected is not included, which is a problem.

An example of a method for setting the size of the obstacle search area will be shown below. Considering the construction limit of the vehicle, it is considered that there will be no problem if there is no obstacle in the area of 3.8 m × 3.8 m on the track. So, 3.8m ahead of 1500m
The size of the m × 3.8 m area on the screen is calculated. This is the number of pixels in the horizontal and vertical directions,
It depends on the horizontal and vertical angles of view of the camera. If the number of pixels is 640 × 486 and the angle of view is 3.5 ° × 2.7 °, the area of 3.8 m × 3.8 m ahead of 1500 m is 27 × 27 (pixels) on the screen. .. Therefore, as shown in FIG. 7, 27 × 2 with the coordinates (m, n) 24 as the center.
An obstacle search area 25 having a size of 7 (pixels) is set. In the present embodiment, the coordinates (m,
Although n) 24 is calculated every time, it is also possible to store the coordinates (m, n) 24 obtained in the above step 203 in association with the distance of about km and use them as route data. That is, the route data may be stored in any form as long as the coordinates (m, n) 24 corresponding to the distance in kilometers are known.

The method shown in FIG. 5 is possible when the influence of vehicle vibration can be ignored. Actually, the position and angle of the camera attached to the vehicle change due to the vibration, which causes a positional shift in the image. That is, an image different from the image when the vehicle is stationary is captured. Step 2 above
The position (m, n) where the point p obtained in 03 is visible in the image 23
24 is the position in the screen when the vehicle is stationary. When the camera mounting position and angle change greatly due to vibration, the position (m, n) 24 is displaced within the screen, which is a problem. Therefore, FIG. 8 shows a processing flow of a method for eliminating the influence of vibration.

Step 201 in FIG. 8 has the same contents as step 201 in FIG. In step 301, the rail image 30 shown in FIG. 9 in which the left and right rails are drawn using the route data stored in the route data memory 16 is generated on the image memory 13. The rail image 30 is, for example, a binary image in which a rail is drawn with a bit “1” on the background of a bit “0”.

An example of the method of generating the rail image 30 is shown below. In step 202 of FIG. 5, not only the point 1500 m ahead, but the point 200-1500 m ahead is 100
It is calculated every m and the coordinates in the screen are obtained in step 203. Since these are the coordinates at the center of the rail, the coordinates of the left and right rails are obtained by adding the number of inter-rail pixels according to the distance. Connect the coordinates with a straight line 200-1
Generate left and right rail images 500m ahead. Needless to say, it is not always necessary to calculate 200 to 1500 m every 100 m, and it is possible to change it flexibly. Further, as described above, the coordinates in the screen of the left and right rails for every 100 m ahead of 200 to 1500 m calculated in the above step 301 are stored in association with the kilometer,
It can also be used as route data. In addition, there is no problem in storing any route data as long as the rail image 30 corresponding to about a kilometer can be generated.

In step 302, the shift amounts ρm and ρn are obtained by correlating the binary image 31 extracted from the rail portion shown in FIG. 10 generated in step 103 with the rail image 30. The binary image 31 is shown in FIG.
As shown in, for example, the rail 34a part has a bit "1".
It is a binary image as represented by. The amount of deviation ρm3
As shown in FIG. 11, 5 m is the number of pixels displaced in the horizontal direction (m-axis direction) between the actual rail 34 and the drawn rail 33, and the deviation amount ρn35n is displaced in the vertical direction (n-axis direction). The number of pixels.

As shown in FIG. 9, a window 32 is set in the rail image 30 so as to surround the drawn rail 33. The correlation value in the window 32 is obtained by shifting one image vertically and horizontally from the state where the two images are completely overlapped. Specifically, a logical product image of the rail image 30 and the binary image 31 is obtained, and the sum total of bits “1” in the window 32 is set as the correlation value. The shift amount in the horizontal direction when the correlation value is maximum is regarded as the shift amount ρm, and the shift amount in the vertical direction is regarded as the shift amount ρn. The method of setting the window for obtaining the correlation does not necessarily have to be the above method, and the deviation amounts ρm, ρn
Any window may be used as long as it can be accurately obtained.

Next, at step 303, the above deviation amount ρm35.
The obstacle search area 25 is set by correcting m and ρn35n. As shown in FIG. 12, a point p21 at the center of the rail ahead of 1500 m can be seen in the image 23 at a position (m, n) 2
4 is corrected by the deviation amounts ρm35m and ρn35n (m
+ Ρm, n + ρn) 36 is set as the center, and the obstacle search area 25 having a size of 27 × 27 pixels is set. That is,
An accurate obstacle search area can be set by obtaining the amount of deviation on the screen and correcting only the amount of deviation.

FIG. 13 shows an example of the obstacle presence / absence determination processing in step 107 of FIG. This apparatus uses an infrared camera 11 as an image pickup means. When an image of an obstacle such as an automobile is taken by the infrared camera 11, a characteristic is that a high temperature portion such as an engine appears bright. Using this feature, if there is a high-luminance portion in the obstacle search area 25, it is determined that there is an obstacle. In step 401, the image data in the obstacle search area 25 is extracted from the digital image 23 stored in the image memory 13. In step 402, the image data is binarized with an appropriate threshold value,
A high-brightness part (a part having a higher brightness than the background) is extracted. In this case, for example, the high-luminance portion is bit "1" and the other portions are "0".

In step 403, the above step 402 is performed.
It is determined whether the high-intensity part is extracted in the binary image generated by. As an example of the determination method, if there is a bit "1" portion, it is determined that there is an obstacle, and the process proceeds to step 404Y. If there is no obstacle, step 404
Proceed to N. This method determines the presence or absence of an obstacle based on the presence or absence of a high-intensity part in the obstacle search area 25, and the binarization method in step 402 is any method as long as the high-intensity part can be extracted. Can be a method,
Further, there is no reason that the method for determining the presence / absence of the high-luminance portion in step 403 is not necessarily the above method.

The obstacle presence / absence determination result 404Y or 404N obtained by the above processing is output to step 108 in FIG. 4, and if there is an obstacle, the process proceeds to step 109 to measure the distance to the obstacle. Here, a method of performing distance measurement using a laser beam will be described. There is no reason that the laser beam must be used as the distance measuring means, and any means can be used as long as the distance to the obstacle can be measured.

An example of the laser beam direction setting process of step 109 in FIG. 4 is shown below. FIG. 14 is an image in which an obstacle exists. The screen size is 2M × 2N (pixels), and the camera angle of view is 2θh (°) in the horizontal direction and 2θv (°) in the vertical direction. The coordinates 41 of the obstacle on the screen are (m0, n0), and the screen center 40 is (M, N). At this time, the direction angle of the obstacle viewed from the optical axis of the camera (the horizontal direction is Θh and the vertical direction is Θv) is given by the following equation. However, in the horizontal direction, the right side from the center is positive,
The vertical direction is positive above the center. Θh = (m0−M) · θh / M θv = − (n0−N) · θv / N

FIG. 15 is a front view of the vehicle. The infrared camera 11 and the laser distance measuring sensor 19 are installed coaxially and on the optical axis thereof, and the optical axes thereof are in the same direction. Therefore, the direction angles Θh ′ and Θv ′ of the obstacle viewed from the laser distance measuring sensor 19 are
And the difference Δh42 in the mounting position height of the infrared camera 11 is corrected. This calculation is given by a mathematical coordinate transformation. On the other hand, the vehicle is moving while capturing the image from the infrared camera 11 and performing the above processing. Therefore, the obstruction direction angle at the time of performing the laser distance measurement is different from the obstruction direction angle at the time of capturing the image. In order to avoid the above situation, the gyro sensor 18 is attached near the infrared camera 11 as shown in FIG. The gyro sensor 18 outputs the yawing angle α and the pitching angle β of the vehicle at the camera mounting position. The yawing angle and the pitching angle at the time of capturing an image are α0 and β0, respectively. The yaw angle immediately before laser ranging is α1,
The pitching angle is β1. Obstacle direction angles Θh ′ and Θv ′ viewed from the laser distance measuring sensor 19 are finally given by the following equations. Θh '= Θh' + (α1-α0) Θv '= Θv' + (β1-β0)

Next, an example of the laser distance measurement in step 110 in FIG. 4 will be shown below. The above Θh ′ and Θv ′ are shown in FIG.
The laser beam direction setting means 14d is transferred to the laser beam direction control driving means 19a to perform laser distance measurement.

FIG. 16 shows a laser beam direction control driving means 1
An example of the principle of 9a is shown. This is called a wedge prism rotation method. Wedge-shaped prism 4
The laser beam is swung in an arbitrary direction by rotating 4, 45. FIG. 17 is a view of the prisms 44 and 45 as seen from the lateral direction. When the angle 48 of refraction of the laser beam at the wedge is θ as shown in the figure, the range 46 in which the beam can be scanned is a conical range of 4θ as shown in FIG. One of the two prisms 44 and 45 can change the position in the circumferential direction, and the other prism can change the position in the radial direction. That is, the prism 4
By controlling the rotation angles of 4, 45, the laser beam can be swung in a predetermined direction.

FIG. 18 shows a laser beam direction control driving means 1
An example of 9a is shown. Direction angle of laser beam Θh ', Θ
v ′ is passed from the laser beam direction setting means 14d to the servo driver 50 in the laser beam direction control driving means 19a. The servo driver 50 rotates the motor 51a by an amount corresponding to the angle Θh 'to drive the wedge prism 4
Rotate 4. An encoder 52a is attached to the motor 51a, and its rotation angle is fed back to the servo driver 50 to be controlled. Similarly, by rotating the motor 51b by an amount corresponding to Θv ', the wedge prism 4
Rotate 5. An encoder 52b is attached to the motor 51b, and its rotation angle is fed back to the servo driver 50 to be controlled. On the other hand, the laser beam 47 output from the laser distance measuring means 19b goes toward the obstacle by passing through the wedge prisms 44 and 45. Thereby, the distance to the obstacle is measured by, for example, the pulse time difference counting method, and the value is passed to the arithmetic unit 14. As a result, in step 111 of FIG.
Thus, the presence of the obstacle and the distance to the obstacle are notified to the driver via the output device 17.

[0036]

As described in detail above, according to the obstacle detecting apparatus for a railway vehicle according to the present invention, by using the route data, the rail ahead of an arbitrary distance is displayed on the screen regardless of the track laying condition in front of the vehicle. You can calculate where you look inside. Therefore, the area to be inspected for the presence of obstacles can be accurately determined in a short time. Therefore, it is possible to accurately determine the presence / absence of an obstacle in a region beyond the braking distance and notify the driver, and high safety is guaranteed. Moreover, since the distance to the obstacle is measured and notified to the driver, the driver can deal with the obstacle according to the distance. Further, it is not necessary to install two cameras, and the problems of system complexity, extension of required calculation time, high cost, etc. do not occur.

[Brief description of drawings]

FIG. 1 is a configuration diagram of an obstacle detection device for a railway vehicle according to an embodiment of the present invention.

FIG. 2 is an explanatory diagram of a vehicle equipped with the obstacle detection device.

FIG. 3 is a diagram showing an example of route data.

FIG. 4 is a flowchart illustrating a processing flow of the obstacle detection device.

5 is a flowchart for explaining the operation of the obstacle search area setting process of step 106 in FIG.

6 is an explanatory diagram of the operation in FIG.

7 is a diagram illustrating a method of setting an obstacle search area in FIG.

FIG. 8 is a flowchart illustrating an example of processing for eliminating the influence of vibration in the obstacle search area setting.

FIG. 9 is a diagram illustrating a method of setting a window on a rail image when obtaining a correlation value.

FIG. 10 is a diagram showing a binary image in which a rail portion is extracted.

FIG. 11 is a diagram illustrating deviation amounts ρm and ρn.

12 is a diagram illustrating a method of setting an obstacle search area in FIG.

FIG. 13 is a flowchart illustrating an obstacle presence / absence determination process in step 107 in FIG.

FIG. 14 is a diagram showing an example of an image when an obstacle is present.

FIG. 15 is a diagram showing an arrangement example of devices when the vehicle is viewed from the front.

16 is a diagram for explaining an example of the principle of the laser beam direction control driving means in FIG.

FIG. 17 is a view of the wedge prism in FIG. 16 seen from the lateral direction.

FIG. 18 is a diagram showing a configuration example of laser beam direction control driving means in FIG. 1.

[Explanation of symbols]

10 ... Vehicle, 11 ... Infrared camera, 12 ... A / D converter, 13 ... Image memory, 14 ... Arithmetic device, 14a ... Binarization means, 14b ... Obstacle search area setting means, 14c ...
Obstacle presence / absence determining means, 14d ... Laser beam direction setting means, 15 ... Kilometer sensor, 16 ... Route data memory, 1
7 ... Output device, 18 ... Gyro sensor, 19 ... Laser distance measuring sensor, 19a ... Laser beam direction control driving means, 1
9b ... Laser distance measuring means, 20 ... Three-dimensional coordinates, 21 ... 15
Rail central point p, 22 ... Rail, 23 ... Digital image stored in image memory, 24 ... Position where point p is visible in digital image, 25 ... Obstacle search area, 30 ... Rail image, 31 ... Binary image, 32 ... Window for obtaining correlation value, 33 ... Rail drawn by using route data, 34 ... Actual rail, 34a ... Actual binarized rail, 35m ... Deviation amount ρm, 35n ... Deviation amount ρn, 36 ... Points corrected by the deviation amounts ρm, ρn, 40 ...
Center of screen (m, n), 41 ... Coordinates of obstacle, 42 ... Height difference Δ between mounting position of laser distance sensor and infrared camera
h, 43 ... Laser light source, 44, 45 ... Wedge prism, 46 ... Laser beam scan range, 47 ... Laser beam, 48 ... Wedge refraction angle θ, 5
0 ... Servo driver, 51a, 51b ... Motor, 52
a, 52b ... Encoder.

[Procedure amendment]

[Submission date] December 2, 1991

[Procedure Amendment 1]

[Document name to be amended] Statement

[Name of item to be corrected] 0022

[Correction method] Change

[Correction content]

The method shown in FIG. 5 is possible when the influence of vehicle vibration can be ignored. Actually, the position and angle of the camera attached to the vehicle change due to the vibration, which causes a positional shift in the image. That is, an image different from the image when the vehicle is stationary is captured. Step 2 above
The position (m, n) where the point p obtained in 03 is visible in the image 23
24 is the position in the screen when the vehicle is stationary. When the camera mounting position and angle change greatly due to vibration, the position (m, n) 24 is displaced within the screen, which is a problem. Therefore, FIG. 8 shows a processing flow of a method for eliminating the influence of vibration.

[Procedure Amendment 2]

[Document name to be corrected] Drawing

[Name of item to be corrected] Fig. 4

[Correction method] Change

[Correction content]

[Figure 4]

Front page continued (72) Inventor Jun Nakamoto 4-6-22 Kannon Shinmachi, Nishi-ku, Hiroshima City, Hiroshima Prefecture Mitsubishi Heavy Industries, Ltd. Hiroshima Research Institute (72) Inventor Hiroshi Yamashita 4-6 Kannon-shinmachi, Nishi-ku, Hiroshima City, Hiroshima Prefecture 22 Mitsubishi Heavy Industries, Ltd. Hiroshima Research Institute (72) Inventor Kuniaki Wakusawa 4-6-22 Kannon Shinmachi, Nishi-ku, Hiroshima-shi, Hiroshima Prefecture Mitsubishi Heavy Industries Ltd. Hiroshima Research Institute (72) Inventor Tamaya Shirahara Itozaki, Mihara City, Hiroshima Prefecture 5007, Machi Mitsubishi Heavy Industries Ltd. Mihara Works (72) Inventor Yasuhisa Iida 1-1-1, Niihama, Arai-cho, Takasago, Hyogo Prefecture Mitsubishi Heavy Industries Ltd. Takasago Research Laboratory

Claims (1)

[Claims] 1. An image input device installed so as to capture an image of the front of a traveling railway vehicle, an image memory for storing images captured by the image input device, and a current position of the traveling vehicle is detected. A position detecting means, and a route data memory that stores in advance how the rail looks in the image captured by the image input device as route data corresponding to the vehicle position detected by the position detecting means,
Based on the route data stored in this memory and the position detection signal from the position detecting means, a judging means for judging whether or not there is an obstacle in the image stored in the image memory, and a vibration of the vehicle are detected. Vibration angle detection means, distance measurement means for measuring the distance to the obstacle, position of the obstacle in the screen obtained by the image input device, and the vibration angle obtained by the vibration angle detection means from the obstacle. Calculating means for calculating the direction, distance measuring direction control means for controlling and driving the distance measuring direction of the distance measuring means based on the direction information obtained by the calculating means, presence or absence of obstacles judged by the judging means, and An obstacle detection device for a railway vehicle, comprising: a notification means for informing a driver of the distance to the obstacle measured by the distance measuring means.