JP2016062285A - Back-and-side-monitor vehicle detection and warning apparatus - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a back-and-side-monitor vehicle detection and warning apparatus capable of improving detection accuracy for other vehicles.SOLUTION: A back-and-side-monitor vehicle detection and warning apparatus 1 comprises: a rear camera 10 imaging detection target ranges in regions at the back and side of an own vehicle; a specific pattern detection unit 16 detecting a specific pattern image which corresponds to another vehicle from the detection target ranges on the basis of an image obtained by imaging the ranges; a white line detection unit 18 detecting a white line that is an end portion of each adjacent traffic lane and that is drawn on a side opposite to a travel lane of the own vehicle from the detection target ranges on the basis of the image obtained by imaging the ranges; and a travel vehicle determination unit 20 determining that another vehicle is traveling in any of positions corresponding to the detection target ranges when the specific pattern detection unit 16 detects the specific pattern image or when the white line detection unit 18 does not detect a white line after the specific pattern image detected so far has disappeared from the detection target ranges.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a rear side vehicle detection alarm device that issues an alarm when another vehicle approaches from the rear side of a running vehicle.

  Conventionally, based on an image captured by a CCD camera or the like, a moving object existing behind the host vehicle is detected, and when there is no white line at the position of the moving object, the moving object is It is determined that the obstacle is a three-dimensional obstacle (another vehicle), and when the direction indicator is on and the degree of proximity between the host vehicle and the other vehicle is excessive, a notification is given to the driver of the host vehicle. Such a rear side vehicle alarm device is known (see, for example, Patent Document 1). In this notification device, for example, an inspection line is set from a white line on the traveling lane side to a position separated from the adjacent lane side by a distance corresponding to 1/3 of the traveling lane width, and an image on the inspection line changes with time (hours). By monitoring the aerial image, a moving body existing behind the host vehicle is detected. Among the moving bodies detected in this way, in addition to the actual moving body (vehicle) to be detected, road surface indications and seams are also included. Display and seams are eliminated.

Japanese Patent Laid-Open No. 9-240397

  By the way, in the alarm device disclosed in Patent Document 1 described above, there is a problem that the detection accuracy of other vehicles may be reduced. As described above, a moving body is detected based on an image on the inspection line, and a vehicle in which no white line exists is used as a vehicle. Indications and seams will be erroneously detected as vehicles. For example, when the white line is disappearing due to secular change, when the white line is dirty due to mud, or when the white line disappears due to road construction or the like, it can be said that the presence or absence of the white line is not clear.

  The present invention was created in view of the above points, and an object thereof is to provide a rear side vehicle detection alarm device capable of improving the detection accuracy of other vehicles.

  In order to solve the above-described problem, a rear side vehicle detection alarm device according to the present invention is based on an imaging unit that images a detection target range of a rear side region of the host vehicle, and an image obtained by imaging by the imaging unit. Based on the specific pattern detection means for detecting the specific pattern image corresponding to the other vehicle from the detection target range, and the image obtained by the imaging by the imaging means, the end of the adjacent lane and the vehicle When the specific pattern image is detected by the white line detecting means for detecting the white line drawn on the side opposite to the traveling lane from the detection target range and the specific pattern detecting means, or the specific pattern image detected until then is When the white line is not detected by the white line detection means after disappearing from the detection target range, it is determined that another vehicle is traveling in the position corresponding to the detection target range. And a traveling vehicle determining means.

  By combining the shape of the specific pattern included in the detection target range and the presence or absence of a white line, not only the shape of the specific pattern included in the image of the detection target range is detected to detect other vehicles traveling in the adjacent lane Since the other vehicle can be detected even when the shape of the specific pattern is out of the detection target range, the detection accuracy of the other vehicle can be improved. In addition, since the detection of other vehicles combined with the presence or absence of white lines on the road is performed only when the specific pattern image disappears from the image of the detection target range, the misdetection of other vehicles due to not detecting the white line Can be reduced, and a decrease in detection accuracy can be suppressed.

  In addition, the traveling vehicle determination unit described above is included in the detection target range when the white line is not detected by the white line detection unit from the time when the specific pattern image detected until then disappears from the detection target range. It is desirable to determine that another vehicle is traveling in the corresponding position. As a result, the specific pattern image disappears from the image in the detection target range because the vehicle is long, but in actuality, other vehicles are detected, including when the other vehicle continues to travel in the adjacent lane. And the detection accuracy of other vehicles can be increased.

  In addition, when the traveling vehicle determination unit described above determines that another vehicle is traveling, the lane change of the host vehicle is performed toward the adjacent lane in which the other vehicle is traveling. It is desirable to further include alarm output means for outputting a predetermined alarm to the driver of the host vehicle. Specifically, the vehicle further includes direction indication determination means for determining the operation state of the direction indicator of the host vehicle, and the warning output means determines the direction indicated by the direction indicator determined by the direction indication determination means as the lane change direction. It is desirable to do. As a result, a warning can be issued only when it is necessary to truly call attention to the driver of the host vehicle, and the driver and other passengers feel with other unnecessary warning output. Discomfort can be reduced.

  Moreover, the imaging means mentioned above is attached to the rear end part of the own vehicle, and is a single unit that simultaneously images both the left detection target range and the right detection target range set in the left and right rear side regions of the own vehicle. A camera is desirable. By detecting other vehicles traveling on the left and right adjacent vehicles using one camera, it is possible to reduce the cost by reducing the number of parts and the number of mounting steps. In addition, a system for imaging the rear of the vehicle and displaying it on a display device in the vehicle interior is widespread. However, in a vehicle equipped with such a system, a camera provided for imaging the rear of the vehicle is used as the camera of the present invention. As a result, the cost can be further reduced.

  Moreover, it is desirable that the specific pattern described above has a shape corresponding to the tire. Since tires have basically the same shape regardless of the type of vehicle, etc., by using a specific pattern having a shape corresponding to the tire, other vehicles can be used with high accuracy regardless of the type of vehicle. It becomes possible to detect.

It is a figure which shows the structure of the rear side vehicle detection alarm apparatus of one Embodiment. It is a figure which shows the mounting position and imaging range of a rear camera. It is a flowchart which shows the operation | movement procedure which detects the other vehicle of the adjacent lane which is approaching to the own vehicle, and outputs a warning. It is a figure which shows the relationship between the image obtained by imaging with a back camera, and the left side partial image and right side partial image in it. It is a figure which shows the relationship between the other vehicle currently drive | working an adjacent lane, and the right side partial image by which viewpoint conversion was carried out. It is a figure which shows the relationship between the other vehicle which is drive | working an adjacent lane, and the other right side partial image by which viewpoint conversion was carried out. It is a figure which shows the relationship between the other vehicle which is drive | working an adjacent lane, and the other right side partial image by which viewpoint conversion was carried out. It is a figure which shows the relationship between the other vehicle with a long full length currently drive | working an adjacent lane, and the right side partial image by which viewpoint conversion was carried out. It is a figure which shows the relationship between the other right-side partial image by which the other vehicle with the full length currently drive | working an adjacent lane and viewpoint conversion was carried out. It is a figure which shows the relationship between the other right-side partial image by which the other vehicle with the full length currently drive | working an adjacent lane, and viewpoint conversion.

  Hereinafter, a rear side vehicle detection alarm device according to an embodiment to which the present invention is applied will be described with reference to the drawings. FIG. 1 is a diagram illustrating a configuration of a rear side vehicle detection alarm device according to an embodiment. As shown in FIG. 1, the rear side vehicle detection alarm device 1 of the present embodiment includes a rear camera 10, an image memory 12, a detection target range image extraction unit 14, a specific pattern detection unit 16, a white line detection unit 18, and a traveling vehicle. A determination unit 20, a direction instruction determination unit 22, and an alarm output unit 24 are provided.

  The rear side vehicle detection alarm device 1 is mounted on a vehicle, and detects that another vehicle has entered the detection target range with a predetermined range on the left and right rear sides as a blind spot of the driver as a detection target range. Then, when the own vehicle is going to change the course to the adjacent lane side where another vehicle is traveling, a warning is given to the driver.

  The rear camera 10 is attached to the rear of the vehicle and has an imaging range including left and right adjacent lanes. For example, it is desirable to use a camera having a wide angle of view by using a CCD image sensor or a CMOS image sensor and attaching a fisheye lens.

  FIG. 2 is a diagram illustrating a mounting position and an imaging range of the rear camera 10. As shown in FIG. 2, the rear camera 10 is attached to the rear center of the host vehicle G, for example, at the top of the license plate. The rear camera 10 has an angle of view A set to an angle close to 180 degrees, and a part of the angle range B (<A) is used for imaging the left detection target range including the left adjacent lane, Another part of the angle range C (= B) is used to image the right detection target range including the right adjacent lane. The rear camera 10 is for displaying an image of the rear of the vehicle on a display device in the passenger compartment when the host vehicle is moving backward, and when the host vehicle is running, a part of the imaging range is set to the left detection target range or the right detection target range. Assigned.

  In the present embodiment, the left detection target range and the right detection target range are imaged using one rear camera 10, but the left camera is mounted on the left side surface of the host vehicle G (for example, the lower part of the left door mirror). The left detection target range may be imaged, and a right camera may be mounted on the right side surface (for example, below the right door mirror) to capture the right detection target range.

  The image memory 12 stores an image obtained by imaging with the rear camera 10. The detection target range image extraction unit 14 reads an image captured by the rear camera 12 and stored in the image memory 12, and from this image, a left partial image corresponding to the left detection target range and a right detection target range are extracted. The corresponding right partial image is extracted.

  The specific pattern detection unit 16 detects an image of a specific pattern included in each of the left partial image and the right partial image extracted by the detection target range image extraction unit 14. As this specific pattern, a shape that can detect the presence of a vehicle traveling in the adjacent lane on the left side or the right side, and the shape of a part that is commonly used for a plurality of vehicles is used. In the present embodiment, the shape of a tire having basically the same shape is used as the specific pattern regardless of the type of the vehicle.

  The white line detection unit 18 detects white lines included in the left partial image and the right partial image extracted by the detection target range image extraction unit 14 (white lines that divide the driving lane and yellow lines drawn in yellow other than white). Included in the target). The white line to be detected is a white line drawn at the end of the adjacent lane and on the side opposite to the traveling lane of the own vehicle (the side far from the own vehicle). This white line includes not only the white line that divides between the adjacent lane and the adjacent lane, but also the white line drawn on the shoulder of the adjacent lane and the white line drawn between the adjacent lane and the median strip. included.

  The traveling vehicle determination unit 20 detects when the specific pattern detection unit 16 detects an image of a specific pattern included in the left detection target range or the right detection target range or detects an image of a specific pattern that has been detected until then. When a white line is not detected by the white line detection unit 18 in the same left detection target range or right detection target range after disappearance from the target range or right detection target range, the left detection target range or right detection is performed. It is determined that another vehicle is traveling in the adjacent lane at the position corresponding to the target range.

  The direction indication determination unit 22 determines the operating state of the direction indicator (blinker) of the host vehicle, that is, whether it is in operation, or when it is in operation, pointing to the left or to the right.

  The warning output unit 24 determines that another vehicle is traveling in the adjacent lane by the traveling vehicle determination unit 20, and the directional indicator in operation is on the side of the adjacent lane in which the other vehicle is traveling. Is output to the driver of the vehicle. For example, the alarm output unit 24 outputs an alarm sound from a speaker or a buzzer (not shown), and displays a predetermined alarm image indicating the direction and position of another running vehicle from a display device (not shown). To do.

  The rear camera 10 described above is an imaging unit, the specific pattern detection unit 16 is a specific pattern detection unit, the white line detection unit 18 is a white line detection unit, the traveling vehicle determination unit 20 is a traveling vehicle determination unit, and the alarm output unit 24 is an alarm. The direction instruction determination unit 22 corresponds to the output instruction determination unit.

  The rear side vehicle detection alarm device 1 of the present embodiment has such a configuration, and the operation thereof will be described next. FIG. 3 is a flowchart showing an operation procedure for detecting another vehicle in the adjacent lane approaching the host vehicle G and outputting an alarm.

  The rear of the vehicle is imaged by the rear camera 10 (step 100). The obtained image is stored in the image memory 12. The detection target range image extraction unit 14 reads out the stored image, and extracts a left partial image corresponding to the left detection target range and a right partial image corresponding to the right detection target range from the image (step). 102).

  FIG. 4 is a diagram illustrating a relationship between an image obtained by imaging with the rear camera 10 and a left partial image and a right partial image therein. In FIG. 4, a rectangular area H indicates a frame buffer of the rear camera 10, and a circular area C lacking a part indicates an image of the rear of the host vehicle captured through a fisheye lens. In such an image, a range S1 in which the position of another vehicle traveling on the left adjacent lane is obliquely behind the vehicle position is extracted as the left partial image. Similarly, in such an image, a range S2 in which the position of another vehicle traveling in the right adjacent lane is obliquely behind the vehicle position is extracted as the right partial image.

  Next, the specific pattern detection unit 16 separately searches for the specific pattern image included in the extracted left partial image and the specific pattern image included in the extracted right partial image (step 104). The search for the specific pattern image is performed by searching for a specific pattern shape, specifically, a circular shape image that matches the tire shape from the left partial image and the right partial image extracted by the detection target range image extracting unit 14. Is done. As shown in FIG. 4, the left partial image and the right partial image obtained by imaging with the rear camera 10 through the fisheye lens are circular in shape because other vehicles traveling in the adjacent lane are seen from the front oblique direction. The tire image is distorted into an elliptical shape. In order to remove the distortion of the image, viewpoint conversion is performed on the left partial image and the right partial image shown in FIG. 4 to create a left partial image and a right partial image in which the vehicle traveling in the adjacent lane is viewed from the side. It is desirable to detect a circular specific pattern image from the inside.

  5-7 is a figure which shows the relationship between the other vehicle currently drive | working an adjacent lane, and the right side partial image by which viewpoint conversion was carried out. These drawings show, for example, a case where the host vehicle G overtakes another vehicle traveling on the right adjacent lane.

  In the example shown in FIG. 5, the right-side partial image includes the rear half of the vehicle body immediately after the host vehicle G overtakes another vehicle and includes the rear wheels of the other vehicle. In the example shown in FIG. 6, the vehicle body G and the other vehicle are separated from each other, and the entire vehicle body including the front and rear wheels of the other vehicle is included in the right partial image. In the example shown in FIG. 7, the distance between the host vehicle G and the other vehicle is further increased, and the front half of the vehicle body including the front wheels of the other vehicle is included in the right side partial image. In this embodiment, when other vehicles up to a total length of the vehicle body of about 5 m are included in the right partial image (or left partial image), as shown in FIGS. The range of the right partial image (or the left partial image) is set so that one tire can be detected.

  Next, the specific pattern detection unit 16 determines whether or not a specific pattern image is detected in at least one of the left partial image and the right partial image (step 106). When a specific pattern image is detected, an affirmative determination is made. Next, the traveling vehicle determination unit 20 determines that there is another vehicle on the adjacent lane corresponding to the partial image in which the specific pattern image is detected (step 108).

  In addition, the direction indication determination unit 22 determines whether the direction indicator indicates whether the direction indicator (blinker) is in operation (step 110) or not (in step 110, affirmative determination). It is determined whether or not the vehicle is pointing to the adjacent lane side where it is determined that there is another vehicle (step 112). If so, a positive determination is made.

  Next, the warning output unit 24 starts warning output (step 114), and alerts the user that there is another vehicle on the adjacent lane to be changed. Further, when the direction indicator is not in operation (negative determination in the determination in step 110), or when the direction indicator of the operation is not pointing to the adjacent lane side where it is determined that there is another vehicle ( In the determination of step 112, a negative output is not issued, and a series of processing ends.

  If no specific pattern image is detected for either the left partial image or the right partial image, a negative determination is made in the determination of step 106. Next, the traveling vehicle determination unit 20 determines whether or not it is within a predetermined time after the specific pattern image disappears from the left partial image or the right partial image (step 116). If it is within a predetermined time after disappearance, an affirmative determination is made, and then the white line detection unit 18 detects a white line included in the partial image on the side where the specific pattern image has disappeared (step 118), and determines the presence or absence of the white line. Determination is made (step 120). A negative determination is made when the white line to be detected is not included in the partial image. In this case, the process proceeds to step 108, and the traveling vehicle determination unit 20 determines that there is another vehicle on the adjacent lane corresponding to the partial image for which the disappearance of the specific pattern image is determined.

  8-10 is a figure which shows the relationship between the other vehicle with the long full length which is drive | working an adjacent lane, and the right side partial image by which viewpoint conversion was carried out. These drawings show, for example, a case where the host vehicle G overtakes another vehicle traveling on the right adjacent lane. Further, as other vehicles having a long overall length, for example, when the length of a wheel base such as a large bus or truck exceeds 5 m, even if the other image is included in the partial image, the front wheel and the rear wheel It is assumed that both tires deviate from the partial image.

  In the example illustrated in FIG. 8, the right side partial image includes the rearmost part of the vehicle body including the rear wheels of the other vehicle immediately after the host vehicle G has overtaken another vehicle. In the example shown in FIG. 9, the position of the host vehicle G precedes the other vehicles, and the portion between the front wheels and the rear wheels of the vehicle body of the other vehicles is included in the right partial image. In the example shown in FIG. 9, the specific pattern image (tire image) included in the right-side partial image immediately before (FIG. 8) has disappeared. In the example illustrated in FIG. 10, the distance between the host vehicle G and the other vehicle is further increased, and the front partial vehicle body including the front wheels of the other vehicle is included in the right partial image.

  In the example shown in FIG. 9, the specific pattern image is not included in the right partial image, but within a predetermined time after the specific pattern image disappears, the white line L of the adjacent lane (the side opposite to the host vehicle G) is If it cannot be detected because it is hidden behind the body of another vehicle, the alarm output is maintained assuming that the other vehicle is continuing to travel on the adjacent lane.

  In addition, after a predetermined time has elapsed since the disappearance of the specific pattern image from the partial image (including the case where the specific pattern image has not been detected immediately before the disappearance), after the negative determination is made in the determination of step 116 Alternatively, after the specific pattern image has disappeared from the partial image within a predetermined time but a white line is detected, and after a positive determination is made in the determination of step 120, the traveling vehicle determination unit 20 It is determined that there is no other vehicle behind the position of the vehicle in the adjacent lane (step 122), and a series of processing relating to detection of other vehicles and alarm output ends.

  Thus, in the rear side vehicle detection alarm device 1 of the present embodiment, the shape of the specific pattern included in the image of the detection target range is combined with the shape of the specific pattern included in the detection target range and the presence or absence of the white line. In addition to detecting other vehicles traveling in the adjacent lane, other vehicles can be detected even when the shape of the specific pattern is out of the detection target range. Detection accuracy can be improved. In addition, since the detection of other vehicles combined with the presence or absence of white lines on the road is performed only when the specific pattern image disappears from the image of the detection target range, the misdetection of other vehicles due to not detecting the white line Can be reduced, and a decrease in detection accuracy can be suppressed.

  Specifically, the traveling vehicle determination unit 20 detects when the white line is not detected by the white line detection unit 18 from the time when the specific pattern image detected until then disappears from the detection target range. It is determined that another vehicle is traveling at a position corresponding to the target range. As a result, the specific pattern image has disappeared from the image in the detection target range because the total length of the vehicle is long, but in reality, other vehicles are detected including when the other vehicle continues to travel in the adjacent lane. It becomes possible, and the detection accuracy of other vehicles can be raised.

  Further, when the traveling vehicle determination unit 20 determines that another vehicle is traveling, and when the lane change of the host vehicle is performed toward the adjacent lane in which the other vehicle is traveling. By outputting a predetermined warning to the driver of the own vehicle, the alarm can be issued only when it is necessary to truly call attention to the driver of the own vehicle. The discomfort felt by the driver and other passengers with the alarm output can be reduced.

  Further, both the left detection target range and the right detection target range set in the left and right rear side areas of the host vehicle are simultaneously imaged using one rear camera 10 attached to the rear of the host vehicle. . As a result, the cost can be reduced by reducing the number of parts and the number of mounting steps. In addition, a system that images the rear of the vehicle and displays it on a display device in the vehicle interior is widespread. However, in a vehicle equipped with such a system, a camera provided for imaging the rear of the vehicle is used as the rear camera 10. Since it can be diverted, further cost reduction is possible.

  In addition, since the tire has basically the same shape regardless of the type of vehicle, the use of a specific pattern having a shape corresponding to the tire allows other tires to be highly accurate regardless of the type of vehicle. The vehicle can be detected.

  In addition, this invention is not limited to the said embodiment, A various deformation | transformation implementation is possible within the range of the summary of this invention. For example, in the above-described embodiment, the case where the host vehicle G overtakes another vehicle traveling on the right adjacent lane has been described with reference to FIGS. 5 to 10. However, the host vehicle G travels on the left adjacent lane. The same applies to the case of overtaking other vehicles in the vehicle or the case of other vehicles traveling on the left or right adjacent lane of the host vehicle G, and the present invention can be applied.

  In the above-described embodiment, the tire shape is used as the specific pattern. However, another vehicle (for example, a front grill shape) may be added to the specific pattern to detect other vehicles.

  As described above, according to the present invention, by combining the shape of the specific pattern included in the detection target range and the presence or absence of a white line, the shape of the specific pattern included in the image of the detection target range is detected to travel in the adjacent lane. In addition to detecting other vehicles in the vehicle, it is possible to detect other vehicles even when the shape of the specific pattern is out of the detection target range, so that the detection accuracy of other vehicles can be improved. Can do. In addition, since the detection of other vehicles combined with the presence or absence of white lines on the road is performed only when the specific pattern image disappears from the image of the detection target range, the misdetection of other vehicles due to not detecting the white line Can be reduced, and a decrease in detection accuracy can be suppressed.

DESCRIPTION OF SYMBOLS 1 Rear side vehicle detection alarm apparatus 10 Back camera 12 Image memory 14 Detection object range image extraction part 16 Specific pattern detection part 18 White line detection part 20 Traveling vehicle determination part 22 Direction instruction | indication determination part 24 Alarm output part

Claims (6)

Imaging means for imaging the detection target range of the rear side area of the host vehicle;
Specific pattern detection means for detecting a specific pattern image corresponding to another vehicle from the detection target range based on an image obtained by imaging by the imaging means;
Based on an image obtained by imaging by the imaging unit, a white line detecting unit that detects a white line drawn at the end of the adjacent lane and on the opposite side of the traveling lane of the own vehicle from the detection target range;
When the specific pattern image is detected by the specific pattern detection means, or after the specific pattern image detected until then disappears from the detection target range, the white line detection means does not detect the white line. A traveling vehicle determination means for determining that another vehicle is traveling at a position corresponding to the detection target range;
A rear side vehicle detection alarm device comprising: In claim 1,
The traveling vehicle determination unit is configured to detect the detection target when the white line is not detected by the white line detection unit from the time when the specific pattern image detected until then disappears from the detection target range. A rear side vehicle detection alarm device characterized by determining that another vehicle is traveling in a position corresponding to the range. In claim 1 or 2,
When the travel vehicle determination means determines that the other vehicle is traveling, and the vehicle changes its lane toward the adjacent lane in which the other vehicle is traveling, A rear side vehicle detection alarm device further comprising alarm output means for outputting a predetermined alarm to a vehicle driver. In claim 3,
It further comprises direction indication determination means for determining the operating state of the direction indicator of the host vehicle,
The rear side vehicle detection warning device, wherein the warning output means uses the direction indicated by the direction indicator determined by the direction instruction determination means as the direction of the lane change. In any one of Claims 1-4,
The imaging means is a single camera that is attached to the rear end of the host vehicle and simultaneously images both the left detection target range and the right detection target range set in the left and right rear side regions of the host vehicle. A rear side vehicle detection alarm device characterized by that. In any one of Claims 1-5,
The rear side vehicle detection alarm device, wherein the specific pattern has a shape corresponding to a tire.

Images