JP2008298501A - Position detector - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To precisely detect a position of a mobile body with simple constitution. <P>SOLUTION: The position detector includes: an image sensor SE1 installed so as to be capable of picking up an image of a train TR together with an image of a background of the train TR; and a position detection part 11 detecting the position of the train TR by comparing a base image FG1 that is an image not containing the image of the train TR and an image FG2 for detection that is the image containing the image of the train TR picked up by the image sensor SE1. The position detection part 11 obtains the actual position PR of a reference part of the train TR by using a position PV on the image of the reference part of the train TR included in the image FG2 for detection and a distance Hd from the background to the reference part, and a distance Hc from the background to the image sensor SE1. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to an apparatus for detecting a position when a moving body such as a train, an automobile, an airplane, a truck, or an article on a conveyor is running or stopped.

  Conventionally, various devices for detecting a stop position of a train have been proposed.

  For example, in Patent Document 1, a first vehicle detection sensor that detects a head portion of a vehicle that is provided at a position shorter than the train stop position and shorter than the vehicle length of one vehicle is provided near the train stop position. Position detecting device comprising five second vehicle detection sensors for detecting the leading portion of the train and a stop position detector for detecting a train stop position from the first and second vehicle sensor detection signal generation timings Has been proposed.

  According to Patent Document 1, an average speed is detected from the time from when it passes through the first vehicle detection sensor to when it passes through the second vehicle detection sensor. The train stop position can be obtained from the average speed and the detection state of the second vehicle detection sensor.

  By detecting the stop position of the train, it is determined whether or not the train door and the door of the safety fence provided on the platform coincide with each other. If they coincide, the door of the safety fence is opened.

In addition, the time from when the head position of the train is detected by a distance sensor, an ultrasonic head is arranged at a predetermined location on the platform, and ultrasonic waves are emitted to the side of the train stopped on the platform and the reflected wave is received. It has also been proposed to detect the position of the passenger door based on them (Patent Document 2).
JP-A-6-278599 JP 2006-264475 A

  However, according to the position detection device of Patent Document 1, each of the first vehicle detection sensor and the second vehicle detection sensor detects the passage at the head of the train, and accurately determines the train position at that time. It is not something to detect.

  Therefore, in order to detect the stop position of the train with high accuracy, it is necessary to densely arrange a large number of vehicle detection sensors in a range where the train may stop. However, providing such a large number of vehicle detection sensors is disadvantageous in terms of cost and maintenance. Moreover, even if a large number of vehicle detection sensors are provided, it is difficult to detect the position with an accuracy greater than the installation interval, and there is a limit to the detection accuracy.

  Moreover, when the stop position of a train is detected by a distance sensor as in Patent Document 2, the detection position varies depending on the shape of the train and the like, so that there is a problem that the detection accuracy cannot be increased so much.

  The present invention has been made in view of the above-described problems, and an object of the present invention is to detect the position of a moving body with a simple configuration with high accuracy.

  According to one aspect of the position detection device of the present invention, an image sensor installed so that an image of a moving body can be captured together with an image of a background of the moving body, and the moving body captured by the image sensor are captured. A position detection unit configured to detect the position of the moving body by comparing a base image that is not an image and a detection image that is an image of the moving body.

  The position detector, for example, a position PV on the image of the reference part of the moving object shown in the detection image, a distance Hd from the background to the reference part, and a distance Hc from the background to the image sensor. Are used to determine the actual position PR of the reference portion of the moving body.

  Note that obtaining the actual position PR of the reference portion is substantially equivalent to obtaining the actual position PR, and it is not always necessary to output data indicating the actual position PR. For example, based on the actual position PR of the obtained reference part, further determine the position of the entire moving body, or the position of a specific part of the moving body, and output data indicating the finally obtained position, The data may be displayed or controlled based on the data.

  In addition, the reference unit in the present invention is a part serving as a reference for detecting the position of the moving body, and may be a part constituting a part of the moving body or a member attached to the moving body.

  According to the present invention, the position of the moving body can be detected with high accuracy with a simple configuration.

[First Embodiment]
FIG. 1 is a block diagram showing a configuration of a position detection apparatus 1 according to a first embodiment of the present invention, FIG. 2 is a diagram for explaining a method for detecting a stop position of a train TR, and FIG. 3 is an installation of an image sensor SE1. FIG. 4 is a diagram illustrating an example of the base image FG1 and the detection image FG2, FIG. 5 is a diagram illustrating a process of detecting a stop position using a detection window, and FIG. 6 is another diagram of the stop position of the train TR. It is a figure for demonstrating a detection method.

  1 and 2, the position detection device 1 includes an image sensor SE1, a moving body sensor SF1, a distance sensor SK1, a position detection unit 11, a stop determination unit 12, a stop position determination unit 13, and the like.

  The image sensor SE1 is installed so that an image of the train TR that is a moving body can be taken together with an image of the background of the train TR. The image sensor SE1 has an imaging range SA, and can image the ground GR as a background over the imaging range SA. For example, the image sensor SE1 is installed at a height of about 5 meters above the ground. In that case, the imaging range SA is, for example, about 6 to 7 meters. The image sensor SE1 is, for example, a video camera that outputs a moving image. For example, the image sensor SE1 continuously outputs images of 30 frames per second. One frame image selected from the moving image is the base image FG1 or the detection image FG2. A plurality of frames may be used to determine the detection image FG2.

  The image sensor SE1 includes an automatic exposure device, and is controlled so that the exposure amount is optimized according to time and weather. However, when the entry of the train TR is detected by the moving body sensor SF1, the automatic exposure device fixes the exposure amount at that time so that the base image FG1 and the detection image FG2 are captured with the same exposure amount. It has become. As a result, it is possible to more accurately detect the tip portion TB corresponding to the reference portion of the present invention without causing an unnecessary image change between the base image FG1 and the detection image FG2.

  The moving body sensor SF1 is installed to detect the approach of the train TR on the near side in the traveling direction of the train TR (arrow M1 direction) from the position where the image sensor SE1 is installed. The moving body sensor SF1 is arranged so as to detect the entry of the train TR before the train TR enters the imaging range SA of the image sensor SE1.

  The moving body sensor SF1 detects the presence or absence of an object in front of the sensor, such as an ultrasonic sensor or an infrared sensor. When the train TR enters in front of the moving body sensor SF1, the moving body sensor SF1 detects it and outputs a detection signal S1. The detection signal S1 is input to the image sensor SE1 as a base image imaging instruction signal, and based on the signal S1, the image sensor SE1 is controlled to capture the base image FG1. As a result, a base image FG1 immediately before the detection image FG2 is imaged is obtained, and the moving object can be detected accurately.

  The distance sensor SK1 measures the height of the front end portion TB of the train TR. As the distance sensor SK1, for example, a plurality of infrared sensors arranged in the height direction are provided on both sides of the track, and the distance Hd is detected by the infrared sensors being shielded by the front end portion TB of the train TR. It is possible to configure. The measured result is output as the distance Hd.

  Since the image sensor SE1 outputs a moving image as described above, “the image sensor SE1 captures the base image FG1” means that the signal S1 is input from the moving image output by the image sensor SE1. One frame image is selected at the same timing. When the image sensor SE1 can output a still image, one still image is output as the base image FG1 at the timing when the signal S1 is input.

  The base image FG1 is an image captured by the image sensor SE1 when the train TR has not entered the imaging range SA of the image sensor SE1. That is, as shown in FIG. 4A, the ground image GR and the track (track) SR are shown in the base image FG1, but the train TR as shown in FIG. 4B is not shown. On the other hand, as shown in FIG. 4B, the detection image FG2 shows the front end portion (leading portion) of the entering train TR. The front end portion of the train TR is an example of the reference portion of the present invention.

  The position detection unit 11 detects the position of the train TR by comparing the base image FG1 and the detection image FG2. For example, the position of the train TR is detected by the difference in luminance or color between the base image FG1 and the detection image FG2. At that time, a rectangular region that is long along the traveling direction of the train TR is extracted as a detection window WD from the base image FG1 and the detection image FG2, and based on the extracted image of the detection window WD, the train TR The actual position PR of the tip is calculated.

  Specifically, the position detection unit 11 includes a position PV (see FIG. 2) on the image of the front end portion TB of the train TR shown in the detection image FG2, and a distance Hd from the ground GR to the front end portion as a background. The actual position PR of the front end portion TB of the train TR is calculated using the distance Hc from the ground GR to the image sensor SE1.

  Since the position PS of the image sensor SE1 is the center of the detection image FG2, the distance from the position PS to the position PV on the image of the tip portion TB is set to LG with reference to the center position PS, and the position PS to the tip. Assuming that the distance to the actual position PR of the part TB is L, the distance LG is expressed by the following equation (1) from the similarity of triangles.

LG = [Hc / (Hc−Hd)] · L (1)
Therefore, by obtaining the position PV on the image and then obtaining the distance LG from this, the distance L can be obtained by the following equation (2).

L = [(Hc−Hd) / Hc] · LG (2)
The distance Hd from the ground surface GR to the front end TB is obtained by actual measurement by the distance sensor SK1, but instead of this, the data of the distance Hd for each train TR is stored in the memory together with the train diagram. These data may be read out in the calculation.

  In addition, an IC tag that records the vehicle number of the train TR, dimensions of each part, other specifications, operation schedule, etc. is attached to the train TR, and the data recorded on the IC tag is read to obtain the data of the distance Hd. Also good.

  Further, instead of the calculation using the formula, necessary data is stored in advance in the table, and the actual position PR is read from the table by inputting the position PV on the image of the tip portion TB. Also good.

  In order to detect the position PV on the image of the tip portion TB, an image of the train TR is obtained by subtracting the base image FG1 from the detection image FG2, and the tip portion TB is detected based on the image.

  As shown in FIG. 5, a detection window WD is extracted from the base image FG1 and the detection image FG2 [FIGS. 5A and 5B], and a difference image FG3 thereof is obtained [FIG. 5C]. An edge image FG4 is generated for the difference image FG3 [FIG. 5 (D)]. Since the tip end portion TB is clearly shown in the edge image FG4, the position becomes the position PV.

  As described above, the position detection unit 11 detects the actual position PR of the tip end part TB based on the detection image FG2. The position detection unit 11 outputs the actual position PR of the detected tip part TB. As the output, the position (stop position) PRB when the train TR is stopped and the occasional position while the train TR is running are output. PRA is output.

  The stop determination unit 12 determines whether the train TR has stopped based on the detection image FG2. When it is determined that the operation has stopped, a stop determination signal S2 is output. Based on the stop determination signal S2, the position detection unit 11 detects a stop position PRB that is the position of the front end of the train TR when the stop determination unit 12 determines that the stop has occurred.

  The stop determination unit 12 compares the image output from the image sensor SE1 with the immediately preceding frame for each frame. More specifically, for example, the position information (position PV or PR) for each frame output from the image sensor SE1 is compared with the position information of the immediately preceding frame. When the amount of change in the position PV or PR of the front end portion TB of the train TR becomes equal to or less than a predetermined value, it is determined that the train TR has stopped.

  That is, for example, the content of the immediately preceding frame is subtracted from the content of the frame output from the image sensor SE1. Since the difference pixel is the amount of change, the movement amount (number of movement pixels) of the difference pixel is obtained, and if the movement amount is within a predetermined range for a plurality of times, it is determined that the pixel is stopped. In this case, the predetermined range may be a range in which displacement due to shaking or the like when the train TR is stopped is not erroneously detected.

  The stop position determination unit 13 determines whether or not the train TR has stopped at a normal position. That is, the stop position determination unit 13 determines that the stop position PRB is stopped at the normal position when the stop position PRB detected by the position detection unit 11 is within the allowable range HMP for the target stop position MP of the train TR. The target stop position MP and the allowable range HMP may be stored in the target stop position storage unit 14 as coordinate data in the x direction in the imaging range of the image sensor SE1, for example. The stop position determination unit 13 outputs a signal S3 indicating whether or not the stop position PRB is a normal position.

  In FIG. 2, the position PR of the front end TB of the train TR is close to the target stop position MP and is within the allowable range HMP. In this case, the signal S3 indicating the normal position is output.

  Note that the target stop position MP and the allowable range HMP may be set as positions on the detection image FG2. In this case, the position PV of the front end portion TB of the train TR in the detection image FG2 is checked.

  Incidentally, in FIG. 2, the image sensor SE <b> 1 is installed on the rear side in the traveling direction (front side in the traveling direction) with respect to the target stop position MP of the train TR. Therefore, normally, when the train TR stops, the image sensor SE1 detects the lower edge of the front surface of the leading vehicle of the train TR as the front end portion TB. In this case, the distance Hd is small, and therefore the difference between the distance LG to the position PV on the image and the distance L to the actual position PR is small, and the detection accuracy of the stop position PRB is improved.

  However, it is also possible to install the image sensor SE1 on the front side in the traveling direction (the rear side in the traveling direction) with respect to the target stop position MP of the train TR. In that case, normally, when the train TR stops, the image sensor SE1 detects the upper edge of the front surface of the head vehicle of the train TR as the front end portion TB.

  That is, in FIG. 6, the train TR is stopped at a position PR ahead of the position PS where the image sensor SE1 is installed. Even in this case, the position detection unit 11 uses the distance Hd from the ground GR to the front end portion TB of the train TR and the distance Hc from the ground GR to the image sensor SE1 to express the equation (2) described above. It is possible to determine the distance L, that is, the actual position PR of the tip portion TB. The position PR is obtained with an accuracy close to the resolution of the pixels of the detection image FG2 of the image sensor SE1.

Thus, according to the position detection device 1 of the present embodiment, the position PR of the train TR can be detected with high accuracy with a simple configuration.
[Second Embodiment]
Next, an example in which a plurality of image sensors SE are installed along the traveling direction of the train TR will be described.

  FIG. 7 is a block diagram showing a configuration of a position detection apparatus 1B according to the second embodiment of the present invention, FIG. 8 is a diagram showing an arrangement example of two image sensors SE1 and SE2, and FIG. 9 shows a plurality of image sensors SE. FIG. 10 is a perspective view illustrating an example of an installed station ST, and FIG. 10 is a diagram illustrating an example of an image FG captured by the lamp image sensor SD1.

  Note that in the second embodiment, descriptions of parts having the same functions as those in the first embodiment will be omitted or simplified. Further, the second embodiment can be appropriately combined with the configuration of the first embodiment. The same applies to the third and subsequent embodiments.

  In FIG. 7, the position detection device 1B includes an image sensor SE1, 2, a moving body sensor SF1, a lamp image sensor SD1, a base image imaging instruction unit 10, a position detection unit 11B, a door opening / closing detection unit 15, and a safety fence door opening / closing instruction unit. 16 and a vehicle speed control instruction unit 17 and the like.

  8 and 9, image sensors SE1 and SE2 are installed at the same high position along the traveling direction of train TR. The image sensors SE1 and SE2 are installed so that the imaging ranges SA1 and SA2 overlap each other. The interval between two adjacent image sensors SE is, for example, about 2 to 3 meters. Although not shown in the figure, even when three or more image sensors SE are installed, the imaging ranges SA of the image sensors SE adjacent to each image sensor SE are installed so as to overlap.

  The lamp image sensor SD1 is installed on the front side in the traveling direction of the train TR relative to the image sensors SE1 and SE2 so as to capture an open / close display lamp DL that displays the open / closed state of the door of the train TR. If the image sensor SE1 is installed at a position where the open / close display lamp DL is imaged, the lamp image sensor SD1 can be omitted by using the lamp image FG4 captured by the image sensor SE1. It is.

  The moving body sensor SF1 is installed on the further front side of the lamp image sensor SD1 so as to detect the entry of the train TR.

  The base image capturing instruction unit 10 receives the detection signal from the moving body sensor SF1, generates base image capturing instruction signals S1a and S1b, and outputs them to the image sensors SE1 and SE2. At the timing when the base image capturing instruction signals S1a and S1b are input, the image sensors SE1 and SE2 output the respective base images FG1. Note that the output timings of the base image capturing instruction signals S1a and S1b are slightly shifted, but they may be the same timing.

  The base image capturing instruction unit 10 also outputs a base image capturing instruction signal S1c to the lamp image sensor SD1. At the timing when the base image capturing instruction signal S1c is input, the lamp image sensor SD1 outputs the base image FG3.

  The position detection unit 11B is provided with comparison units 21 and 22 for comparing the base image FG1 and the detection image FG2. The comparison units 21 and 22 compare these images and detect the position of the train TR.

  The door open / close detection unit 15 detects the open / closed state of the door of the train TR based on the detection window WD (see FIG. 10) of the base image FG3 and the lamp image FG4 captured by the lamp image sensor SD1. The door opening / closing detection unit 15 is provided with a comparison unit 23 for comparing the base image FG3 and the lamp image FG4. The comparison unit 23 compares these images and detects the open / closed state of the door. The door open / close detection unit 15 detects the open / closed state of the door based on, for example, a change in color or luminance of the image of the open / close display lamp DL. When it is detected that the door is opened, a door opening signal S4 indicating that is output from the door opening / closing detection unit 15.

  The safety fence door opening / closing instruction unit 16 is based on the door opening / closing state detected by the door opening / closing detection unit 15 and the stop position PRB of the train TR detected by the position detection unit 11, and the platform PF where the train TR stops. A signal S5 for controlling the opening / closing drive of the door DR of the safety fence SK provided in the is output.

  For the safety fence SK provided with a large number of doors that can be opened and closed at an arbitrary position as disclosed in Patent Document 2, for example, the stop position PRB and the door open signal S4 or signal S5 are used. Based on the stop position PRB, it is possible to control to open and close only the door DR at a position corresponding to the door of the train TR. In that case, the control device for the safety fence SK may acquire data about the positional relationship between the position of the front end TB of the train TR and the position of the door in advance.

  The vehicle speed control instruction unit 17 calculates the speed of the train TR based on the change in the position of the front end portion TB of the train TR in the frame output from the image sensor SE. Then, in order to stop the train TR at the target stop position, the calculated speed signal S6 is output to the speed control device of the train TR. Note that the position of the front end TB of the train TR in the frame can be detected as the position PRA as described above, for example.

Thus, the position PR of the train TR can be detected with high accuracy with a simple configuration also by the position detection device 1B of the present embodiment. Further, it is possible to control so that only the door at the position corresponding to the stop position of the train TR is opened and closed at an appropriate timing. Furthermore, it becomes easy to stop the train TR at the target stop position by detecting the speed of the train TR and outputting it to the speed control device of the train TR.
[Third Embodiment]
Next, an example in which a plurality of image sensors SE are configured to control on / off of the operation as the train TR enters will be described.

  FIG. 11 is a block diagram showing the configuration of a position detection apparatus 1C according to the third embodiment of the present invention, and FIG. 12 is a timing chart showing how a plurality of image sensors SE are turned on / off as the train TR enters.

  In FIG. 11, the position detection apparatus 1C includes image sensors SE1 to SE4, a moving body sensor SF1, a position detection unit 11C, a moving body passage determination unit 18, an operation control unit 19, and the like.

  The image sensors SE1 to SE4 are installed so that the respective imaging ranges SA overlap along the traveling direction of the train TR as in the second embodiment. The moving body sensor SF1 is installed on the front side of the image sensor SE4 so as to detect the entry of the train TR.

  The moving body passage determination unit 18 determines that the train TR has passed in front of each of the image sensors SE1 to 4 when the front end TB of the train TR has passed the imaging range SA of each of the image sensors SE1 to SE4. . The moving body passage determination unit 18 outputs a passage signal S7 corresponding to the image sensor SE when the passage of the train TR is detected.

  The operation control unit 19 controls on / off of the operations of the image sensors SE1 to SE4 based on the detection signal S1 from the moving body sensor SF1 and the passing signal S7 from the moving body passage determination unit 18. That is, the operation of the image sensors SE1 to SE4 is controlled by the operation control unit 19, and the operation is turned on only when imaging is necessary, and the operation is turned off when imaging is not necessary. Yes.

  As shown in FIG. 12, the image sensors SE1 to SE4 are normally off. When the moving body sensor SF1 detects the approach of the train TR and outputs the detection signal S1, the operation control unit 19 includes the image sensors SE4 and SE4. Turn 3 on. First, the image sensor SE4 detects the front end portion TB of the train TR, but the train TR passes through the imaging range SA of the image sensor SE4.

  When the moving body passage determination unit 18 detects that the front end TB of the train TR has passed the imaging range SA of the image sensor SE4, the moving body passage determination unit 18 turns off the image sensor SE4 and the traveling direction of the train TR with respect to the image sensor SE4. The second image sensor SE2 is turned on.

  Next, the image sensor SE3 detects the front end portion TB of the train TR. When the train TR passes through the imaging range SA of the image sensor SE3, the moving body passage determination unit 18 detects it, and the image sensor SE3 is detected. While turning off, the image sensor SE1 installed in the second traveling direction of the train TR with respect to the image sensor SE3 is turned on.

  Furthermore, when the train TR passes through the imaging range SA of the image sensor SE2, the moving body passage determination unit 18 turns off the image sensor SE2.

  When the train TR stops in the imaging range SA of the image sensor SE1, the position detection unit 11C detects the position of the front end TB of the train TR at that time as the stop position PRB. Then, when the train TR departs and the leading end TB passes through the imaging range SA of the image sensor SE1, the operation control unit 19 turns off the image sensor SE1.

  Thus, the image sensors SE1 to SE4 are controlled to be turned off and on by the operation control unit 19, thereby reducing power consumption and extending the life of the image sensors SE1 to SE4.

  Note that as the on / off control of the operation of the image sensor SE by the operation control unit 19, the power supply of the image sensor SE may be controlled, and control is performed so that only the imaging operation is turned on / off while the power is supplied. May be.

In the above example, the train TR has been described as stopping in the imaging range SA of the image sensor SE1 out of the four image sensors SE1 to SE4. However, the train TR may be stopped in the imaging range SA of the other image sensors SE2 to SE4. Of course it is possible.
[Embodiments for other moving objects]
In the 1st-3rd embodiment described above, the case where a mobile body was train TR1 was demonstrated. The train TR travels on the track SR, and such a train TR includes a train, a monorail car, a truck, and the like. Further, the moving body may be an automobile and an airplane that travel on a road surface without a track, an article that moves on a conveyor, and the like. Next, various methods for detecting the position of the moving body will be described.

  In the following embodiments, since the method for detecting the position of the moving body is basically the same as that of the first embodiment described above, the description will focus on points specific to each moving body.

  FIG. 13 is a diagram for explaining a method for detecting a stop position when the moving body is a truck TRT.

  In FIG. 13, the position L of the front end TB of the truck TRT is detected in the same manner as in the case of the train TR. When the position of the truck TRT is detected, the load on the truck TRT can be sent out from the hopper corresponding to the position. Such a hopper may be provided so as to be movable along the traveling direction of the truck TRT, for example, above the truck TRT. If it does in that way, according to the stop position PRB of the truck TRT, a load can be discharged | emitted from right above the truck TRT and can be loaded on the truck TRT. In addition, even when the truck TRT is traveling, it is possible to run the hopper along the position PRA of the truck TRT and load the load on the truck TRT without stopping the truck TRT.

  FIG. 14 is a diagram for explaining a method for detecting a stop position when the moving body is an automobile TRC.

  In FIG. 14, the position L of the front end portion TB of the automobile TRC is detected in the same manner as in the case of the train TR. In the case of the automobile TRC, since the position of the automobile TRC in the direction perpendicular to the traveling direction varies, it is desirable to make the width of the detection window WD larger than that of the train TR.

  FIG. 15 is a diagram for explaining a stop position detection method when the mobile body is an airplane TRH, and FIG. 16 is a diagram for explaining another example of a stop position detection method when the mobile body is an airplane TRH. It is a top view. 15 shows an example in which the image sensor SE1 is installed above the airplane TRH, and FIG. 15 shows an example in which the image sensor SE1 is installed on the side of the airplane TRH.

  In FIG. 15, the position L of the front end portion TB of the airplane TRH is detected in the same manner as in the case of the train TR. In the case of the airplane TRH, the image sensor SE1 and the moving body sensor SF1 may be attached to, for example, a building or tower near the boarding gate.

  In FIG. 16, the image sensor SE1 is installed so that the tire TAY of the airplane TRH is imaged from the side, that is, the imaging direction is the horizontal direction. On the back side of the route on which the airplane TRH enters, a background plate GRB to be the background of the airplane TRH is provided up to the same height as the tire TAY.

  The image sensor SE1 can image the tire TAY of the airplane TRH together with the background plate GRB in the imaging range SA. The distance Hc from the background plate GRB to the image sensor SE1 is known in advance. A distance Hd from the background plate GRB to the tip portion TB of the tire TAY of the airplane TRH is measured by a distance sensor.

  Therefore, also in this case, the position L of the front end TB of the airplane TRH is detected in the same manner as in the case of the train TR.

  FIG. 17 is a diagram for explaining a method for detecting a stop position when the moving body is the article TRS on the conveyor GRC.

  In the case of the article TRS shown in FIG. 17, since the actual position PR of the tip end portion TB is the same as the position PV on the image, the distance L is equal to the distance LG.

  Next, a flow of processing for detecting the stop position PRB in the position detection devices 1, 1B, 1C will be described with reference to a flowchart.

  FIG. 18 is a flowchart showing the flow of processing for detecting the stop position PRB in the position detection device.

In FIG. 18, when the moving body sensor SF1 detects the entry of the moving body (Yes in # 11), the image sensor SE1 acquires the base image FG1 (# 12), and acquires the detection image FG2 (# 13). Based on these images, the position PV on the image of the tip portion TB of the moving body is detected (# 14), and the actual position PR of the tip portion TB is detected by calculation (# 15). The actual position PR may be output from time to time. When a stop determination signal indicating that the moving body has stopped is output (Yes in # 16), the stop position PRB of the moving body is detected and output (# 17).
[Other examples for acquiring base images]
In each of the embodiments described above, the image sensor SE1 acquires the base image FG1 when the moving object sensor SF1 detects the ingress of the moving object. However, when a plurality of image sensors SE are provided, when another image sensor SE2 provided in front of one image sensor SE1 detects the entry of the moving body, the image sensor SE1 detects the base image FG1. May be obtained. That is, the presence or absence of the moving body is determined based on the image captured by the image sensor SE2, and when it is determined that the moving body has entered, the adjacent image sensor SE1 acquires the base image FG1. In this case, the image sensor SE2 also functions as the moving body sensor SF1.

  In addition, the base image FG1 can be acquired based on an image (frame) captured by each image sensor SE without providing the moving body sensor SF1 and without substituting the function of the image sensor SE2. It is.

  That is, the frame output from the image sensor SE is compared with the immediately preceding frame, and the frame before the frame in which the moving object appears appears as the base image FG1. Next, the position detection device 1D configured as described above will be described.

  FIG. 19 is a block diagram showing a configuration of a position detection apparatus 1D according to the fourth embodiment of the present invention, FIG. 20 is a flowchart showing a position detection procedure in the position detection apparatus 1D, and FIG. 21 shows a base image acquisition procedure. A flowchart and FIG. 22 are flowcharts showing a procedure for detecting a stop position.

  In FIG. 19, the position detection device 1D includes an image sensor SE1 and a position detection unit 11D.

  Referring also to FIGS. 20 to 22, the image sensor SE <b> 1 continuously outputs, for example, 30 frames per second image by imaging (# 21).

  The position detection unit 11D acquires the base image FG1 based on the frame (image) output from the image sensor SE1 (# 22), and calculates and outputs the position PR of the moving body (# 23).

  That is, the position detection unit 11D is provided with the base image acquisition unit 31, the comparison unit 32, the stop determination unit 33, and the calculation unit 34.

  The base image acquisition unit 31 compares the mth frame with the (m + 1) th frame (m is an integer) for the frame output from the image sensor SE1 (# 31). The presence / absence of a change is determined from the difference between both frames, and when there is a change, it is determined that a moving body has entered (# 32). At this time, in order to reduce false detection, the presence or absence of the change is determined by paying attention to the end of the imaging range SA of the image sensor SE1 on the side where the moving body enters.

  If it is determined that a moving body has entered, the immediately preceding frame, that is, the m-th frame in this example, is acquired as the base image FG1 (# 33). Note that a slightly previous frame such as the (m-1) th or (m-2) th may be used as the base image FG1. The acquired base image FG1 is stored in the memory.

  After acquiring the base image FG1, the comparison unit 32 sets the frame output from the image sensor SE1, that is, the nth frame (n is an integer, n> m) as the detection image FG2, and the detection image FG2 and the base image The image FG1 is compared (# 41). The comparison unit 32 acquires the edge image FG4 based on those differences (# 42).

  The stop determination unit 33 compares the edge image FG4 (position information) of the adjacent frames, that is, the edge image FG4 of the nth frame and the edge image FG4 of the (n + 1) th frame (# 43), Based on the difference, it is determined whether or not there is a change. If there is no change (Yes in # 44), it is determined that the moving body has stopped (# 45).

  Based on the edge image FG4, the calculation unit 34 calculates the actual position PR of the tip portion TB using the above-described equation (2) (# 46). When the calculation is performed based on the edge image FG4 determined to be stopped by the stop determination unit 33, the stop position PRB is obtained. In this way, the position detection unit 11D calculates and outputs the position PRA and the stop position PRB of the moving body.

Note that the base image FG1 acquired by the base image acquisition unit 31 may be updated when the next base image FG1 is detected.
[Other examples of reference part]
Next, another example of the reference portion will be described.

  In each embodiment described above, the example which used front-end | tip part TB, such as train TR, as a reference | standard part was demonstrated. That is, the tip portion TB in each of the above embodiments is a portion that is located at the most distal portion of the train TR when imaged by the image sensor SE. In the example of FIG. 2, it is the lower edge of the front end surface of the leading vehicle of the train TR, and in the example of FIG. 6, it is the upper edge of the front end surface of the train TR.

However, various other parts or objects can be used as the reference part. An example will be described below.
(1) As shown in FIG. 23, a vertex portion TB2 that is an intersection with a vertical line (vertical line) on the lower edge of the train TR and closer to the platform is used as a reference portion. In this case, marking may be performed by applying a color paint or a fluorescent paint to the apex portion TB2 or attaching a phosphor or a fluorescent seal. Or you may attach to the vertex part TB2 by making a light-emitting body, a reflecting plate, etc. into a target.

Further, as shown in FIGS. 24A and 24B, a reference target constituted by housing a plurality of light emitting diodes 53, 53,... Mounted on a substrate 52 in a casing 51 and attaching a transparent cover 54. The reference target 50 may be attached to the apex TB2 such that the cover 54 is on the upper surface.
(2) A horizontal reference line is drawn using color paint or fluorescent paint at a predetermined height position at the center or lower part of the front end surface of the train TR, and this reference line is used as a reference part. Alternatively, instead of drawing a reference line, a rod-like reference member made of metal or synthetic resin is attached as a reference portion. In this case, the distance Hd can be set in advance.
(3) Draw a reference line using a color paint or fluorescent paint on the front end surface, roof, or lower side of the train TR, or attach an appropriate reference member or reference target 50 in advance. That is, since the position can be specified by imaging with the image sensor SE1 within the range of the field of view of the image sensor SE1, the position within the range that can be imaged with the image sensor SE1 is used as a reference according to the mounting position of the image sensor SE1. It is only necessary to apply a paint to the reference portion, attach a seal, or attach an appropriate reference member or reference target 50.
(4) If there is almost no gap between the platform on which the train TR stops and the car body of the train TR and the images of both appear to overlap, the vertical line on the edge of the train TR or the edge near the platform on the front face The intersection with the horizontal line on the edge of the upper surface of the platform may be used as the reference portion. In this case, since the error due to the gap between the platform and the train TR can be ignored, the distance L can be calculated using the height of the platform as the distance Hd. Therefore, the distance Hd can be set in advance, and the distance L can be easily calculated.

  According to each embodiment described above, the position of the moving body can be detected with high accuracy with a simple configuration. Both the traveling position PRA and the stop position PRB can be detected as the position of the moving body.

  Further, by detecting the open / close state of the door of the train TR, the door at a position corresponding to the stop position of the train TR can be automatically opened / closed at an appropriate timing. It is also possible to control the train TR to stop at the target stop position by detecting the speed of the train TR based on the detection image FG2.

  Further, by installing a plurality of image sensors SE along the approach direction of the moving body, detection is possible even when the stop position of the moving body varies greatly. In that case, by controlling the image sensor SE to operate only when a moving body enters, it is possible to reduce power consumption and extend the life of the image sensor SE.

  In the embodiment described above, a digital camera that captures a still image may be used as the image sensor SE. In that case, a configuration may be adopted in which imaging is performed by an external signal and an image is output. The position detectors 11, 11B, 11C, 11D, etc., constituting the position detectors 1, 1B, 1C, 1D, and the like are based on a computer system configured such that an appropriate program stored in a ROM or the like is executed by a CPU or DSP. It is possible to realize. In that case, a semiconductor memory, a magnetic disk, other peripheral elements, an image processing circuit, a logic processing circuit, an interface circuit, a display device, an input device, and the like may be provided as necessary.

  In addition, the image sensor SE, the moving body sensor SF, the lamp image sensor SD, the position detection units 11, 11B, 11C, and 11D, the stop determination unit 12, the entire position detection devices 1, 1B, 1C, and 1D, or the structure and configuration of each unit The shape, dimensions, number, arrangement, image content, type of moving object, content or order of detection processing, and the like can be appropriately changed in accordance with the spirit of the present invention.

It is a block diagram which shows the structure of the position detection apparatus of 1st Embodiment which concerns on this invention. It is a figure for demonstrating the detection method of the stop position of a train. It is a figure which shows the example of installation of an image sensor. It is a figure which shows the example of a base image and the image for a detection. It is a figure explaining the detection process of the stop position by a detection window. It is a figure for demonstrating the other detection method of the stop position of a train. It is a block diagram which shows the structure of the position detection apparatus of the 2nd Embodiment of this invention. It is a figure which shows the example of arrangement | positioning of two image sensors. It is a perspective view which shows the example of the station in which the several image sensor was installed. It is a figure which shows the example of the image which the lamp image sensor imaged. It is a block diagram which shows the structure of the position detection apparatus of the 3rd Embodiment of this invention. It is a timing diagram which shows the operation state of an image sensor. It is a figure for demonstrating the detection method of the stop position of a truck. It is a figure for demonstrating the detection method of the stop position of a motor vehicle. It is a figure for demonstrating the detection method of the stop position of an airplane. It is a figure for demonstrating the other example of the detection method of the stop position of an airplane. It is a figure for demonstrating the detection method of the stop position of the articles | goods on a conveyor. It is a flowchart which shows the flow of the position detection process of a position detection apparatus. It is a block diagram which shows the structure of the position detection apparatus of the 4th Embodiment of this invention. It is a flowchart which shows the procedure of the position detection in a position detection apparatus. It is a flowchart which shows the procedure of base image acquisition. It is a flowchart which shows the procedure of a stop position detection. It is a figure which shows the vertex part of a train. It is a figure which shows the example of the structure of a reference | standard target.

Explanation of symbols

1, 1B, 1C, 1D Position detection device 11, 11B, 11C, 11D Position detection unit 12 Stop determination unit 13 Stop position determination unit 14 Target stop position storage unit 15 Door open / close detection unit 16 Safety fence door open / close instruction unit 17 Vehicle speed Control instruction unit 18 Moving object passage determination unit 19 Operation control unit 31 Base image acquisition unit 32 Comparison unit 33 Stop determination unit 34 Calculation unit 50 Reference target (reference unit)
SE Image sensor SF Moving body sensor SD Lamp image sensor SK Distance sensor FG1 Base image FG2 Detection image PR Actual position PRB at the tip PRB Stop position PV Position TR on the image TR Train (mobile)
TB tip (reference part)
TB2 vertex (reference part)
WD detection window SA Imaging range

Claims (16)

A position detection device for detecting the position of a moving body,
An image sensor installed so as to be able to capture an image of the moving body together with an image of a background of the moving body;
A position at which the position of the moving body is detected by comparing a base image that is captured by the image sensor and that is an image in which the moving body is not captured and a detection image that is an image in which the moving body is captured. A detection unit;
A position detection device comprising: The position detector is
Using the position PV on the image of the reference part of the moving object shown in the detection image, the distance Hd from the background to the reference part, and the distance Hc from the background to the image sensor, the moving object is used. The actual position PR of the reference part of
The position detection device according to claim 1. A stop determination unit for determining whether or not the moving body has stopped,
The position detection unit detects a stop position that is a position of a reference unit of the moving body when the stop determination unit determines that the stop is performed;
The position detection device according to claim 2. The image sensor outputs frames constituting an image at predetermined time intervals,
The stop determination unit compares the position information for each frame output from the image sensor with the position information of the immediately preceding frame, and when the amount of change in the position of the reference unit of the moving body becomes a predetermined value or less, Determining that the moving body has stopped;
The position detection device according to claim 3. A stop position determination unit for determining whether or not the movable body has stopped at a normal position;
The stop position determination unit determines that the stop position is stopped at a normal position when the stop position detected by the position detection unit is within an allowable range for the target stop position of the moving body.
The position detection device according to claim 3 or 4. The image sensor is installed on the rear side in the traveling direction from the target stop position of the moving body,
The position detection device according to claim 3. A plurality of the image sensors are installed along the traveling direction of the moving body,
The position detection device according to claim 2. The image sensor is installed so that each imaging range overlaps with an imaging range of an adjacent image sensor,
The position detection device according to claim 7. The image sensor outputs frames constituting an image at predetermined time intervals,
The frame output from the image sensor is compared with the immediately preceding frame, and the frame before the frame in which the moving object appears appears as the base image.
The position detection device according to claim 2. A moving body sensor for detecting the approach of the moving body is installed on the front side in the traveling direction of the moving body from the position where the image sensor is installed,
The image sensor is controlled to capture the base image when the moving body sensor detects the entry of the moving body.
The position detection device according to claim 2. The position detection unit extracts, as a detection window, a rectangular region that is long along the traveling direction of the moving body from an image captured by the image sensor, and based on the extracted image of the detection window, Find the actual position PR of the reference part,
The position detection device according to claim 2. The position detection unit detects the position of the moving body based on a difference in luminance or color between the base image and the detection image.
The position detection device according to claim 2. A moving body passage determining unit that determines that the moving body has passed when the reference portion of the moving body has passed the imaging range of the image sensor;
When the moving body passes, the operation of the image sensor is turned off.
The position detection device according to claim 7 or 8. A moving body passage determining unit that determines that the moving body has passed when the reference portion of the moving body has passed the imaging range of the image sensor;
When the moving body passes, the operation of the image sensor installed in the second moving direction of the moving body with respect to the image sensor is turned on.
The position detection device according to claim 7 or 8. A distance sensor for measuring a distance Hd from the background to the reference portion is installed;
The position detection device according to claim 2. The moving body is a traffic vehicle;
The image sensor outputs frames constituting an image at predetermined time intervals,
The speed of the moving body is calculated based on a change in the position of the reference portion of the moving body in the frame, and the calculated speed is sent to the speed control device of the moving body to stop the moving body at a target stop position. Output for
The position detection device according to claim 2.

Images