JP2015191437A - Driving support device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a technique capable of appropriately performing a notification when there is a possibility that an object such as a head and a hand sticking out of an opening constituting a window and a sliding roof may come into contact with an obstacle in the vicinity of a vehicle.SOLUTION: On the assumption that a self vehicle reaches a position of an obstacle 12 in coordinate axes set in a travel direction, when a vertical or horizontal interval H or W between the self vehicle 1 and the obstacle 12 is within a prescribed range, determination means determines a possibility of contact therewith, and notification means performs a notification. This constitution can appropriately perform the notification only when there is a possibility that an object such as the head and the hand sticking out of an opening constituting a window and a sliding roof may come into contact with the obstacle 12 in the vicinity of the own vehicle.

Description

  The present invention relates to a technique for recognizing an obstacle around a host vehicle by processing a captured image of a shooting unit mounted on the host vehicle.

  Conventionally, the passenger's head or hand jumping out of the window or sunroof without the driver noticing that the passenger, such as a child sitting in the passenger seat or rear seat, has put his head or hand out of the window or sunroof. However, there was a risk of contact with obstacles such as guards (viaducts, overpasses), tunnels, utility poles / telegraph poles, road signs, and oncoming vehicles. As described above, in order to prevent the passenger's head or hand jumping out of the window or sunroof from coming into contact with an obstacle around the host vehicle, the following techniques have been proposed (for example, Patent Documents). 1). That is, there is provided detection means for detecting that an object such as a passenger's head or hand is protruding from an opening forming a window or sunroof. Then, when it is determined that the vehicle is traveling based on the vehicle speed or the like, if the detection means detects that the object is popping out from the opening such as a window or sunroof, the object pops out of the vehicle from the opening. The driver is notified by an alarm sound or the like.

Japanese Utility Model Publication No. 7-46571 (pages 1 to 3, FIGS. 1 to 4)

  In the above-described conventional apparatus, when an object is popping out from an opening such as a window or a sunroof during traveling, a notification by an alarm sound or the like is always executed, which causes the following problem. That is, a) When the surrounding environment is complex when the road environment is complicated, the face is exposed to the outside from the opening such as a window, etc. b) The distance confirmation between the own vehicle and the roadside groove when passing through a narrow road C) When the face is out of the vehicle through the opening of the window, etc. c) In the situation where the face is out of the vehicle through the opening of the window or the like when traveling in parallel parking or entering the garage In addition, the driver may feel annoyance because the alarm sound or the like is notified.

  The present invention provides a technique capable of appropriately notifying only when an object such as a head or a hand protruding from an opening forming a window or a sunroof may come into contact with an obstacle around the host vehicle. The purpose is to do.

  In order to achieve the above-described object, the driving support device of the present invention includes an acquisition unit that acquires a captured image around the host vehicle by a shooting unit mounted on the host vehicle, and a plurality of captured images each having a different shooting angle. Recognizing means for recognizing an obstacle included in the vehicle, and determining a relative position and a relative distance of the obstacle from the host vehicle based on the position of the obstacle in the captured images recognized by the recognizing means. A calculating means for calculating; a detecting means for detecting whether or not an object is protruding from an opening forming a window or sunroof of the own vehicle; and the calculating means when the detecting means detects the object. Determining means for determining the possibility of contact between the object and the obstacle based on the relative position and the relative distance calculated by the step, and the contact by the determining means An informing means for informing when it is determined that the own vehicle has reached the position of the obstacle on the coordinate axis set in the traveling direction; and It is characterized in that the possibility of contact is determined when a vertical or horizontal interval with the obstacle is within a predetermined value (Claim 1).

  Further, the acquisition unit acquires a plurality of time-series captured images obtained by a monocular camera as the imaging unit, and the recognition unit includes the respective captured images acquired by the acquisition unit. Recognizing an obstacle, the calculating means is configured to determine the relative position and the relative position based on the time-series movement of the obstacle in the captured images recognized by the recognition means and the movement of the host vehicle. The relative distance may be calculated (claim 2).

  Further, the determination means determines that there is a possibility of contact when the time for the host vehicle to reach the obstacle calculated based on the relative distance and the vehicle speed is within a predetermined time. (Claim 3).

  According to the first aspect of the present invention, a captured image around the host vehicle by the shooting unit mounted on the host vehicle is acquired by the acquiring unit, and obstacles included in each of the plurality of captured images having different shooting angles are provided. Recognized by recognition means. The relative position and relative distance of the obstacle from the host vehicle are calculated by the calculation means based on the position of the obstacle in each captured image recognized by the recognition means. In addition, when the detection unit detects whether or not an object has protruded from the opening of the vehicle's window or sunroof, and the detection unit detects the object, the relative position and relative value calculated by the calculation unit are detected. Based on the distance, the possibility of contact between the object and the obstacle is determined by the determining means.

  At this time, contact is made when the vertical or horizontal distance between the host vehicle and the obstacle is within a predetermined value, assuming that the host vehicle has reached the position of the obstacle on the coordinate axis set in the traveling direction. The determination means determines that there is a possibility. When the determination unit determines that there is a possibility of contact, the notification unit performs notification. Therefore, it is possible to appropriately notify only when there is a possibility that an object such as a head or a hand protruding from an opening forming a window or sunroof may come into contact with an obstacle around the host vehicle.

  According to the second aspect of the present invention, a plurality of time-series photographed images obtained by the monocular camera as the photographing means are obtained by the obtaining means, and the obstacle included in each of the photographed images obtained by the obtaining means is obtained. Recognized by recognition means. Then, based on the time-series movement of the obstacle in each captured image recognized by the recognition means and the movement of the own vehicle (shooting position), the relative position and the relative distance of the obstacle viewed from the own vehicle. Is calculated by the calculating means. Therefore, it is possible to calculate the relative position and distance of an obstacle from the host vehicle by the so-called motion stereo method using an inexpensive monocular camera without using multiple cameras or an expensive stereo camera. can do.

  According to the invention of claim 3, in addition to the conditions described in claim 1, when the time for the host vehicle to reach the obstacle calculated based on the relative distance and the vehicle speed is within a predetermined time, It is determined by the determining means that there is a possibility of contact between an object such as a head or hand protruding from an opening such as a window or sunroof and an obstacle around the host vehicle. Therefore, it is possible to appropriately notify only when an object such as a head or a hand protruding from the opening is more likely to contact an obstacle around the host vehicle.

It is a block diagram of one embodiment of a driving support device of the present invention. It is a figure which shows an example of a picked-up image. It is a figure for demonstrating the method to judge whether the object which protrudes from the opening parts, such as a window and a sunroof, may collide with the obstruction contained in the picked-up image of FIG. It is a flowchart for operation | movement description of FIG.

  An embodiment of the present invention will be described with reference to FIGS.

  FIG. 1 is a block diagram of an embodiment of the driving support apparatus of the present invention, and FIG. 2 is a diagram showing an example of a captured image. 3 is a diagram for explaining a method for determining whether or not there is a possibility that an object protruding from an opening such as a window or a sunroof collides with an obstacle included in the captured image of FIG. FIG. 4 is a flowchart for explaining the operation of FIG. FIG. 3 is a diagram schematically showing the situation of the captured image shown in FIG.

1. Configuration The driving support apparatus 2 shown in FIG. 1 processes a captured image IMG of a monocular camera 3 (corresponding to the “photographing means” of the present invention) mounted on the host vehicle 1 and shooting the periphery of the host vehicle 1. 1 to recognize obstacles 12 around the road (for example, a tunnel or a road sign shown in FIG. 2). The driving support device 2 has an opening portion when an object such as a child's head or hand, a pet's head, or the like protrudes from an opening portion that forms a window or a sunroof while the host vehicle 1 is traveling. When there is a possibility that an object protruding from the vehicle will come into contact with the obstacle 12 around the host vehicle 1, a warning sound, a monitor display, etc. are notified to the driver.

  The driving support device 2 is acquired by the monocular camera 3, the steering angle sensor 4, the vehicle speed sensor 5, the opening degree sensor 6, the object sensor 7, the captured image IMG by the monocular camera 3, and the sensors 4-7. Storage means 8 for storing various data, ECU 9 having a microcomputer configuration for processing the captured image IMG of the monocular camera 3 and determining the possibility of collision based on outputs from the sensors 4 to 7, for example, an instrument panel And a travel monitor 10 (corresponding to “notification means” of the present invention) such as a CRT and a liquid crystal display, and a warning speaker 11 (corresponding to “notification means” of the present invention).

  In this embodiment, the monocular camera 3 is composed of a monochrome or color camera capable of photographing the front of the host vehicle 1, and is a captured image IMG continuously and momentarily photographed every predetermined time (for example, 30 ms). A signal is output (see FIG. 2). In this embodiment, the monocular camera 3 is formed by a wide-angle camera or a fish-eye camera so that the widest possible range in front of the host vehicle 1 can be photographed.

  The steering angle sensor 4 detects the steering angle of the steering wheel, and outputs data relating to the detected steering angle to the ECU 9. The vehicle speed sensor 5 detects the vehicle speed and outputs data relating to the detected vehicle speed to the ECU 9. The opening degree sensor 6 detects whether or not an opening (not shown) that forms a window or sunroof of the host vehicle 1 is opened, and outputs data relating to the opening degree of the opening to the ECU 9.

  The object sensor 7 (corresponding to the “detecting means” of the present invention) is whether or not an object (not shown) such as a child's passenger's head, hand, or pet's head protrudes from an opening such as a window or sunroof. To detect. For example, a light-blocking optical unit formed by a light projecting unit disposed at one end edge of the opening and a light receiving unit disposed at the other end edge to receive light from the light projecting unit. The object sensor 7 can be formed by the sensor. In this way, when the object is projected from the opening, the light of the light projecting unit that is projected to the light receiving unit is blocked. Therefore, the object sensor 7 determines whether or not the object is projected from the opening. Can be detected. Note that the object sensor 7 may be configured by a monocular camera for monitoring the side of the host vehicle 1 disposed on the side mirror, an ultrasonic sensor disposed near an opening of a window, a sunroof, or the like.

  The storage unit 4 stores image data of the captured image IMG output from the monocular camera 3 every moment, various data obtained by performing various processes on the captured images IMG by the ECU 9, and the sensors 4. Store various data output from .about.7.

  The ECU 9 includes the following units by, for example, executing a preset driving support program when the ignition key is turned on.

(1) Acquisition means 9a
The acquisition unit 9 a acquires image data of a time-series captured image IMG that is captured every moment by the monocular camera 3 and stored in the storage unit 8. FIG. 2 shows a captured image IMG in front of the host vehicle 1 captured by the monocular camera 3 disposed in the front of the host vehicle 1. Various data related to the captured image IMG captured in time series by the monocular camera 3 is stored in the storage unit 8.

(2) Recognizing means 9b
The recognizing means 9b is a fault which is a stationary object such as a guard (bypass, overpass), tunnel, utility pole / telegraph pole, road sign, curb, pole, wall, etc. included in each captured image IMG acquired by the acquiring means 9a. The object 12 is recognized by pattern matching. Templates such as guards (viaducts, overpasses), tunnels, utility poles / telegraph poles, road signs, curbs, poles, walls, etc. for recognizing the obstacles 12 in the captured image IMG by pattern matching by the recognition means 9b. The image is stored in the storage unit 8. In the example shown in FIG. 2, a tunnel exists in front of the host vehicle 1, a road sign exists in front of the host vehicle 1, and an opening portion of the tunnel included in the captured image IMG in front of the host vehicle 1 The road sign is recognized by the recognition means 9b as an obstacle 12 by pattern matching. Data relating to the obstacle 12 recognized by the recognition means 9b is stored in the storage means 8.

  Note that the pattern recognition method by the recognition means 9b is not limited to the pattern matching method described above, and the obstacle 12 can be recognized by detecting its contour line, such as a method using a density histogram or a method using edge detection. Any method may be adopted as long as it can be performed. In this embodiment, the obstacle 12 that is a stationary object is a target for obstacle recognition by the recognition unit 9b. However, a moving object such as an oncoming vehicle or the like may be a target for obstacle recognition by the recognition unit 9b. Of course.

(3) Calculation means 9c
The calculating unit 9c calculates the relative position and the relative distance of the obstacle 12 from the host vehicle 1 based on the positions of the obstacle 12 in the plurality of captured images IMG recognized by the recognition unit 9b and having different shooting angles. To do. Specifically, in this embodiment, the calculation unit 9c includes a change in time-series movement of the obstacle 12 in the plurality of time-series captured images IMG recognized by the recognition unit 9b, and a steering angle sensor. 4 and the change of the movement (shooting position) of the host vehicle 1 between the shots of the shot images IMG calculated based on the outputs of the vehicle speed sensor 5, the host vehicle 1 of the obstacle 12 by the so-called motion stereo method. The relative position and relative distance seen from the above are calculated.

(4) Determination means 9d
The determination unit 9d determines the possibility of contact between the obstacle 12 and an object protruding from an opening such as a window or a sunroof, based on the calculation result by the calculation unit 9c. Specifically, as shown in FIG. 3, the determination means 9 d is based on the relative distance between the host vehicle 1 and each obstacle 12 calculated by the calculation means 9 c and the vehicle speed detected by the vehicle speed sensor 5. When the calculated times T1 and T2 for the host vehicle 1 to reach each obstacle 12 are within a predetermined time (for example, within 3 seconds), the head or hand of a passenger such as a child from an opening such as a window or a sunroof If an object such as the above has popped out, it is determined that there is a possibility that the object and the obstacle 12 are in contact with each other.

  Therefore, the determination means 9d determines whether or not the openings such as windows and sunroofs are opened based on the output of the opening degree sensor 6 if the times T1 and T2 to reach each obstacle 12 are within a predetermined time. Judging. Further, the determination unit 9d determines whether or not an object is protruding from an opening such as a window or a sunroof based on the output of the object sensor 7 when the opening is open.

  Also, as shown in FIG. 3, the judging means 9d can detect each obstacle 12 on the coordinate axis in which the host vehicle 1 is set in the traveling direction (in the figure, if an object is popping out from an opening such as a window or a sunroof). The vertical distance H between the host vehicle 1 and each obstacle 12 and the horizontal position when it is assumed that the position (the area surrounded by the dotted line) has reached the position (the position of the host vehicle 1 indicated by the dotted line in the figure) The direction interval W is calculated. The determination means 9d then determines the vertical distance H between the host vehicle 1 and each obstacle 12 (e.g., the distance between the ceiling of the tunnel and the host vehicle 1) or the horizontal distance between the host vehicle 1 and each obstacle 12. When the distance W in the direction (e.g., the distance between the road sign on the right side and the own vehicle 1, the distance between the wall surface on both sides of the tunnel and the own vehicle 1) is within a predetermined value (for example, 50 cm), the opening It is determined that there is a possibility of contact between the object protruding from the obstacle and the obstacle 12.

(5) Notification means 9e
When the determination unit 9d determines that there is a possibility of contact, the notification unit 9e can contact the vehicle by displaying a warning display superimposed on the car navigation map being displayed on the travel monitor 10, for example. The presence of the sexual obstacle 12 is visually notified to the driver. In addition, the speaker 11 generates a voice message and a warning sound as a result of determination by the determination unit 9d to audibly notify the driver of the possibility of contact. The driver's driving assistance is performed by the visual or auditory notification to the driver.

2. Operation Next, the operation of the driving support apparatus 2 configured as described above will be described with reference to FIG. In this embodiment, as shown in FIG. 2 and FIG. 3, the state in which the host vehicle 1 is traveling will be described as an example.

  First, when the monocular camera 3 starts shooting every certain time by turning on the ignition key of the host vehicle 1, a signal of the captured image IMG in front of the host vehicle 1 by the monocular camera 3 as shown in FIG. It is taken in by means 9a (step S1). The obstacle 12 in the captured captured image IMG is recognized by the recognition unit 9b (step S2).

  Subsequently, the data of the obstacle 12 recognized by the recognition unit 9b is registered in the storage unit 8 (step S3). Then, the obstacle stored in the storage unit 8 indicates whether or not the same obstacle 12 recognized by the recognition unit 9b in the current captured image IMG is included in the previous captured image IMG. The calculation means 9c determines based on the data related to the object 12 (step S4). If the same obstacle 12 is not included (NO in step S4), the process returns to step S1 and the process is repeatedly executed.

  On the other hand, if the same obstacle 12 is included (YES in step S4), the relative position and the relative distance of the obstacle 12 viewed from the host vehicle 1 are calculated and calculated by the calculation means 9c by the motion stereo method. Based on the relative position and the relative distance, the height of the obstacle 12 is calculated by the calculating means 9c (step S5). Subsequently, based on the vehicle speed detected by the vehicle speed sensor 5, times T1 and T2 for reaching the obstacle 12 are calculated by the determination means 9d (step S6).

  Next, whether or not the calculated arrival times T1 and T2 are within a predetermined time is determined by the determination means 9d (step S7). If the arrival times T1 and T2 exceed the predetermined time (NO in step S7). ), Returning to the process of step S1, the process is repeatedly executed. On the other hand, if the arrival times T1 and T2 are within a predetermined time (YES in step S7), it is determined by the determination means 9d based on the output of the opening degree sensor 6 whether or not an opening such as a window or a sunroof is opened. (Step S8).

  If an opening such as a window or sunroof is not opened (NO in step S8), the process returns to step S1, and the process is repeatedly executed. If opened (YES in step S8), the output of the object sensor 7 is output. Based on the above, it is judged by the judging means 9d whether or not an object is protruding from the opening (step S9). If no object has jumped out of the opening (NO in step S9), the process returns to step S1 and the process is repeatedly executed. If the object has jumped out of the opening (YES in step S9), whether or not the distances H and W between the host vehicle 1 and the obstacle 12 in the vertical direction or the horizontal direction when passing through the obstacle 12 are within a predetermined value. Is determined by the determination means 9d (step S10).

  If both the vertical and horizontal intervals H and W between the host vehicle 1 and the obstacle 12 exceed a predetermined value (NO in step S10), the process returns to step S1 and is repeatedly executed. If the vertical or horizontal distances H and W between the host vehicle 1 and the obstacle 12 are within a predetermined value (YES in step S10), the head of a passenger such as a child can be seen from an opening such as a window or a sunroof. The notification means 9e notifies that the hand is popping out (step S11), and the process returns to step S1 to repeatedly execute the process.

  As described above, according to this embodiment, the captured image IMG around the host vehicle 1 by the monocular camera 3 mounted on the host vehicle 1 is acquired by the acquisition unit 9a, and each of the plurality of captured images IMG having different shooting angles is used. Is recognized by the recognition means 9b. Further, the relative position and the relative distance of the obstacle 12 from the host vehicle 1 are calculated by the calculating means 9c based on the position of the obstacle 12 in each captured image IMG recognized by the recognition means 9b. Further, when the object sensor 7 detects whether or not an object such as a passenger's head or hand is protruding from the opening of the vehicle 1 or the sunroof, and the object sensor 7 detects the object. Based on the relative position and relative distance calculated by the calculation means 9c, the possibility of contact between the object and the obstacle 12 is determined by the determination means 9d.

  At this time, vertical and horizontal distances H and W between the host vehicle 1 and the obstacle 12 when the host vehicle 1 is assumed to have reached the position of the obstacle 12 on the coordinate axis set in the traveling direction are predetermined. When it is within the value, it is determined by the determining means 9d that there is a possibility of contact. When the determination unit 9d determines that there is a possibility of contact, notification by the notification unit 9e is executed. Therefore, it is possible to appropriately notify only when there is a possibility that an object such as a head or a hand protruding from an opening forming a window or a sunroof may come into contact with an obstacle 12 around the host vehicle 1.

  A plurality of time-series captured images IMG obtained by the monocular camera 3 are acquired by the acquisition unit 9a, and the obstacle 12 included in each of the captured images IMG acquired by the acquisition unit 9a is recognized by the recognition unit 9b. . Then, the obstacle 12 viewed from the own vehicle 1 based on the time-series movement of the obstacle 12 in each captured image IMG recognized by the recognition means 9b and the movement of the own vehicle 1 (shooting position). Relative position and relative distance are calculated by the calculating means 9c. Therefore, the relative position and relative position of the obstacle 12 from the host vehicle 1 can be determined by the so-called motion stereo method using an inexpensive monocular camera 3 without using a plurality of cameras or an expensive stereo camera. The distance can be calculated.

  Further, in addition to the condition about the vertical or horizontal distances H and W between the host vehicle 1 and the obstacle 12, the time for the host vehicle 1 to reach the obstacle 12 calculated based on the relative distance and the vehicle speed. Is within the predetermined time, it is determined by the determining means 9d that there is a possibility that an object such as a head or a hand protruding from an opening such as a window or sunroof and an obstacle 12 around the host vehicle 1 may come into contact with each other. The Therefore, it is possible to appropriately notify only when an object such as a head or a hand protruding from the opening is more likely to come into contact with the obstacle 12 around the host vehicle 1.

  In addition, since the obstacle 12 is a stationary object such as a guard (a viaduct or a bridge), a tunnel, a utility pole / telegraph pole, a road sign, etc., the relative position and relative distance of the obstacle 12 from the host vehicle 1 are more accurate. Driving with a practical configuration that can be calculated well and can determine with high accuracy the possibility of contact between an object protruding from an opening such as a window or sunroof and an obstacle 12 around the host vehicle 1 The support device 2 can be provided.

  It should be noted that the present invention is not limited to the above-described embodiment, and various modifications other than those described above can be made without departing from the spirit thereof. The monocular camera 3 is provided not only in the front part of the vehicle 1 but also in the rear part of the own vehicle 1 for the purpose of preventing contact between the object 12 protruding out of the vehicle 1 and the obstacle 12 during back travel. Also good. A monocular camera may be disposed on the side of the host vehicle 1 for the same purpose.

  Further, the photographing means may be configured by a stereo camera, and the relative position and relative distance of the obstacle 12 as viewed from the own vehicle 1 may be calculated based on a plurality of photographed images IMG photographed simultaneously by the stereo camera.

  The present invention can be applied to driving assistance for various vehicles.

1 self-vehicle 2 driving support device 3 monocular camera
7 Object sensor (detection means)
9a Acquisition means 9b Recognition means 9c Calculation means 9d Judgment means 9e Notification means 10 Travel monitor (notification means)
11 Speaker (notification means)
12 Obstacle IMG image

Claims (3)

Obtaining means for obtaining a photographed image around the own vehicle by photographing means mounted on the own vehicle;
Recognizing means for recognizing an obstacle included in each of a plurality of photographed images with different photographing angles,
Calculation means for calculating a relative position and a relative distance of the obstacle from the host vehicle based on the position of the obstacle in each captured image recognized by the recognition means;
Detecting means for detecting whether or not an object is protruding from an opening forming the window or sunroof of the host vehicle;
A determination unit that determines the possibility of contact between the object and the obstacle based on the relative position and the relative distance calculated by the calculation unit when the detection unit detects the object;
An informing means for informing when the judging means judges that the contact possibility exists,
The determination means assumes that a vertical or horizontal interval between the host vehicle and the obstacle when the host vehicle has reached the position of the obstacle on the coordinate axis set in the traveling direction is a predetermined value. It is determined that there is a possibility of contact when it is within the range. The acquisition means acquires a plurality of time-series captured images by a monocular camera as the imaging means,
The recognizing unit recognizes the obstacle included in each of the captured images acquired by the acquiring unit;
The calculation means calculates the relative position and the relative distance based on the time-series movement of the obstacle in each captured image recognized by the recognition means and the movement of the host vehicle. The driving support device according to claim 1, wherein   The determination means determines that there is a possibility of contact when a time for the host vehicle to reach the obstacle calculated based on the relative distance and a vehicle speed is within a predetermined time. The driving support device according to claim 1 or 2.

Images