JP2014174903A - Object detection system, object detection method, and object detection program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To accurately detect, from an image, an object located in a detection region by reducing the opportunity of imaging a user or the like located in front of the detection region in a state in which the user is overlaid on the detection region.SOLUTION: An object detection system 11 includes cameras 31 and 32 (33 and 34), sensors 41-44, and an object detection unit 18. The cameras 31 and 32 (33 and 34) are installed in front of a track 23 (24), and image the tracks 23 and 24. The sensors 41-44 detect a train located on the tracks 23 and 24. When no train is detected in imaging areas of the cameras 31 and 32 (33 and 34), the object detection unit 18 detects an object located on the track 24 (23) from the images of the cameras 31 and 32 (33 and 34).

Description

  The present invention relates to an object detection system, an object detection method, and an object detection program for detecting an object located on a moving path from an image obtained by capturing the moving path of a moving body.

  2. Description of the Related Art Conventionally, an object detection system that detects an object from an image obtained by imaging a track may be used in order to detect a user's belongings or a user falling on a track at a station or the like (for example, Patent Documents). 1). In the object detection system, a camera that captures an adjacent track is installed for each home, and an object on the adjacent track is detected from an image captured by each camera.

JP 2012-11989

  In such an image of the track, the user's head located on the platform may be picked up by overlapping the track, and in this case, it is erroneously detected that the object is positioned on the track. was there. Further, when a part of the track enters the blind spot of the home, an object located on the track may not be detected.

  Accordingly, an object of the present invention is to reduce the chance that an object positioned in front of the moving path on which the moving body moves is reflected in the image and detected as an object positioned on the moving path, and moves with high accuracy. The object is to provide a technique for detecting an object located on a road.

  The object detection system according to the present invention includes an image acquisition unit, a moving body detection unit, and an object detection unit. In addition, the object detection method according to the present invention is a method in which processing corresponding to the configuration of the image acquisition unit, the moving body detection unit, and the object detection unit is executed by the information processing apparatus. Furthermore, the object detection program according to the present invention is a program installed in the information processing apparatus, and causes the information processing apparatus to execute processing corresponding to the configuration of the image acquisition unit, the moving object detection unit, and the object detection unit. It is a program.

  The image acquisition unit acquires an image in which the moving path is captured by a camera installed on the side of the moving path. The image acquisition unit may acquire a plurality of images captured by a plurality of cameras installed on both sides of the moving path or a plurality of images captured by a plurality of cameras installed along the moving path. Good. The camera may be a far infrared camera.

  A moving body detection part detects the moving body located in the imaging area of a camera on a moving path. The moving body detection unit can be configured using, for example, ultrasonic sensors that are provided at both ends of the platform of the station and detect a train located on the track.

  When the moving object is not detected, the object detection unit detects, from the image, an object located in a facing area on the opposite side of the moving path from the camera side. Further, the object detection unit stops detecting the object located in the facing area from the image when the moving body is detected.

  Thus, by detecting the object located in the facing area, it is possible to prevent the object located in front of the moving path from being reflected in the image and being erroneously detected as the object located on the moving path. In addition, if there is a moving body located in the imaging area of the camera on the moving path, it will be difficult to accurately detect an object located in the facing area. Not performed. For this reason, the object located in the opposing area can be detected accurately.

  Note that it is preferable to acquire a plurality of images with a plurality of cameras that are installed on both sides of the moving path and in which part of the imaging areas overlap each other. By detecting objects from both sides of the moving path in this way, it is possible to accurately detect objects located in each area of the moving path.

  Further, the images may be acquired by a plurality of cameras that are installed along the moving path and in which the imaging areas partially overlap each other. In this case, the camera is preferably a far infrared camera. By detecting the object from both ends along the moving path, it is possible to detect the object even from an area in the vicinity of the camera that is difficult to capture with a single image from each camera. Further, the far-infrared camera can detect an object in a very long distance range. Therefore, by combining these configurations, even if the range where the object needs to be detected is very wide, the number of imaging points can be suppressed and the object can be detected from the entire range where the object should be detected. .

  Moreover, you may determine whether a mobile body is located in an opposing area | region. In this case, when it is determined that the moving body is located in the facing area, it is preferable to detect an object located in the near area opposite to the facing area from the image. Thereby, even if the moving body is located in the opposing area, the object located in the near area can be detected.

  Different output signals may be output when an object located in the near area is detected and when an object located in the facing area is detected. In this way, it becomes possible to perform appropriate processing later according to the accuracy with which the object is positioned on the moving path. When an object located in the near area is detected, it is determined whether the position where the object is detected is an area close to the facing area, and the position where the object located in the near area is detected is the facing area. Different output signals may be output when it is determined that the region is close to the region and when it is determined that the region is not close to the opposite region.

  According to the present invention, when a moving body such as a train is not located on a moving path such as a track, an object located in the facing area is detected, so an object located in front of the moving path is reflected in the image. Thus, the chance of being detected as an object located on the moving path can be reduced, and an object located on the moving path can be detected with high accuracy.

It is the schematic which shows the platform of the station in which the camera is installed. It is a schematic diagram which illustrates the near area | region and opposing area | region set to the image which a camera images. It is a schematic diagram which shows the structure of the principal part of an object detection system. It is a flowchart which shows an example of an object detection process. It is a flowchart which shows an example of an object detection step. It is a schematic diagram explaining the range of the detection area set to each image in an object detection step. It is a flowchart which shows an example of an output step.

  Hereinafter, an object detection system, an object detection method, and an object detection program according to embodiments of the present invention will be described.

  In the example of the object detection system described in this embodiment, a camera installed on a platform of a station images a track adjacent to the platform, and detects an object located on the track from the image. That is, the object detection system will be described using a train as a moving body and a track on which the train moves as a moving path. In this object detection system, when an object located on the track is detected, a signal to that effect is output to the external device, and the external device is caused to perform an alarm or the like.

  FIG. 1 is a schematic diagram showing a set of homes where cameras of the object detection system are installed. FIG. 1A is a side view of a set of homes as viewed from the direction in which the track extends. FIG. 1B and FIG. 1C are plan views of a pair of homes as viewed from above.

  Station 1 has platforms 21 and 22 and tracks 23 and 24. The home 21 and the home 22 are opposed to each other with the lines 23 and 24 interposed therebetween. The home 21 faces the track 23. The home 22 faces the track 24.

  The home 21 is provided with far-infrared cameras 31 and 32 near both ends in the direction in which the line 23 extends. The home 22 is provided with far-infrared cameras 33 and 34 near both ends in the direction in which the line 24 extends.

The far-infrared cameras 31 to 34 each take a temperature distribution image (hereinafter simply referred to as an image). FIG. 2 is a schematic view illustrating an example of an image captured by the far-infrared camera 31. Note that images captured by the far-infrared cameras 32 to 34 are similar to the images illustrated in FIG.
As shown in the image illustrated in FIG. 2, the far-infrared camera 31 captures the vicinity of the center in the extending direction of the tracks 23 and 24 near the center of the image and places the end of the home 21 facing the track 23 in the image. Further, as shown in FIG. 1, the imaging area A on the obliquely lower side is imaged from the upper imaging point near one end of the home 21.
Similarly, the far-infrared camera 32 is located above the other end of the home 21 so that the vicinity of the center of the tracks 23 and 24 is reflected in the vicinity of the center of the image and the end of the home 21 facing the track 23 is included in the image. The imaging area B on the diagonally lower side from the imaging point is imaged. The far-infrared camera 33 is viewed from an upper imaging point near one end of the home 22 so that the center of the tracks 23 and 24 appears in the vicinity of the center of the image and the end of the home 22 facing the track 24 is included in the image. The imaging area C on the diagonally lower side is imaged. The far-infrared camera 34 has an upper imaging point near the other end of the home 22 so that the vicinity of the center of the lines 23 and 24 appears in the vicinity of the center of the image, and the end of the home 22 facing the line 24 is included in the image. The imaging area D on the diagonally lower side is imaged.

  Further, in order to detect a train or the like located on the track 23, the home 21 is provided with ultrasonic sensors 41 and 42 near both ends in the direction in which the track 23 extends. The home 22 is provided with ultrasonic sensors 43 and 44 in the vicinity of both ends in the direction in which the track 24 extends in order to detect a train or the like located on the track 24. Specifically, the ultrasonic sensor 41 is disposed near one end of the platform 21 toward the track 23 and detects a train entering the track 23. The ultrasonic sensor 42 is disposed near the other end of the platform 21 toward the track 23, and detects a train that goes out from the track 23. The ultrasonic sensor 43 is disposed near one end of the home 22 toward the track 24 and detects a train entering the track 24. The ultrasonic sensor 44 is disposed near the other end of the platform 22 toward the track 24 and detects a train that leaves the track 24.

Although details will be described later, detection areas 51 and 52 that are targets for detecting the position of the object are set in each image of the imaging areas A, B, C, and D captured by the far-infrared cameras 31 to 34. The detection area 51 is an image of the space on the line 23 as shown in the area corresponding to FIGS. The detection area 52 is an image of the space on the line 24 as shown in the area corresponding to FIGS.
The detection area 51 corresponds to a front area when an object is detected from each image in the imaging areas A and B, and corresponds to a facing area when an object is detected from each image in the imaging areas C and D. is there. On the other hand, the detection area 52 corresponds to a facing area when an object is detected from each image in the imaging areas A and B, and corresponds to a front area when an object is detected from each image in the imaging areas C and D. is there.

Further, as will be described in detail later in each of the imaging areas A, B, C, and D captured by the far-infrared cameras 31 to 34, as shown in FIGS. 1B and 2, attention areas 53 and 54 are displayed. And warning areas 55 and 56 are set. The attention areas 53 and 54 are areas in which images of the imaging areas A, B, C, and D are highly likely to be captured by overlapping the heads of the users passing through the homes 21 and 22. Specifically, the caution area 53 is an area corresponding to a part of the space along the edge facing the home 21 of the detection area 51 shown in FIGS. The attention area 54 is an area corresponding to a part of the space along the edge facing the home 22 of the detection area 52 shown in FIGS. In addition, the warning areas 55 and 56 are areas where there is a low possibility that the heads of the users passing through the homes 21 and 22 overlap and are captured in the images of the imaging areas A, B, C, and D. The warning area 55 is an area corresponding to the space excluding the attention area 53 of the detection area 51 shown in FIGS. The warning area 56 is an area corresponding to the space excluding the attention area 54 of the detection area 52 shown in FIGS.
As shown in FIG. 2, in the image of the imaging area A, there is a possibility that the head of the user 100 passing through the home 21 overlaps in the attention area 53 on the home side. On the other hand, in the attention area 54 on the opposite home side, there is no possibility that the users who pass through the home 22 overlap and are imaged. Therefore, in the image of the imaging area A, only the attention area 53 located on the own home side is treated as the attention area in the image, and the attention area 54 located on the opposite home side is not treated as the attention area in the image. The entire area 52 may be handled as a warning area in the image.

  Further, the far infrared cameras 31 to 34 may be provided in each of the homes 21 and 22 because the distance range that can be imaged is long (for example, about 150 m). However, in that case, it is difficult to image an area in the vicinity of each far-infrared camera (for example, about 40 m). Therefore, in this embodiment, as shown in FIG. 1, the far-infrared cameras 31 and 32 are provided in the same home 21 so that the area | region of a mutual vicinity is located inside each imaging area. Further, the far-infrared cameras 33 and 34 are provided in the same home 22 so that the regions in the vicinity of each other are located in the respective imaging areas. As a result, the entire detection areas 51 and 52 are captured by the far-infrared cameras 31, 32, 33, and 34, respectively. Moreover, the blind spot by the structure on the homes 21 and 22 in each far-infrared camera 31-34 can be supplemented with another far-infrared camera. Thereby, it is possible to prevent an object detection from being overlooked.

  FIG. 1C shows areas corresponding to the imaging areas A, B, C, and D of the far-infrared cameras 31, 32, 33, and 34, respectively. For example, the partial area 511 located near one end of the detection area 51 is out of the imaging areas A and C, and is captured only in the images of the imaging areas B and D. The partial area 512 adjacent to the partial area 511 of the detection area 51 is out of the imaging area C and is captured only in the images of the imaging areas A, B, and D. A partial area 513 adjacent to the partial area 512 of the detection area 51 is imaged in the images of the imaging areas A, B, C, and D. A partial area 514 adjacent to the partial area 513 of the detection area 51 is out of the imaging area D and is captured only in the images of the imaging areas A, B, and C. A partial area 515 adjacent to the partial area 514 of the detection area 51 and located near the other end of the detection area 51 is out of the imaging areas B and D, and is captured only in the images of the imaging areas A and C.

  Further, the partial area 521 located near one end of the detection area 52 is out of the imaging areas A and C, and is captured only in the images of the imaging areas B and D. A partial area 522 adjacent to the partial area 521 of the detection area 52 is out of the imaging area A, and is captured only in the images of the imaging areas B, C, and D. A partial area 523 adjacent to the partial area 522 of the detection area 52 is imaged in the images of the imaging areas A, B, C, and D. A partial area 524 adjacent to the partial area 523 of the detection area 52 is out of the imaging area B and is captured only in the images of the imaging areas A, C, and D. A partial area 525 adjacent to the partial area 524 of the detection area 52 and located near the other end of the detection area 52 is out of the imaging areas B and D, and is captured only in the images of the imaging areas A and C.

  FIG. 3 is a block diagram of the object detection system 11.

  The object detection system 11 includes an information processing apparatus 12 together with the far-infrared cameras 31 to 34 and the ultrasonic sensors 41 to 44 described above.

  The far-infrared cameras 31 to 34 operate according to control signals input from the information processing apparatus 12, take images of the respective imaging areas A, B, C, and D, and output those images to the information processing apparatus 12. . For example, the far-infrared cameras 31 to 34 periodically receive control signals from the information processing apparatus 12 so as to capture about 5 to 10 images per second. Note that a normal camera using visible light may be used instead of the far-infrared cameras 31 to 34. Further, a stereo camera or a TOF camera that can capture a distance image may be used.

  The ultrasonic sensors 41 to 44 detect trains and the like located in the homes 21 and 22 and output a moving body detection signal to the station service system and the information processing apparatus 12 that manage the operation of the trains. The moving body detection signal may be directly input to the information processing apparatus 12 from the ultrasonic sensors 41 to 44, or indirectly from the ultrasonic sensors 41 to 44 via the station service system or the like. May be entered. In addition, the moving body detection signal may be repeatedly output to the information processing apparatus 12 at a constant period, or may be output to the information processing apparatus 12 every time the ultrasonic sensors 41 to 44 are changed.

  The information processing apparatus 12 is installed with an object detection program for executing object detection processing. The object detection process is a process for detecting an object located in the detection areas 51 and 52 using the images of the imaging areas A, B, C, and D received from the far-infrared cameras 31 to 34.

  Therefore, the information processing apparatus 12 includes a control unit 13 and a storage unit 14. The control unit 13 includes an image acquisition unit 16, a moving body detection unit 17, an object detection unit 18, and an output unit 19 as functional blocks for controlling the entire information processing apparatus 12 and executing object detection processing. .

  The storage unit 14 includes various program codes including an object detection program, images received from the far-infrared cameras 31 to 34, moving body detection signals received from the ultrasonic sensors 41 to 44, and the imaging areas A, B, C, Data such as area defining parameters that define the ranges of the detection areas 51 and 52, the attention areas 53 and 54, and the warning areas 55 and 56 in each image of D are stored.

  FIG. 4 is a schematic flowchart of object detection processing.

  First, the control unit 13 performs the image acquisition step by causing the image acquisition unit 16 to function (S11). Specifically, the image acquisition unit 16 outputs a control signal to the far-infrared cameras 31 to 34, and each image of the imaging areas A, B, C, and D captured by the far-infrared cameras 31 to 34 that have received the control signal. Is received and stored in the storage unit 14.

  Next, the control unit 13 performs the moving body detection step by causing the moving body detection unit 17 to function (S12). The mobile body detection unit 17 detects the presence or absence of a mobile body such as a train located in each of the homes 21 and 22 based on the mobile body detection signal received from the ultrasonic sensors 41 to 44 and stored in the storage unit 14. . The presence or absence of a moving body such as a train located in each of the homes 21 and 22 is determined by the moving body detection unit 17 based on time-series data indicating time transitions of the moving body detection signals expressed as on / off of the ultrasonic sensors 41 to 44. Can be estimated.

  In addition, when the information processing apparatus 12 estimates the presence or absence of a train or the like located on the tracks 23 and 24, the information processing apparatus 12 corresponds to the moving body detection unit described in the claims. Moreover, in the station administration apparatus which manages operation management of a train etc., the presence or absence of the train etc. which are located in the tracks 23 and 24 is estimated from the output of the ultrasonic sensors 41-44, and the moving body detection data which shows the estimated result May be output from the station service device to the information processing device 12. In this case, the station service device corresponds to the moving object detection unit described in the claims.

  Next, the control unit 13 performs the object detection step by causing the object detection unit 18 to function (S13). FIG. 5 is a more detailed flowchart of the object detection step. The object detection unit 18 executes the following steps for each image.

  First, the object detection unit 18 reads out the moving body detection signal, the i-th image, and the area defining parameter from the storage unit 14 (S21).

  If the moving body detection signal indicates that the train is located in the near area (one of the detection areas 51 and 52) of the image read in step (S21), the object detection unit 18 The object detection step for the image is finished as it is.

  On the other hand, the object detection unit 18 indicates that the moving body detection signal indicates that the train is not located in the front area (one of the detection areas 51 and 52) of the image read in step (S21), and is opposed to the object detection section 18. When the moving body detection signal indicates that the train is not located in the area (the other of the detection areas 51 and 52), the facing area is set in the read image based on the area defining parameter (S22). ).

  FIG. 6A is a diagram specifically showing the facing areas set in the images of the imaging areas A, B, C, and D in this step. For example, in the image of the imaging area A captured by the far-infrared camera 31, an area where the partial areas 523, 524, and 525 are captured is set as the facing area. In the image of the imaging area B captured by the far-infrared camera 32, the area where the partial areas 523, 522, and 521 are captured is set as the facing area. In the image of the imaging area C captured by the far-infrared camera 33, the areas where the partial areas 513, 514, and 515 are captured are set as opposing areas. In the image of the imaging area D captured by the far-infrared camera 34, the areas where the partial areas 513, 512, and 511 are captured are set as the facing areas.

  Next, in the flowchart of FIG. 5, the object detection unit 18 detects an object from the facing area set in the image (S23). In this case, since a moving body such as a train is not located between the two platforms, the detection accuracy of the object can be increased by detecting the object from the facing area where the user or the like is not easily reflected.

  Then, when an object is detected from the facing area set in the image, the object detection unit 18 stores object detection data indicating that the object is in a warning state in which the object is located in the facing area in the storage unit 14. (S24).

  Further, the object detection unit 18 indicates that the moving body detection signal indicates that the train is not located in the front area (one of the detection areas 51 and 52) of the image read in step (S21), and is opposed to the object detection section 18. When the moving body detection signal indicates that the train is located in the area (the other of the detection areas 51 and 52), the front area is set in the read image based on the area defining parameter (S25). ).

  FIG. 6B is a diagram specifically showing a near area set in each image of the imaging areas A, B, C, and D in this step. For example, in the image of the imaging area A captured by the far-infrared camera 31, the area where the partial areas 512, 513, 514, and 515 are captured is set as the front area. In addition, in the image of the imaging area B captured by the far-infrared camera 32, the area where the partial areas 514, 513, 512, and 511 are captured is set as the front area. In the image of the imaging area C captured by the far-infrared camera 33, an area where the partial areas 522, 523, 524, and 525 are captured is set as the near area. In addition, in the image of the imaging area D captured by the far-infrared camera 34, the area where the partial areas 524, 523, 522, and 521 are captured is set as the front area.

  Next, in the flowchart of FIG. 5, the object detection unit 18 detects an object from the near area set in the image (S26). In this way, when a moving body such as a train is located on the opposite platform side and each far-infrared camera 31 to 34 cannot capture the opposite area, the user or the like is not in the near area. However, by detecting the object, it is possible to always detect the object regardless of the presence or absence of the vehicle in the facing area.

  Then, when an object is detected from the near area set in the image, the object detection unit 18 sets the attention area and the warning area in the image again based on the area defining parameter (S27). Then, the object detection unit 18 determines whether or not the object detected from the near area set in the image is located in the attention area (S28). When the object is not located in the attention area, the object detection unit 18 stores object detection data indicating that there is a warning state in which there is a risk that the object is located in the near area in the storage unit 14. When the object is located in the attention area, the object detection data indicating that the object is in a state of caution that may be captured by overlapping the object is stored in the storage unit 14 (S30). ). As described above, when an object is detected using the front area, an object area is detected by further setting a warning area in which a user or the like is easily reflected and a warning area in which the user or the like is not easily reflected. Thus, the object detection accuracy can be increased.

  For detecting an object from each image, background image data is registered in advance for each image in the imaging areas A, B, C, and D, and each image in the imaging areas A, B, C, and D and each background are detected. You may perform based on the difference with image data. Alternatively, each image in the imaging areas A, B, C, and D may be compared with an image captured immediately before, and an object may be detected based on the difference.

  Next, in the flowchart of FIG. 4, the control unit 13 performs an output step by causing the output unit 19 to function (S14). FIG. 7 is a more detailed flowchart of the output step.

  The output unit 19 determines whether object detection data is stored in the storage unit 14 (S31). When the object detection data is stored in the storage unit 14, the output unit 19 determines whether the object detection data indicates a caution state or a warning state (S32). When the object detection data indicates a caution state, the output unit 19 outputs a caution signal to the external device (S33). If the object detection data indicates a warning state, the output unit 19 outputs a warning signal to the external device (S34).

  The information processing apparatus 12 executes an object detection process for detecting an object located on the detection areas 51 and 52 by causing the control unit 13 to repeat the flowchart shown in FIG. Thereby, the object located in each of the detection areas 51 and 52 can be accurately detected from the images of the far-infrared cameras 31 to 34 provided on the homes 21 and 22.

  The external device that has received the warning signal from the information processing device 12 performs a process such as displaying a warning sound or displaying a warning display, and the station clerk or the like places the object in the detection areas 51 and 52. It is good to warn you. In addition, the external device that has received the attention signal performs a process such as displaying a notification with a warning sound or displaying a warning display, and the user who passes the end of the platforms 21 and 22 to the station staff or the like. It is advisable to pay attention to objects located in the areas 53 and 54. In this way, the information processing device 12 detects the position of the object, and switches and outputs the caution signal and the warning signal according to the accuracy with which the object is positioned on the track, so that appropriate processing can be performed by the external device. Can be realized.

In the above description, the workflow for detecting the object by dividing the near area into the attention area and the warning area in each image has been described. However, the entire near area may be handled as the attention area. In addition, although the workflow for detecting an object using the entire opposite area as a warning area is shown, the opposite area may be further divided into a caution area and a warning area. For example, in the image of the imaging area A shown in FIG. 2, the attention area 54 on the opposite home side may be handled as it is as a part of the attention area in the image. In addition, the entire facing area may be treated as a warning area, and the area where the vertical wall of the facing home located above the facing area on the image is imaged may be treated as a caution area. Although an example in which a set of cameras is arranged at the home has been described, one camera may be arranged at each home. Further, although an example in which a set of cameras is installed in the opposing home has been described, a camera may be installed only in one home. In addition, each camera may be installed in a direction orthogonal to the home in a plan view. Moreover, although the example which utilizes an ultrasonic sensor as a moving body detection part was shown, you may utilize another sensor. Regarding other configurations, different configurations may be adopted as appropriate.

DESCRIPTION OF SYMBOLS 1 ... Station 11 ... Object detection system 12 ... Information processing apparatus 13 ... Control part 14 ... Memory | storage part 16 ... Image acquisition part 17 ... Moving body detection part 18 ... Object detection part 19 ... Output part 21,22 ... Home 23,24 ... Lines 31-34 ... Cameras 41-44 ... Ultrasonic sensors 51, 52 ... Detection areas 53, 54 ... Caution areas 55, 56 ... Warning areas

Claims (8)

An image acquisition unit that acquires an image of the moving path captured by a camera installed on a side of the moving path on which the moving body moves;
A moving body detection unit for detecting a moving body located in the imaging area of the camera on the moving path;
When the moving body is not detected by the moving body detection unit, an object located in the facing area on the opposite side of the moving path from the camera side is detected from the image acquired by the image acquisition unit, An object detection unit that stops detecting an object located in the facing area when the moving object is detected by a moving object detection unit;
An object detection system comprising: The image acquisition unit acquires a plurality of images with a plurality of cameras that are installed on both sides of the moving path and in which a part of the imaging area overlaps each other.
The object detection system according to claim 1. The image acquisition unit acquires images with a plurality of cameras that are installed along the moving path and in which a part of the imaging area overlaps each other;
The object detection system according to claim 1 or 2. The moving body detection unit determines whether or not the moving body is located in the facing area;
When the moving body detecting unit determines that the moving body is located in the facing area, the object detecting unit moves from the image acquired by the image acquiring unit to the near side on the camera side in the moving path. Detect the objects located,
The object detection system in any one of Claims 1-3. An output unit that outputs different output signals when the object detection unit detects an object located in the near area and when the object detection unit detects an object located in the facing area; ,
The object detection system according to claim 4. When the object detection unit detects an object located in the near area, the object detection unit determines whether the position where the object is detected is an area close to the facing area;
The output unit outputs different output signals when it is determined that the position where the object is detected is in an area close to the facing area and when it is determined that the position is not close to the facing area. Output,
The object detection system according to claim 5. An image acquisition step of acquiring an image in which the moving path is captured by a camera installed on the side of the moving path on which the moving body moves;
A moving object detecting step for detecting a moving object located in the imaging area of the camera on the moving path;
When the moving body is not detected in the moving body detecting step, an object located in an opposite area opposite to the camera side in the moving path is detected from the image acquired in the image acquiring step, An object detection step for stopping detection of an object located in the facing area when the moving object is detected in the moving object detection step;
An object detection method for the information processing apparatus to execute. A program installed in an information processing apparatus,
An image acquisition step of acquiring an image in which the moving path is captured by a camera installed on the side of the moving path on which the moving body moves;
A moving object detecting step for detecting a moving object located in the imaging area of the camera on the moving path;
When the moving body is not detected in the moving body detecting step, an object located in an opposite area opposite to the camera side in the moving path is detected from the image acquired in the image acquiring step, An object detection step for stopping detection of an object located in the facing area when the moving object is detected in the moving object detection step;
An object detection program for causing the information processing apparatus to execute.

Images