WO2016063309A1

WO2016063309A1 - Obstacle detecting device and obstacle detecting method

Info

Publication number: WO2016063309A1
Application number: PCT/JP2014/005309
Authority: WO
Inventors: 小沼　吉樹
Original assignee: 日産ライトトラック株式会社
Priority date: 2014-10-20
Filing date: 2014-10-20
Publication date: 2016-04-28

Abstract

The invention detects an obstacle with which a moving body is likely to come into contact as a result of the movement of the moving body, on the basis of an image of the view in the direction of travel of the moving body. An obstacle detecting device (10) is provided with: an image receiving unit (11) which receives a first image and a second image in which the view in the direction of travel of a moving body (1) is captured at a first time point and a second time point, after the first time point, respectively; a zone setting unit (12) which sets a first detection zone (34) within the first image and a second detection zone (36) within the second image, in accordance with a zone (33) occupied, in the zones captured in the first image and the second image respectively, by a vertical cross section of the moving body (1) perpendicular to the front-rear direction of the moving body (1) if the moving body (1) is located in a predicted location (Pa) at a third time point after the second time point; and a determining unit (13) which determines whether or not an object detected at a first location within the first detection zone (34) is an obstacle, depending on whether or not the object is detected in a second location, associated in advance with the first location, within the second detection zone (36).

Description

Obstacle detection device and obstacle detection method

The present invention relates to an obstacle detection device and an obstacle detection method.

2. Description of the Related Art As a technique for detecting an obstacle on a moving object, for example, an automobile obstacle detection device described in Patent Document 1 is known. The obstacle detection device for an automobile includes a detection area setting unit that sets an obstacle detection area according to a predicted course of the vehicle and a specific road structure around the vehicle. The obstacle detection device for an automobile includes image information acquisition means for acquiring image information taken by a camera mounted on the own vehicle and obstacle detection means for detecting an obstacle based on the image information.

JP 2009-271766 A

When an obstacle of a moving object is detected based on an image of a scene in the traveling direction of the moving object, the distance to the object shown in this image may not be known. For example, distance information to the subject is not included in the two-dimensional image. In such a case, it is difficult to determine from the position of the object on the image whether there is a possibility that the object and the moving object shown in the image may come into contact with the moving object. For example, if a nearby obstacle on the path of the moving object, a distant object not on the path of the moving object, and the photographing device are in a straight line, the positions of the obstacle and the distant object on the image are the same. It can be seen from becoming.
An object of the present invention is to detect an obstacle that may come into contact with a moving body when the moving body moves based on an image of a scene in the traveling direction of the moving body.

The obstacle detection device according to an aspect of the present invention receives an image reception of a first image and a second image obtained by capturing a scene in a moving direction of a moving object at a first time and a second time after the first time, respectively. And when the moving body is located at an expected position at the third time after the second time, the vertical cross section perpendicular to the front-rear direction of the moving body occupies the ranges occupied by the shooting ranges of the first image and the second image, respectively. A range setting unit for setting the first detection range in the first image and the second detection range in the second image, and the object detected at the first position in the first detection range at the first position. A determination unit configured to determine whether the object is an obstacle according to whether the object is detected at a second position within a second detection range associated in advance.

According to the present invention, it is possible to detect an obstacle that may come into contact with the moving body when the moving body moves based on the image of the scene in the traveling direction of the moving body.

It is a figure showing an example of composition of vehicles provided with an obstacle detection device concerning an embodiment of the present invention. It is a figure which shows the function structure of the obstruction detection apparatus which concerns on 1st Embodiment of this invention. (A) is explanatory drawing of the imaging position of a 1st image, (b) is explanatory drawing of the imaging position of a 2nd image. (A) is a figure which shows an example of a 1st image, (b) is a figure which shows an example of a 2nd image. It is a figure which shows the function structure of the judgment part which concerns on 1st Embodiment of this invention. (A) is a figure which shows an example of a 1st partial image, (b) is a figure which shows an example of a 2nd partial image. It is explanatory drawing of an example of operation | movement of the obstruction detection apparatus which concerns on 1st Example of this invention. (A) is a figure which shows another example of a 1st image, (b) is a figure which shows another example of a 2nd image. It is a figure which shows the modification of a 2nd detection range. It is a figure which shows the function structure of the obstruction detection apparatus which concerns on 2nd Embodiment of this invention. (A) And (b) is explanatory drawing of an example of operation | movement of a space | interval change part. (A) And (b) is explanatory drawing of an example of operation | movement of the modification of a space | interval change part. It is a figure which shows the function structure of the range setting part which concerns on 3rd Embodiment of this invention. It is explanatory drawing of the setting operation | movement of a 1st detection range.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.
(First embodiment)
(Constitution)
Please refer to FIG. The obstacle detection device according to the first embodiment of the present invention is mounted on, for example, a vehicle 1 that is a moving body, and is used to detect an obstacle that may come into contact with the vehicle 1. For example, the vehicle 1 may be a truck. For example, the vehicle 1 may be a panel van mounting vehicle having a cargo space with a ground clearance higher than the cabin height. The application of the obstacle detection device disclosed in this specification is not limited to this. The obstacle detection device disclosed in the present specification is used to detect an obstacle that may come into contact with a moving body as the moving body moves based on an image of a scene in the traveling direction of the moving body. can do.

The vehicle 1 includes a steered wheel 2, wheels 3 other than the steered wheel 2, a wheel speed sensor 4, a control unit 5, a steering wheel 6, a rudder angle sensor 7, a photographing device 8, and an alarm device 9. And an obstacle detection device 10.
The wheel speed sensor 4 detects the wheel speed of the steered wheel 2 or the wheel 3 and outputs a detection signal to the control unit 5. The wheel speed sensor 4 may include, for example, a pulse generator such as a rotary encoder that measures wheel speed pulses.
The control unit 5 controls, for example, an ABS (Antilock Brake System) based on the wheel speed detected by the wheel speed sensor 4. Further, the control unit 5 calculates the speed and travel distance of the vehicle 1 based on the wheel speed detected by the wheel speed sensor 4. The control unit 5 outputs a speed signal indicating the speed of the vehicle 1 and a travel distance signal indicating the travel distance of the vehicle 1 to the obstacle detection device 10.
The steering angle sensor 7 detects the steering angle of the steering wheel 6 and calculates the steering angle of the steered wheels 2. The steering angle sensor 7 outputs a steering angle signal indicating the steering angle of the steered wheel 2 to the obstacle detection device 10.

The imaging device 8 captures an image of a scene in the traveling direction of the vehicle 1. The image captured by the imaging device 8 may be a two-dimensional image, for example. When the vehicle 1 moves forward, that is, when the traveling direction of the vehicle 1 is in front of the vehicle 1, the imaging device 8 may capture an image of a scene in front of the vehicle 1. In this case, for example, the photographing device 8 may be a front camera installed in the vicinity of the rear mirror of the vehicle 1. For example, the imaging device 8 may be a camera used in a Lane Departure Warning System (LDWS) that notifies the driver of a danger when the vehicle 1 deviates from the lane.
When the vehicle 1 moves backward, that is, when the traveling direction of the vehicle 1 is behind the vehicle 1, the imaging device 8 may capture an image of a scene behind the vehicle 1. In this case, for example, the imaging device 8 may be a rear camera installed at the rear portion of the vehicle 1.

The alarm device 9 outputs an alarm to the driver of the vehicle 1. For example, the alarm device 9 may be an audible alarm device that gives an audible alarm to the driver. The auditory alarm device may be a buzzer or a speaker, for example. For example, the warning device 9 may be a visual warning device that gives a visual warning to the driver. The visual alarm device may be a meter or an indicator, for example.
The obstacle detection device 10 is an electronic circuit device including a CPU (Central Processing Unit) and CPU peripheral components such as a ROM (Read Only Memory) and a RAM (Random Access Memory). The obstacle detection device 10 may be configured by, for example, a general-purpose or dedicated logic circuit, an ASIC (Application Specific Integrated Circuit), or an FPGA (Field-Programmable Gate Array).
The obstacle detection device 10 detects an obstacle that may come into contact with the vehicle 1 when the vehicle 1 moves based on an image of a scene in the traveling direction of the vehicle 1 taken by the imaging device 8.

Hereinafter, the configuration of the obstacle detection apparatus 10 will be described. Please refer to FIG. The obstacle detection device 10 includes an image reception unit 11, a range setting unit 12, a determination unit 13, and an alarm generation unit 14. In addition, the below-mentioned process by each component with which the obstacle detection apparatus 10 is provided is implemented, for example, when CPU contained in the obstacle detection apparatus 10 runs a computer program. This computer program is stored in a ROM or RAM included in the obstacle detection apparatus 10. The computer program may be recorded on a machine-readable recording medium, read by a medium reading device (not shown), and installed in the obstacle detection device 10.
Note that FIG. 2 mainly shows the configuration related to the functions described in this specification with respect to the obstacle detection apparatus 10. The obstacle detection apparatus 10 may include other components other than the illustrated components. The same applies to FIG.
The image receiving unit 11 receives a plurality of images of the scene in the traveling direction of the moving object that the imaging device 8 has captured at a plurality of times. For example, in the first embodiment, the image receiving unit 11 receives a moving image of a scene in the traveling direction of the moving body from the imaging device 8. The moving image includes a plurality of image frames as images taken at a plurality of times. The image reception unit 11 outputs the received plurality of images to the determination unit 13.

The range setting unit 12 sets a range for detecting an obstacle in the first image at the first time and the second image at the second time included in the plurality of images received by the image receiving unit 11. Hereinafter, the range in which the obstacle is detected in the first image and the range in which the obstacle is detected in the second image may be referred to as “first detection range” and “second detection range”, respectively.
Reference is made to FIG. 3A and FIG. An arrow 30 indicates the traveling direction of the vehicle 1. Reference sign P1 indicates the position of the vehicle 1 when the first image is taken at the first time. Reference sign P2 indicates the position of the vehicle 1 when the second image is taken at the second time. In the following description, the position of the vehicle 1 when the first image is taken and the position of the vehicle 1 when the second image is taken are denoted as “shooting position P1” and “shooting position P2”, respectively. Sometimes.
The second time is later than the first time, and the vehicle 1 at the second time is in a position advanced from the position of the vehicle 1 at the first time. Reference sign Pa is an expected position of the vehicle 1 at a third time later than the second time. In the following description, the predicted position of the vehicle 1 at the third time may be simply expressed as “expected position Pa”.
FIGS. 4A and 4B show examples of the first image and the second image, respectively. For example, it is assumed that the airplane 31 and the tree 32 at the end of the road are shown in the first image and the second image.

4 (a) shows a range in which the vertical cross section of the vehicle 1 occupies the first image when the vehicle 1 located at the expected position Pa is temporarily shown in the first image. In other words, the alternate long and short dash line 33 indicates a range that the vertical cross section of the vehicle 1 located at the predicted position Pa occupies in the imaging range of the first image. Here, the vertical cross section of the vehicle 1 indicates a cross section of the vehicle 1 along a plane orthogonal to the front-rear direction of the vehicle 1. Hereinafter, the vertical cross section of the vehicle 1 may be simply referred to as “vertical cross section”. An alternate long and short dash line 35 in FIG. 4B indicates a range in which the vertical cross section of the vehicle 1 occupies the second image when the vehicle 1 located at the predicted position Pa is temporarily captured in the second image. In other words, the alternate long and short dash line 35 indicates a range in which the vertical cross section of the vehicle 1 located at the predicted position Pa occupies the shooting range of the second image.

The range setting unit 12 sets the first detection range 34 according to the range 33 in which the vertical cross section of the vehicle 1 located at the expected position Pa occupies the shooting range of the first image. For example, the first detection range 34 may be equal to the range 33. The range setting unit 12 may set the first detection range 34 by, for example, perspective-converting a vertical section of the vehicle 1 located at the predicted position Pa into a region in the first image.
Further, for example, the first detection range 34 may include a margin range extending outward of the range 33, that is, a range in which the margin range extending outward of the vertical cross section of the vehicle 1 occupies the first image capturing range. For example, in the first embodiment, the first detection range 34 is a range in which the margin areas extending outward from the upper edge region, the left edge region, and the right edge region of the vertical section of the vehicle 1 occupy the shooting range of the first image. including.

Further, the range setting unit 12 sets the second detection range 36 according to the range 35 in which the vertical cross section of the vehicle 1 located at the expected position Pa occupies the shooting range of the second image. For example, the second detection range 36 may be equal to the range 35.
For example, the second detection range 36 may include a margin range extending outward of the range 35, that is, a range where the margin range extending outward of the vertical cross section of the vehicle 1 occupies the imaging range of the second image. In this case, the dimensional ratio between the range occupied by the vertical cross section in the second image and the margin range and the dimensional ratio between the range occupied by the vertical cross section in the first image and the margin range are mutually equal. The width of the margin range of the second detection range 36 may be set.
For example, in the first embodiment, the second detection range 36 is a range in which the margin areas extending outward from the upper edge area, the left edge area, and the right edge area of the vertical section of the vehicle 1 occupy the shooting range of the second image. including.

Please refer to FIG. The range setting unit 12 outputs detection range signals indicating the first detection range 34 and the second detection range 36 to the determination unit 13.
Please refer to FIG. The determination unit 13 includes an image extraction unit 20 and an obstacle determination unit 21. The image extraction unit 20 receives a travel distance signal from the control unit 5. The image extraction unit 20 receives a moving image of a scene in the traveling direction of the moving object from the image reception unit 11. The image extraction unit 20 receives the detection range signal from the range setting unit 12.
The image extraction unit 20 recognizes the first image and the second image captured at the imaging position P2 that is separated from the imaging position P1 of the first image by a predetermined travel distance, included in the received moving image. For example, the image extraction unit 20 determines a combination of a certain first image and a second image photographed at a photographing position P2 separated from the photographing position P1 of the first image by a predetermined traveling distance based on the traveling distance signal. recognize.

Since the obstacle detection device 10 continuously detects obstacles, the image extraction unit 20 is separated from the first images at different times by a predetermined distance from the shooting positions of the first images. A plurality of second images photographed at the photographing position are recognized. For this reason, the image extraction unit 20 may recognize a certain first image as a second image taken at a shooting position that is a predetermined distance away from the shooting position of the other first image. In this case, the first detection range 34 and the second detection range 36 may be set in the same image.
The image extraction unit 20 extracts a partial image of the first detection range 34 of the first image and a partial image of the second detection range 36 of the second image. In the following description, the partial image of the first detection range 34 of the first image and the partial image of the second detection range 36 of the second image are referred to as “first partial image” and “second partial image”, respectively. There is. The image extraction unit 20 outputs the first partial image and the second partial image to the obstacle determination unit 21.

Reference is made to FIG. 6A and FIG. 6B.

Reference numerals

50 and 51 denote a first partial image and a second partial image, respectively. In FIG. 6A, the first partial image 50 is enlarged and displayed. Therefore, the size ratio between the first partial image 50 shown in FIG. 6A and the second partial image 51 shown in FIG. 6B is the actual size of the first partial image 50 and the second partial image 51. Different from size ratio. Further, on the actual first partial image 50 and the second partial image 51, the alternate long and

short dash lines

33 and 35 indicating the vertical cross section of the vehicle 1 may not be displayed.
For example, each in-image position of the second partial image 51 may be associated in advance with one of the in-image positions of the first partial image 50. For example, a case is assumed where the image position 54 of the second partial image 51 is associated with the image position 52 of the first partial image 50. For example, when the first partial image 50 is enlarged to the same size as the second partial image 51, the relative positional relationship between the reference point 53 of the vertical section of the vehicle 1 and the in-image position 52 is as follows. The relative positional relationship with the position 54 may be equal.

When the same object is detected at both the in-image position 52 and the in-image position 54, it can be determined that the object is a stationary object within the vertical section or margin of the vehicle 1 located at the expected position Pa. . For this reason, the obstacle determination unit 21 detects an object detected at both the in-image position 52 and the in-image position 54 as an obstacle that may come into contact with the vehicle 1 when the vehicle 1 moves to the expected position Pa. to decide.
Further, for example, the first partial image 50 and the second partial image 51 may have a similar relationship. The in-image position 52 and the in-image position 54 may be corresponding positions in the first partial image 50 and the second partial image 51 that are similar to each other. In other words, the in-image position 52 and the in-image position 54 may be corresponding positions in the first detection range 34 and the second detection range 36 that are similar to each other.

In the first embodiment, the first partial image 50 and the second partial image 51 are similar to each other, and are divided into a plurality of blocks by the same number of divisions. The dotted lines in FIG. 6A and FIG. 6B indicate dividing lines that divide the first partial image 50 and the second partial image 51. In the example of FIGS. 6A and 6B, the number of horizontal divisions and the number of vertical divisions are “6”. In the following description, a plurality of blocks included in the first partial image 50 and a plurality of blocks included in the second partial image 51 may be referred to as “first block” and “second block”, respectively.

Each of the second blocks is associated with one of the first blocks in advance. For example, one of the first blocks and one of the second blocks corresponding to the first block are arranged at corresponding positions in the first partial image 50 and the second partial image 51 that are similar to each other. It's okay. In other words, any of the first blocks and any of the second blocks corresponding to the first block are arranged at corresponding positions in the first detection range 34 and the second detection range 36 that are similar to each other. It may be. For example, the second block in the m-th column from the left and the n-th row from the top in FIG. 6B is the m-th column from the left and the n-th row from the top in FIG. May be associated with the first block.
The obstacle determination unit 21 may contact an object detected by both the first block and the second block corresponding to the first block when the vehicle 1 moves to the expected position Pa. Judge that there is an obstacle. For example, the obstacle determination unit 21 may determine the same image pattern or objects having an image pattern with a similarity equal to or higher than a predetermined threshold as the same object.

In the example of FIGS. 6A and 6B, the tree 32 includes the first block in the sixth column from the left and the first block in the fourth row from the top and the sixth column from the left in the first partial image 50. It is detected in the first block in the fifth row from the top and the top. The tree 32 is also detected in the second partial image 51 in the sixth block from the left and the second block in the fourth row from the top, and in the sixth column from the left and the second block in the fifth row from the top. . Therefore, the obstacle determination unit 21 determines that the tree 32 is an obstacle.
On the other hand, the airplane 31 is detected in the third column from the left and the first block in the first row from the top, the fourth column from the left and the first block in the first row from the top of the first partial image 50. Is done. However, in the second partial image 51, the airplane 31 is detected in the second block in the third column from the left and the fourth row from the top. For this reason, the obstacle determination unit 21 determines that the airplane 31 is not an obstacle.
Please refer to FIG. The obstacle determination unit 21 outputs the determination result of the presence / absence of an obstacle to the alarm generation unit 14.
Please refer to FIG. When the obstacle determination unit 21 determines that there is an obstacle, the alarm generation unit 14 generates an alarm signal and outputs the alarm signal to the alarm device 9. The alarm device 9 that has received the alarm signal outputs an alarm notifying the driver of the vehicle 1 that an obstacle has been detected. The alarm device 9 may be an auditory alarm or a visual alarm.

(Operation)
Next, an example of the operation of the obstacle detection apparatus 10 according to the first embodiment will be described. Please refer to FIG. In step S 10, the image extraction unit 20 recognizes the first image included in the moving image received from the image reception unit 11.
In step S11, the image extraction unit 20 extracts the first partial image 50 of the first detection range 34 set by the range setting unit 12 from the first image.
In step S 12, the obstacle determination unit 21 determines whether an object is detected in any of the plurality of first blocks included in the first partial image 50. If no object is detected (step S12: N), the operation returns to step S10. If an object is detected (step S12: Y), the operation proceeds to step S13.

In step S 13, the image extraction unit 20 extracts a second image captured at a shooting position P 2 that is a predetermined distance away from the shooting position P 1 of the first image recognized in step S 10 among the moving images received from the image receiving unit 11. Recognize images.
In step S14, the image extraction unit 20 extracts the second partial image 51 of the second detection range 36 set by the range setting unit 12 from the second image.
In step S15, the obstacle determination unit 21 detects an object having the same or similar image pattern as the object detected in any of the first blocks of the first partial image 50 in the second block corresponding to the first block. Determine whether or not. If no object is detected (step S15: N), the operation returns to step S10. If an object is detected (step S15: Y), the operation proceeds to step S16.
In step S 16, the alarm generation unit 14 outputs an alarm signal to the alarm device 9. The alarm device 9 that has received the alarm signal outputs an alarm notifying the driver of obstacle detection.

(Effect of 1st Embodiment)
(1) The obstacle detection device 10 according to the first embodiment takes a first image and a second image of a scene in the traveling direction of the vehicle 1 at a first time and a second time, respectively. The obstacle detection device 10 performs the first detection according to a range in which the vertical cross section of the vehicle 1 located at the predicted position Pa at the third time after the second time occupies the shooting range of the first image and the second image. A range 34 and a second detection range 36 are set. The obstacle detection device 10 determines whether or not the object detected at the first position in the first detection range 34 is detected at the second position in the second detection range 36 previously associated with the first position. In response to this, it is determined whether or not the object is an obstacle.
Therefore, by appropriately associating the first position and the second position, the relative positional relationship between the object detected at the first position and the second position and the vertical cross section of the vehicle 1 located at the expected position Pa can be determined. . Therefore, it can be determined that the detected object is an obstacle that may come into contact with the vehicle 1.
For example, the height of the luggage compartment of the panel van bodywork vehicle is higher than the height of the cabin. For this reason, when a panel van framed vehicle is driven toward the end of the road, there is a possibility that trees and the like that protrude from the road come into contact with the luggage compartment. According to the obstacle detection device 10 according to the first embodiment, it is possible to determine the relative positional relationship between an object in the vicinity of the predicted position Pa ahead and the vertical section of the cargo compartment of the panel van-mounted vehicle at the predicted position Pa. Therefore, it can be determined that an object in the vicinity of the predicted position Pa ahead is an obstacle that may come into contact with the panel van mounted vehicle.

(2) In the obstacle detection apparatus 10 according to the first embodiment, the first position and the second position associated with the first position are similar to each other in the first detection range 34 and the second detection range 36. The corresponding position in
By associating the second position with the first position in this way, it is possible to determine the relative positional relationship between the object detected at both the first position and the second position and the vertical section of the vehicle 1 located at the expected position Pa. For example, it can be determined that the detected object is a stationary object within the vertical cross section of the vehicle 1 or within a margin range provided in the vertical cross section. Therefore, it can be determined that the detected object is an obstacle that may come into contact with the vehicle 1.

(3) In the obstacle detection apparatus 10 according to the first embodiment, the first detection range 34 may be divided into a plurality of first blocks, and the second detection range 36 corresponds to each of the plurality of first blocks. May be divided into a plurality of second blocks. Any one of the first blocks and any one of the second blocks corresponding to the first block may be arranged at corresponding positions in the first detection range 34 and the second detection range 36 having a similar relationship to each other. . The first position and the second position associated with the first position may be an arrangement position of the first block and the second block associated with the first block.
By associating the second block with the first block in this way, the relative positional relationship between the object detected in both the corresponding first block and the second block and the vertical section of the vehicle 1 located at the expected position Pa is determined. it can. For example, it can be determined that the detected object is a stationary object within the vertical cross section of the vehicle 1 or within a margin range provided in the vertical cross section. Therefore, it can be determined that the detected object is an obstacle that may come into contact with the vehicle 1.
(4) In the obstacle detection device 10 according to the first embodiment, the first detection range 34 and the second detection range 36 are the upper edge region, the left edge region, and the right edge of the vertical section of the vehicle 1 that is located at the expected position Pa. Each of the ranges may be included in a margin area extending outward from the edge area. By providing a margin area in the first detection range 34 and the second detection range 36, an object outside the first detection range 34 and the second detection range 36 in which no obstacle is detected by the obstacle detection device 10 collides with the vehicle 1. Fear can be reduced.

(Modification)
Reference is made to FIG. 8A and FIG. 8B. In the example of the first image shown in (a) of FIG. 8 and the second image shown in (b) of FIG. 8, the preceding vehicle 37 is shown near the center of the

vertical sections

33 and 35 of the vehicle 1. Since the preceding vehicle 37 travels in the same direction as the vehicle 1, even if the vehicle 1 moves to the expected position Pa, there is little possibility of contact with the vehicle 1, and it may not be detected as an obstacle. However, since the preceding vehicle 37 is detected in both the first block and the second block that are associated with each other in the vicinity of the center of the

vertical sections

33 and 35, the preceding vehicle 37 may not be distinguished from the obstacle.
Thus, there is a possibility that an obstacle and other objects cannot be distinguished near the center of the

vertical cross sections

33 and 35.

For this reason, the obstacle determination unit 21 shown in FIG. 5 has the size of the object detected in the first detection range 34 on the first image and the size of the object detected in the second detection range 36 on the second image. Based on the ratio between the two, it may be determined whether the detected object is an obstacle. For example, the difference between the ratio between the size on the first image and the size on the second image and the ratio between the first detection range 34 and the second detection range 36 is not within the allowable range. In addition, the obstacle determination unit 21 may determine that the detected object is not an obstacle. By making such a determination, it is possible to distinguish a stationary object near the expected position Pa from other objects.

Further, the first detection range 34 and the second detection range 36 may be set so as not to include the vicinity of the center of the

vertical cross sections

33 and 35. Please refer to FIG. The second detection range 36 may be, for example, an inverted U shape formed by an upper side portion, a left side portion, and a right side portion. By making the second detection range 36 into an inverted U shape, the second detection range 36 does not include the vicinity of the center of the vertical section 35. For this reason, it is possible to prevent an object that is not an obstacle from being determined as an obstacle near the center of the vertical section 35.
The second detection range 36 may include a range occupied by the upper edge region, the left edge region, and the right edge region of the vertical section 35, and extend outward from the upper edge region, the left edge region, and the right edge region, respectively. The range occupied by the margin area may be included.
Further, the first detection range 34 may be set in an inverted U shape similarly to the second detection range 36.

(Second Embodiment)
Next, the obstacle detection apparatus 10 according to the second embodiment of the present invention will be described. Please refer to FIG. Constituent elements similar to those in the first embodiment shown in FIG. The obstacle detection device 10 includes a speed signal receiving unit 15 and an interval changing unit 16.
The speed signal receiving unit 15 receives a speed signal indicating the speed of the vehicle 1 from the control unit 5. The speed signal receiving unit 15 outputs the received speed signal to the interval changing unit 16.
The interval changing unit 16 generates an interval specifying signal that specifies the interval between the shooting position P1 of the first image and the shooting position P2 of the second image according to the speed of the vehicle 1. The interval of the interval between the shooting position P1 and the shooting position P2 specified by the interval specifying signal varies depending on the speed of the vehicle 1.

The interval changing unit 16 outputs an interval designation signal to the range setting unit 12 and the determination unit 13. The range setting unit 12 changes the position of at least one of the first detection range 34 and the second detection range 36 according to the interval designated by the interval designation signal. For example, when the position of the first detection range 34 is fixed, the range setting unit 12 may change the position of the second detection range 36 according to the interval designated by the interval designation signal. For example, when the position of the second detection range 36 is fixed, the range setting unit 12 may change the position of the first detection range 34 according to the interval designated by the interval designation signal. The range setting unit 12 may change the positions of both the first detection range 34 and the second detection range 36 according to the interval designated by the interval designation signal.
The image extraction unit 20 of the determination unit 13 recognizes the second image captured at the imaging position P2 that is separated from the imaging position P1 of the first image by the interval specified by the interval designation signal. In this way, the interval changing unit 16 changes the interval between the shooting position P1 of the first image and the shooting position P2 of the second image according to the speed of the vehicle 1.
Reference is made to (a) of FIG. 11 and (b) of FIG. The interval changing unit 16 may reduce the interval between the shooting position P1 of the first image and the shooting position P2 of the second image when the speed of the vehicle 1 increases.

For example, as shown in FIG. 11A, when the speed V of the vehicle 1 is “V1”, the interval D between the shooting position P1 of the first image and the shooting position P2 of the second image is “D1”. To do. As shown in FIG. 11B, when the speed V of the vehicle 1 is “V2”, which is faster than “V1”, the interval changing unit 16 captures the first image photographing position P1 and the second image photographing position P2. The distance D may be “D2” shorter than “D1”.
When the speed of the vehicle 1 increases, for example, the second time at which the second image is shot is advanced by reducing the interval between the shooting position P1 of the first image and the shooting position P2 of the second image. An alarm can be output at an early stage.
The interval changing unit 16 may increase the interval between the shooting position P1 of the first image and the shooting position P2 of the second image when the speed of the vehicle 1 increases. As a result, it is possible to reduce the time interval at which the determination unit 13 recognizes the first image and the second image as the speed of the vehicle 1 increases. Thereby, for example, inconveniences such as a processing waiting time for the first image in the image processing for the second image can be avoided.

(Effect of 2nd Embodiment)
The obstacle detection apparatus 10 according to the first embodiment changes the interval between the shooting position P1 of the first image and the shooting position P2 of the second image according to the speed of the vehicle 1. For example, the obstacle detection device 10 may reduce the interval between the shooting position P1 of the first image and the shooting position P2 of the second image when the speed of the vehicle 1 increases. As a result, for example, when the speed of the vehicle 1 increases, the second time at which the second image is captured can be advanced and an alarm can be output earlier.
Further, for example, the obstacle detection device 10 may increase the interval between the shooting position P1 of the first image and the shooting position P2 of the second image when the speed of the vehicle 1 increases. Thereby, for example, it is possible to reduce the time interval at which the determination unit 13 recognizes the first image and the second image as the speed of the vehicle 1 increases.

(Modification)
The interval changing unit 16 may generate a designation signal that designates an interval between the shooting position P2 of the second image and the expected position Pa according to the speed of the vehicle 1. The interval between the photographing position P2 designated by the designation signal and the predicted position Pa varies depending on the speed of the vehicle 1. The range setting unit 12 changes at least the position of the second detection range 36 in accordance with the interval specified by the specification signal. The range setting unit 12 may change the position of the first detection range 34 as the position of the second detection range 36 is changed. In this way, the interval changing unit 16 changes the interval between the shooting position P2 of the second image and the expected position Pa according to the speed of the vehicle 1.

Reference is made to FIG. 12A and FIG. The interval changing unit 16 may increase the interval between the shooting position P2 and the expected position Pa when the speed of the vehicle 1 increases.
For example, as shown in FIG. 12A, when the speed V of the vehicle 1 is “V1”, the interval D between the imaging position P2 and the expected position Pa is set to “D3”. As shown in FIG. 12B, when the speed V of the vehicle 1 is “V2”, the interval changing unit 16 sets “D4”, which is longer than “D3”, with the interval D between the imaging position P2 and the expected position Pa. As good as
By increasing the interval between the shooting position P2 and the predicted position Pa when the speed of the vehicle 1 increases, for example, the second time at which the second image is shot can be advanced to output an alarm earlier. It becomes possible.
The interval changing unit 16 may reduce the interval between the shooting position P2 and the expected position Pa when the speed of the vehicle 1 increases. When the speed of the vehicle 1 increases, by reducing the interval between the shooting position P2 and the expected position Pa, for example, the second time at which the second image is shot is delayed, so that a clearer shot at a closer distance is obtained. Obstacles can be detected based on the image. Thereby, when the speed of the vehicle 1 increases, detection accuracy can be improved.

(Third embodiment)
Next, an obstacle detection apparatus 10 according to the third embodiment of the present invention will be described. Please refer to FIG. The range setting unit 12 includes a turning radius estimation unit 40 and a position setting unit 41. The turning radius estimation unit 40 receives a steering angle signal from the steering angle sensor 7. The turning radius estimation unit 40 estimates the turning radius of the vehicle 1 based on the turning angle of the steered wheel 2 indicated by the steering angle signal. The turning radius estimation unit 40 may receive an image of a scene in the traveling direction of the vehicle 1 from the image receiving unit 11. The turning radius estimation unit 40 may detect the road shape of the travel path of the vehicle 1 based on the received image. For example, the turning radius estimation unit 40 may detect a road shape based on a lane marking line that appears in the received image, a white line such as a roadway outer line, or a lane marker. The turning radius estimation unit 40 may estimate the turning radius of the vehicle 1 based on the detected road shape. The turning radius estimation unit 40 outputs a turning radius signal indicating the turning radius of the vehicle 1 to the position setting unit 41.

The position setting unit 41 changes the horizontal position of one or both of the first detection range 34 and the second detection range 36 according to the turning radius of the vehicle 1. For example, the position setting unit 41 may determine the predicted position Pa according to the turning radius of the vehicle 1. When the horizontal position of the first detection range 34 is changed, the position setting unit 41 uses the first detection range in the first image of the vertical cross section of the vehicle 1 located at the expected position Pa determined according to the turning radius. Perspective conversion to 34 may be performed. When the horizontal position of the second detection range 36 is changed, the position setting unit 41 uses the second detection range in the second image of the vertical cross section of the vehicle 1 located at the expected position Pa determined according to the turning radius. Perspective conversion to 36 may be performed. As in the first embodiment, the first detection range 34 and the second detection range 36 may include a margin range.
Refer to FIG. A dotted rectangle 38 indicates an example of the position of the first detection range 34 when the vehicle 1 is traveling straight, and a solid rectangle 39 indicates the first detection range 34 when the vehicle 1 is turning right. An example of the position is shown. The horizontal position of the position 39 when the vehicle 1 is turning to the right is corrected to the right side of the position 38 when traveling straight according to the radius of rotation of the vehicle 1. Similarly, the position setting unit 41 may correct the position of the second detection range 36 to the right side of the position 38 when traveling straight according to the turning radius of the vehicle 1.

(Effect of the third embodiment)
The obstacle detection device 10 according to the third embodiment estimates the turning radius of the vehicle 1 and changes the horizontal position of at least one of the first detection range 34 and the second detection range 36 according to the turning radius of the vehicle 1. To do. Thereby, the position of the 1st detection range 34 or the 2nd detection range 36 can be set appropriately even if the vehicle 1 is turning. For this reason, it is possible to detect an obstacle that may come into contact with the vehicle 1 even while the vehicle 1 is turning.

DESCRIPTION OF SYMBOLS 1 Vehicle 4 Wheel speed sensor 7 Steering angle sensor 8 Imaging device 9 Alarm device 10 Obstacle detection device 11 Image receiving part 12 Range setting part 13 Judgment part 14 Alarm generation part

Claims

An image receiving unit that receives a first image and a second image, respectively, taken at a first time and a second time after the first time;
A range in which a vertical section perpendicular to the front-rear direction of the moving body occupies the shooting range of the first image and the second image when the moving body is located at an expected position at a third time after the second time. And a range setting unit for setting a first detection range in the first image and a second detection range in the second image, respectively.
Depending on whether or not the object detected at the first position within the first detection range is detected at the second position within the second detection range previously associated with the first position, the object A determination unit for determining whether or not is an obstacle;
An obstacle detection device comprising:
A speed signal receiving unit for receiving a speed signal indicating the speed of the moving body;
An interval changing unit that changes an interval between the shooting positions of the first image and the second image according to the speed of the moving body;
The obstacle detection device according to claim 1, further comprising:
A speed signal receiving unit for receiving a speed signal indicating the speed of the moving body;
An interval changing unit that changes an interval between the shooting position of the second image and the predicted position according to the speed of the moving body;
The obstacle detection device according to claim 1, further comprising:
A rotation radius estimation unit for estimating a rotation radius of the moving body;
The range setting unit changes a horizontal position of at least one of the first detection range and the second detection range according to the turning radius. Obstacle detection device.
The first position and the second position are positions corresponding to each other in the first detection range and the second detection range, which are in a similar relationship to each other. The obstacle detection apparatus according to 1.
The first detection range is divided into a plurality of first blocks,
The second detection range is divided into a plurality of second blocks respectively corresponding to the plurality of first blocks,
Any one of the first blocks and any one of the second blocks corresponding to any one of the first blocks are in a similar relationship with each other and corresponding positions in the first detection range and the second detection range. Placed in
The first position is an arrangement position of any one of the first blocks, and the second position is an arrangement position of any one of the second blocks. The obstacle detection apparatus according to 1.
The first detection range and the second detection range each include a range occupied by a margin region extending outward from the upper edge region, the left edge region, and the right edge region of the vertical section, respectively. The obstacle detection device according to any one of claims 6 to 6.
The obstacle according to any one of claims 1 to 7, wherein the first detection range and the second detection range have an inverted U shape formed by an upper side portion, a left side portion, and a right side portion. Object detection device.
Taking a first image and a second image of the scene in the traveling direction of the moving body at a first time and a second time after the first time,
A range in which a vertical section perpendicular to the front-rear direction of the moving body occupies the shooting range of the first image and the second image when the moving body is located at an expected position at a third time after the second time. And setting a first detection range in the first image and a second detection range in the second image, respectively,
Depending on whether or not the object detected at the first position within the first detection range is detected at the second position within the second detection range previously associated with the first position, the object An obstacle detection method characterized by determining whether or not an obstacle.